Wednesday assorted links

1. “Variation in skill can explain 44 percent of the variation in diagnostic decisions, and policies that improve skill perform better than uniform decision guidelines.”  Not a Covid-19 paper, but relevant of course, link here.

2. Which states are practicing social distancing the most? (NYT)

3. Human challenge studies to accelerate a vaccine.

4. My Bloomberg column on how the macroeconomics of Covid-19 do and do not resemble WWII.  Oops, correct link here.

5. The idea of “group testing” actually came from economist Robert Dorfman of Harvard (who taught me history of economic thought way back when).  And more on pooled tests.  And Nebraska is doing pooling.

6. “Use Surplus Federal Real Property to Expand Medical and Quarantine Capacity for COVID-19.

7. Why scaling up testing is so hard (New Yorker).

8. We still don’t know the CFR for H1N1.

9. “Overlooked is the possibility that beauty can influence college admissions.”  But not for Chinese it seems.

10. Mullainathan and Thaler with some deregulatory suggestions (NYT).

11. “The Food and Drug Administration will allow doctors across the country to begin using plasma donated by coronavirus survivors to treat patients who are critically ill with the virus, under new emergency protocols approved Tuesday.

12. Benjamin Yeoh on early vaccine use.

13. James Stock: “The most important conclusion from this exercise is that policy hinges critically on a key unknown
parameter, the fraction of infected who are asymptomatic. Evidence on this parameter is scanty, however
it could readily be estimated by randomized testing.”

14. Two elite factions in tension with each other (nasty stuff, please do not read).


From the paper TC linked to:

"Evidence on this parameter is scanty, however it could readily be estimated by randomized testing."

And then:

"Based on this limited evidence, I adopt two values of the asymptomatic rate, 0.30 (for example, used here) and 0.86."

That's a wide interval!

People good at advocacy: What's the best way to advocate for national representative (or even statewide representative) testing?

a. Directly email state health director?

b. Directly email firms who engage in this kind of thing - a local lab or something?

I imagine $ would flow rapidly to this, given its importance, RoI.

"That's a wide interval!"
Yes, it's wide. It reflects the uncertainty. Still even a 30% rate would suggest there may be significant natural immunity out there.

Right - yes, great point.

I think there's experiments popping up in diff places - e.g.

a. Telluride -

b. Many other examples in WSJ (including the NBA!) today via Stanford's Eran Bendavid and Jay Bhattacharya -

- and so I wonder how to get faster, more aggressive here. Cordon off 1000 tests (I know there's scarcity - but there has to be a funder or lab who can find 1000). Find a random U.S. sample. Or a random, I don't know, Rhode Island sample. Test; check asymptomatic carrier rate; make some decisions about how to proceed?

CDC, local health guidelines - "only test if you have symptoms." This seems foolish even as more and more tests become available in light of asymptomatic data.

Pandemics are not wars - no factories are being destroyed, no cargo ships are being torpedoed, no cities are being fire bombed. Which has already been written in a comment here, in the same fashion that the link is not to Bloomberg, but to an earlier MR post.

While literally true, doesn't that rather undersell the destruction of human capital and organizational knowledge if and when companies are broken up and sold?

The thing about "go do something else" is that you may not have 20-30 years of experience making you super efficient and productive at "something else."

Labor is not *fully* fungible, nor is industrial capacity.

When stock prices of automakers, steel and aluminum refiners go down, clearly factories are being destroyed!

Factories are capital built by paying workers, financed by selling shares to investors deploying the savings of workers.

When workers are flush, they will save more, generating more investment in playing workers to build more factories.

The only reason the share price goes down is the factories get used up, aka depreciate, or are physically destroyed like with bombs, or innovators uses savings from other workers to pay workers to build improved factories that produce more for lower unit cost (but higher overall cost to the economy), ie, destroyed by obsolescence.

Only if you see workers as a liability and customers getting all the money they spend from government, and you believe not paying workers to work (profits) creates jobs, can you see falling stock prices as "wealth destruction".

Falling stock prices merely indicates loss of mostly government sanctioned monopoly profits and rent seeking, not "destruction of wealth". As Adam Smith put in in a version of his intro, the wealth of a nation is the productivity of it's people, ie, workers working. He was arguing against the mercantilism that today dominates Wall Street today, with one major change.

In Adam Smith's time, capital was mostly stuff,, eg, gold, storehouses of cotton, cloth, grain, things that were readily converted into labor, by way of selling them to workers for money workers had to earn by working.

Today, money is held to be equivalent to capital.

So, you no longer need to own stuff, whether warehouses of cotton goods or factories that turn natural gas from a pipeline into nylon, polypropylene, or rayon, etc fabric and clothing.

Thus, today, a billion dollars in cash means a billion dollars of SARS-Cov2 vaccine, EXCEPT GOVERNMENT BANS VACCINES. It's no longer about paying workers to work. Wealth is destroyed by paying workers to work.

Today evil government, or nature, is requiring paying workers to work for years or decades to stop disease from harming rich people, or making consumers fearful and thus not spending money paying the monopolists and rent seekers.

If we are talking about the U.S., no factories were being bombed during the war either, so what's your point?

13- yes, it’s important to know. Iceland and Decode had the proportion of infected at 0.86% in a ( fairly) large random sample with 1/2 asymptomatic.
I assume they took care of false positives.

Henderson needs to never again use "one reason is because."

#14 is not a very good "bad take" to promote discussion, though
#1-#13 definitely represent his usually superlative taste - which is why I keep coming back to read this blog.

From my years of lurking, Cowen's format is:

1. He lists the most conventionally insightful link first, with periodic demotions to #2 to have a link to attempt to raise the status of artistic triumphs (which I think he believes is important for greater civilization).

2. He lists the most important culturally controversial provocation last (often it is a troll post in the vein of Cunningham's Law, but that's to throw you off when he occasionally subscribes to the controversial belief).

3. If it is *super* culturally controversial (e.g. genetic sex differences) but would otherwise be important he buries it in the middle instead.

4. Everything else in between.

Why link #14 here when there are so many better provocations right now? He even adds the reverse psychology bit not to read it so I thought it must be an especially super great bad take, and now I feel super guilty blathering about feeding the controversy virus about content that doesn't deserve it.

I don't usually take Cunningham's Law bait because I don't agree with the tactic, especially poorly executed, but the stakes of the "raise-lower status" debate is a bit higher in a pandemic and I kind of was planning to do this regardless.

So here is my suggested superior provocation:

Archive for proof of original tweet:

Obligatory bait comment: "Definitely don't read this take, really, and what does this say about elites in our society?"

I otherwise am not grumpy and really generally enjoy the content here, but I really hate how this tactic has spread throughout society and I wish to lower its status if I can LMAO.

6. “Use Surplus Federal Real Property to Expand Medical and Quarantine Capacity for COVID-19.”
They mean isolate the homeless camps in LA. By now this virus is endemic to homeless camps.

Link above didn't work for me:

And then I posted an incorrect link. (Wouldn't an Edit feature be nice.)

7 seems a bit strange, considering this current news from the UK

Thousands of 15-minute home tests for coronavirus will be delivered by Amazon to people self-isolating with symptoms or will go on sale on the high street within days, according to Public Health England (PHE), in a move that could restore many people’s lives to a semblance of pre-lockdown normality.

a move that could restore many people’s lives to a semblance of pre-lockdown normality.

Individuals that test negative for the virus and return to pre-lockdown normality will be exposing themselves to infection.

For some reason, much of the article is apparently blocked. However, maybe this will pass - Thousands of 15-minute home tests for coronavirus will be delivered by Amazon to people self-isolating with symptoms

The test is intended to be used for those with symptoms who are currently at home.

This sounds good, and there is a right way to use this. If you are isolating at home, and get a positive result, use it to get your doctor to take you seriously.

But there's a bad way to use it too, and that's treating it as a get out of jail card for you and your family.

What are manufacturer's recommendations for a symptomatic negative? Retest every day for a week? Remain isolated in that time?

"7. Why scaling up testing is so hard (New Yorker)."

This seems out of touch. Testing is scaling up rapidly. At this point, most states have tested 5x-10x the number of confirmed cases. If every tested case (or a huge percentage of them) test positive, it's a likely sign that you have a vast pool of unconfirmed infected.

If only 1 in 10 of test cases is infected, then that's probably not the case.

This is from the article:

"According to the covid Tracking Project, the number of daily tests has grown from just under a thousand, on March 4th, to more than sixty-five thousand, on Monday. "

There's was huge screw up early by the CDC and FDA, but we're past than and the numbers are ramping up. The article addresses a whole lot of issues, that don't seem to be an issue in other countries. So, they probably aren't as significant as the article paints them.

NJ and LA managed to pull out from the slums. Watch out for MD and AZ. NY is chasing Lombardia.

Bad copying - here is more detail.

“Several million tests have been purchased for use. These are brand new products. We have to be clear they work as they are claimed to do,” she said. “Once they have been tested this week and the bulk of tests arrive, they will be distributed into the community.”

Amazon has agreed to carry out distribution and the tests will also go on sale in chemist shops.

11. “The Food and Drug Administration will allow doctors across the country to begin using plasma donated by coronavirus survivors to treat patients who are critically ill with the virus, under new emergency protocols approved Tuesday.”
We have robust anti-bodies because the virus does not mutate much. This means game over for New York, with 25% carrying the virus then most of New York will be immune within eight weeks. I am suspecting the immunity is long lasting since the virus does not mutate.

"I am suspecting the immunity is long lasting since the virus does not mutate." That will change its incentive to mutate (using the usual handy metaphor for describing such things).

True; but there's no indication Cov19 is particularly good at mutating, and from a low unexposed population of humans and a small Cov19 population, hard to get much velocity going again.

3. Fauci and the other "experts" who didn't realize human challenges would be necessary to speed vaccine development and clung to the "it just can't be done in less than 18 months" mantra need to be held accountable after all this is over.

It was obvious a month ago that we should begin testing every credible vaccine in humans the moment they were developed and that we should expose all study participants to the virus as soon as blood tests indicated immune response to vaccine candidates.

Shouldn't you wait at least until there is an effective vaccine before holding anyone accountable for something that is not yet true?

And the Chinese, who knew considerably more than a month ago that a vaccine was critical, only reached thos point in the last week - "China has commenced a Phase I clinical trial of a vaccine against Covid-19, the infection caused by the novel coronavirus that is so far responsible for 341,700 infections and 14,750 deaths globally.

According to Chinese media, a staff member associated with the government-funded project said that participants in the trial are already being vaccinated.

The trial is designed to enrol 108 volunteers aged 18-60 years who are residents of Wuhan, the city where the virus originated. Participants will be divided into three groups and administered with different dosages.

The experimental candidate is a recombination vaccine developed by biotechnology company CanSino Biologics in alliance with a research team from the PLA Academy of Military Medical Sciences.

Last week, CanSino Biologics received regulatory approval to conduct a Phase I trial of its recombinant novel coronavirus vaccine (adenovirus type 5 vector) candidate.

Global Times reported that participants will not be infected with the novel coronavirus following the vaccination in the Phase I trial.

The aim is to investigate whether the vaccine will trigger antibody generation and make participants immune to the new coronavirus.

Trial investigators will quarantine volunteers for 14 days and follow them for six months to monitor any adverse reactions."

"Shouldn't you wait at least until there is an effective vaccine before holding anyone accountable for something that is not yet true?"

No. You should hold people accountable for testing vaccine candidates as quickly as possible.

Moderna sent the NIH all the vaccine needed for a phase I trial on Feb 25. The first patient was injected on March 15, and everyone was patting each other on the back for acting fast.

That's insane. The first person should have been injected by February 26, and the rest should have been injected the instant it was clear the vaccine candidate wasn't a fatal poison.

We should be a few days away from data on immune response and, if that immune response looks good, a few more days away from exposure to live virus.

You mean the Chinese were dragging their feet? They were injecting people right around March 15 also.

I do not know when Chinese officials first had access to enough doses of experimental vaccine to say whether they were dragging their feet.

I know officials here waited 18 days to begin testing a credible vaccine candidate, which is too slow.

And I know they're almost certainly going to forbid testing vaccine efficacy via direct exposure, at least initially, so we're going to have to wait much longer than necessary to know if they work.

Under normal conditionals, these precautions are perfectly reasonable. Under current circumstances, they're criminal.

you sed " the rest should have been injected the instant it was clear the vaccine candidate wasn't a fatal poison."
fortunately that's not how it is done
mebbe you should read up about Guillain-Barré syndrome

Yes. Clearly this whole Covid thing isn't a big enough deal to subject a couple dozen healthy volunteers to a small-but-significant risk of adverse vaccine response. Better delay for several months.

usually they start with animal testing

Surely there are some Uighurs who the CCP can get to "volunteer" to be exposed to the CCP-Coronavirus.

There are plenty of volunteers here. Trials filled hours after they opened.

4. I think this link is broken as it points to

'uniform decision guidelines': I have the impression that in the US "guidelines" are often interpreted as instructions. Is that wrong?

“Overlooked is the possibility that beauty can influence college admissions.” And so it should. My rule was that if I had two equally good candidates for the last place on a course then if they were male I'd take the younger, if female the prettier, and if one of each I'd toss a coin.

That was before all the other genders we're invented, of course.

...but seriously folks. If I had two equally good candidates and they were male or female I would take the one more likely to produce stimulating conversation in the classroom. I am a normal male and appreciate female beauty as much as anyone, probably more than most, but with 60% of college students being female, outside of some STEM fields there are plenty of pretty girls and it is not my job to increase that number anyway. If they were male you'd take the younger? Why? Or was your whole post satirical?

13- The asymptomatic ( some of which may never develop symptoms) may have to be estimated eventually by a serological test. Because these asymptomatic may only test PCR positive for 3 to 4 days.
After this, the PCR test may find them negative when in fact they have immunity.
This is important to determine the true fatality rate which may possibly be much lower than the figures commonly thrown around

Very disappointed with the lack of international comparisons and contrasts. e.g. James Stock: " Evidence on this parameter is scanty, however it could readily be estimated by randomized testing.” Doesn't South Korea or Singapore etc. have this data.

New Yorker article: no comparison contrast with South Korean, Taiwan, Hong Kong and Singapore.

Where are they getting swabs and reagents from or is that they have flattened the curve so much they don't need very much?

No, they don't. They tested many people, but to get this data, you don't need to test many people, you need to test *random* people.

Only Iceland has some related data apparently. At 0.9%, if we assume the same rate for the US, there are 3M infected people in the US, not 60,000 as reported. I suspect the truth is still higher.

Rio de Janeiro City has beaten the coronavirus and will end its lockdown.

#14) Correct for group 1, but off for Group 2. It doesn't really matter what Group 2 "hopes" to do. Markets ensure that Group 2 can only make $ by providing valuable goods and services. Consumers hold all the economic power.

Regardless, the cultural battle is not between cultural and economic elites. It's between those with an abstain-and-atone mindset and a human-flourishing mindset. The abstain-and-atoners view American consumerist excess as a sin in need of correction. They view extreme social distancing --- no matter how long in duration, even until a vaccine becomes available (>= 1 year), if necessary --- as a necessary corrective to American excess. Whether climate change or coronavirus, they have a natural inclination to believe that the solution to every problem is American sacrifice through abstinence and atonement. The Flourishers view American liberal capitalism as the greatest source of human flourishing in history. They view social distancing as a strictly temporary measure to buy enough time to defeat the virus through American ingenuity and productivity: production of more masks, PPE, tests, ventilators, hospital capacity, etc. and use of same to develop more targeted measures to slow virus growth rather than blunt country-wide, universal economic shutdown. They have a natural inclination to view the solution to every problem as *more* economic productivity and innovation, which they recognize cannot happen during a prolonged shutdown.

#9 Anyone who saw Legally Blonde can attest to it being true.

Here's a fun story re: number 9, but it involves hiring not college application:

I had to hire an engineer to replace my position as I had just become the manager over that department. Several interviews with local candidates went nowhere and we cast a wider net to out-of-state applicants. I had a Skype interview with one candidate, but I couldn't figure out on my phone how to get her picture up on the screen, so I just conducted the whole interview over the phone even though she could see me (I just pretended everything was OK because I didn't want to seem incompetent). Well turns out she was a good enough candidate and I hired her. This was also in the late 2000's when it wasn't de rigeur to scan social media and LinkedIn profiles.

So she shows up for her first day of work and this woman is just drop dead beautiful. Of course I had no idea because I had never seen her but EVERYBODY at work gave me so much crap for it because they thought I had seen her over skype and used that to make my decision


I have a more general question here, but I suppose it's related to #8. Given that Italy, Spain, and France appear to have peaked in deaths as of today, at far below what would put them on track for typical annual flu deaths, what makes Richard Epstein wrong?

More generally, why does he have to shout to be heard? He's no crank.

Isn't this what makes him wrong? "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."

One day with less deaths than the day before is not sufficient to call a peak. All three countries have been in lockdown for some time. So the relevant question is what happens if they lift the lockdown, right?

It's premature to say deaths from COVID-19 will be less than annual flu deaths. Death rates could well remain elevated for quite some time. What are annual flu deaths for those countries anyway?

Is this a rebranding of disinformer Huloo? Has Huloo been banned?

" What are annual flu deaths for those countries anyway?"

In Italy, it's anywhere between 10,000 and 20,000. The error you note he has since corrected, but at the moment no country is even approaching annual flu deaths, not even Italy, to say nothing of exceeding them (or exceeding them by a factor of 10, which is the "conservative" figure that is often thrown around).

Epstein more or less conceded that that number, 500, was a math error or typo. My question is whether the analysis is wrong?

"at far below what would put them on track for typical annual flu deaths" seems then very incorrect.

Inconceivable that the headline number could have been a typo. "Math error" does not inspire much confidence. Essentially, he is backpedaling and admits he will continue to backpedal as new numbers come in ("it is possible both figures will move up in a rough proportion from even that revised estimate").

As far as analysis, I don't see much. All he says is it's not certain that progression will continue to be exponential, which isn't saying much.

If you've read other stuff, he actually says it is certain that growth will NOT be exponential. The main reason? It never has been exponential in the case of any other viral infection known to humans. It begins exponentially, reaches an inflection point, and then declines.

Why everyone seems to have forgotten this is beyond me (be sure to watch the video):

See the fuller discussion by Epstein here:

Infections follow a sigmoid function ((logistics curve). They go exponential until they run into constraints (lack of available hosts) then they start a decaying exponential function till they level off. It's rare that the limit is actually the total population.

That's simplified, but gives you an idea of what to expect. I'm hardly an expert of course, maybe one of the medical experts can chime in with more detailed information.

Well if you compare Epstein's revised estimate of 5K American deaths versus claims of 1 million deaths, sure he makes a lot more sense. At this point, I think the very high claims are just ignorant of the available data. However, I'm not convinced that he's nailed it . The most fundamental flaw is he is going off of Chinese data, which is probably intentionally under counted and in any case, the US will never weld people into their apartment buildings.

"the US will never weld people into their apartment buildings."

Just curious, did this literally happen in China?

I don't know. I just know that I'm not a gun guy, but a heavily armed populace seems like a good idea right now.

"Just curious, did this literally happen in China?"

There are dozens of videos and comments from China claiming this was going on. Personally, I suspect it was happening, but am doubtful they were closing every entrance. The most probable explanation is that they locked or welded up all the secondary entrees and that way they only had to post guards at the main entrance. Then all personal traffic was temperature checked and only allowed with valid reasons.

Here's one such post:

Epstein originally estimated world deaths at 50,000 up 8 fold and in the USA if the death rate increases at the same rate current 67 deaths should reach about 500. the 500 figure is about 8 x 67. If he originally meant to estimate total USA deaths as a proportion of the world population he wouldn't have included the then current USA deaths.

Mr Epstein claims that deaths will level off as people voluntary adapt responses to slow the spread of disease and that the virus will become less virulent so that it is more likely to spread to a new host before the current host dies.

This overlooks the fact for 90% of the people under 40 it seems to suffer nothing more than a cold so there is not much pressure for it to become less virulent.

They have not peaked at all - maybe the number of dead per day will not significantly increase in Italy, but going from a death rate of 700+ a day to a death rate of 500+ a day is not what most people consider peaking. Since three days at the lower rate of 500+ is still 1500 more dead people. Which would be around 3% of his global total in three days, in Italy alone, after the rate of increase reverses.

And the current American death toll is 878, compared to the number of 537 Americans noted in that March 24 article.

He may not be a crank, but he is clearly and obviously wrong, in a way that is a bit more than a 'gaffe' when referring to a pandemic.

"... but going from a death rate of 700+ a day to a death rate of 500+ a day is not what most people consider peaking."

Actually, it's what everyone who understands what the word "peaked" means that it has peaked. It means that deaths per day are growing at a factor of less than 1.0.

What definition are you using?

Total deaths, of course. Just like Epstein, who still clings to the idea that this pandemic will peak at around 50,000 deaths, according to your link.

When you look at daily death numbers in Italy, you can plot them on a graph. At some point the curve reaches a high point (what most people call the "peak") and then declines. In viral epidemics, the back end of the curve tends to look like the front end.

"Given that Italy, Spain, and France appear to have peaked in deaths as of today, "

I presume by peaked you mean their reported deaths are less than they were yesterday. Assuming a trend off of one data point seems precarious.

As to Epstein, he's clearly made a wrong early prediction. The US is going to have far more than 500 deaths. We are already over 700 and there were 225 deaths yesterday alone. He's vastly underestimated the impact in the US. Being clearly wrong (on a prediction from a week ago) should result in a devaluation of his opinions.

To be clear: "reported additional daily deaths."

In Italy we have four days of data that show a peak. We've been making assumptions about growth on almost as little data. Why suddenly be skeptical when the data shows that growth has slowed or is reversing?

And during the four days since that highest single day death toll, between March 21 and March 21, almost 2000 Italians have died, or 2% of Epstein's predicted total of 50,000. 3 days, one country, 2%. The global death toll is not really peaking, even if the increase in one county is declining.

Ah, March 24 of course.

Ooops again - not 2%, that is 4% of Epstein's ludicrous number.

3 days, March 21-24, one country, 4%

Slowing or reversing growth rates are exactly what we expect when lockdowns are implemented. We certainly have not been making assumptions based on a single data point, though. Nor should we.

The virus was widespread in Italy before the lockdown. And the lockdown is much less respected in Italy than in, say China. How much of the decline is to be attributed to the lockdown, since not all of it can be?

I would say all of it or almost. Why not?

I actually gave two reasons for why it may not be to your none for why all or most of the reversal was a result of the lockdown.

Well sorry, I don't understand why either of your "reasons" actually supports your claim at all. My "reason" is that this is the purpose of lockdowns and there is nothing else that I am aware of that could have resulted in the slowdown. Certainly there is no good reason to believe that Italy has achieved herd immunity.

Nah I think when it comes to death toll predictions the only way that makes sense is let’s make a deal rules. Whoever gets closest without going over. Far too many boomers spouting out absurdly high death tools to private health benefits and socialize societal misery.

14. What makes people in universities, media, or Hollywood “cultural elites”? They don’t have any power. Plenty of people do just fine with no contact whatsoever with universities, media, or Hollywood. Cultural “power” is like Tinkerbell, and exists only in the minds of people who believe in it.

The only real elites are people in government because they are the only ones who have power over other people. The law and its coercive power exists whether you believe in it or not. Economic elites can also exist because the resources needed for survival also exist whether you believe in them or not, and because there is a revolving door among the highest economic echelons and government.

On the coronavirus point, the tradeoff between containment versus economy is not what the herd immunity herd wants it to be. If there is a dangerous virus circulating around, people are not going to go out and spend money even if they were free to do so. In fact, most private companies in my area told their employees to work from home and most people who could started social distancing at least two weeks before the state-mandated order came down. The state shutdown order really had little effect as most people were already voluntarily shut down before that, and repealing the order isn’t going to get people to come out of voluntary shutdown. There will be no economic recovery until the virus can be reduced to manageable levels, which unfortunately requires a big-bang approach because we were late in enacting containment measures. This doesn’t need to and shouldn’t last for many months—I’d say tell everyone we’re going to have a lockdown, and it will end at the end of April regardless of results. We know this can work because it has worked in China, which is now restarting its economy after less than two months of extreme containment (except in Wuhan City). If we can get numbers down to manageable numbers, we can use a South Korea strategy to keep them low without shutting the economy.

They have the power to control public opinion and therefore politicians

No this is silly. Indeed, the fact you are wrong about the immense power wielded by Hollywood underscores what is so wrong about his analysis.

Hollywood is just so much more powerful than the media and academic elites that it doesn’t make sense to group them.

Hollywood shares similar politics to media and academic elites but not because of any kind of propinquity- leftist views are high status and because Hollywood is a new money high status anxiety kind of place left wings are widespread. Compare Hollywood in 1957 to the New York Times and for CCNY faculty.

Or look at Bernie Sanders. Hollywood is pretty enthusiastic about Bernie- the media ramped into overdrive to defeat him after his early victories. It wasnt Hollywood that told them to do that- it was the economic elites many of whom are married to media elites. Look at the spouses of media elites far more economic elites than Hollywood elites.

Wrong. The contemporary term is "soft power". It's religion, Plato's cave, and Keynes statement that "practical men, who believe themselves to be quite exempt from any intellectual influences, are usually slaves of some defunct economist." Just because it's soft power does not make it any less influential.

GOP Senators Demand Fix For "Massive Error" In Coronavirus Bill Which Would Create "Perverse Incentive"


The entire bill, start to finish, is a pack of disincentives. Republicans led the charge on piling on.

#5: Pooled testing is a nice idea, I wonder if its time has already passed given the rapid rollout of tests now. (OTOH there seem to still be questions about the accuracy of the new tests?)

It sounds like Nebraska is taking individuals' samples and combining five of them into an aggregate sample to be tested. There might be two disadvantages to that process: are the patients told that their test is not really "their" test but part of a group test?

And if the group result comes back positive, the group needs to be re-tested which means that time has been lost. If the virus were slow to spread, or if the patients were already quarantined, this loss of time would not be a big deal. But if these are free-ranging people, the epidemiological models have shown us that even one day of delay has a large effect on the eventual result.

On yet another hand, a good proportion of the country is indeed doing something close to self-quarantine now.
Just 0.2% of Chinese patients in their twenties died of COVID-19, compared to 8% of those in their seventies and almost 15% of those in their eighties.
Bring out your dead.
Of those young people are many who acquired the virus and went into remission, never going to hospital. We need wider testing before jumping to conclusions about death rates.

These rates need to be presented in terms of excess death over "normal" rates. In the US, 25 year old males have about 0.16% annual death rate without the Wu Flu, 75 year olds about 3.5%, and 85 year olds about 10%.

What fraction of the 85 year old population shows up in the denominator of 15% computation?

The way Michael Levitt seems to have described it is that it doubles your risk of death in the next two months. Therefore I guess if I am following the reasoning that it would be that if 85 year olds go Cov19, in an average year, where their risk of death would be 10%, it would be 12% (assuming risk of death through the year is cumulative?). Seems very low though.

My term life insurance company is trying to tell me (through my insurance rats) that, at age almost 66, my likelihood of dying in the next year is about 1.1%. (This includes a profit for them, so it's less.) That means that doubling my risk of dying in the next two months increases the total risk of dying in the next two months from 0.2% to 0.4%. While that's not insignificant, especially if I die, it doesn't seem worth sending the world into a second Great Depression for.

Case in point - -

(Professor Neil Ferguson) told the committee that the latest research suggested as many as half to two-thirds of deaths from coronavirus might have happened this year anyway, because most fatalities were among people at the end of their lives or with other health conditions.

Not saying it's wrong (I'm not sure of my feeling on it), but say overall mortality is lower than the 1% and perhaps in the 0.2% (mainly through more a greater ratio of asymptomatic and mild cases than we know, relative to unknown deaths) and of that mortality we're buying most of the dead 6-9 months. Hell of a sacrifice and maybe *we* can afford it as the right thing to do ("we'll bounce back to trend!"), but for a younger country closer to the margin of subsistence and with greater wealth impacts on health (India, Brazil, even China)...

I am an influenza survivor

I'm a cold survivor, thirty times over.

#14 This factor is the real reason for US underperforming during the COVID19 crisis. In other countries, especially China and East Asia, these two factions are actually under the same tent, as opposed to the US where they are allowed to rabidly attack each other out in the open. I'm sure in the long run it makes us more dynamic and creative, but it's not helping in this crisis.

The only creativity happening is in the savageness of their open conflict.
In many countries, coups and civil wars have happened due to inter-elite competition.

"nasty stuff, please do not read"
Straussians in free-speech countries are usually just trolls with sophistry to self-justify, as in this case Tyler is.

#13: One obvious implication of high fraction asymptomatic individuals, and should this be confirmed by finding low estimated CFR, is that vaccine development should be a least a bit easier than we would think.

If Cov19 is already kind of a wimp and you're starting at 0.4% CFR (or 0.2% estimate per Nuffield -, then a vaccine should be less hard than if we're starting at 4% CFR.

Though what's the acceptable level of fatalities for a vaccine? 0.001% 1/100,000)? Universal vaccination at that level would hit 3,000 US deaths. Smallpox vaccine is about 0.0001% (1 per million; or 300 deaths in a universal vaccination situation in the US). What will we accept in this situation?

Related but off-topic question for any of those that know, does the power of CRISPR make it easier to produce attenuated vaccines, where you can work out the nasty bits of a virus and then systematically cut them out? Or is this, for practical purposes, just not a very relevant tool?

8. From the CDC:

Updated Estimates from April 2009 – March 13, 2010
Using the same methodology CDC has again updated the estimates to include the time period from April 2009 through March 13, 2010 on April 19, 2010.

CDC estimates that between 43 million and 88 million cases of 2009 H1N1 occurred between April 2009 and March 13, 2010. The mid-level in this range is about 60 million people infected with 2009 H1N1.
CDC estimates that between about 192,000 and 398,000 H1N1-related hospitalizations occurred between April 2009 and March 13, 2010. The mid-level in this range is about 270,000 2009 H1N1-related hospitalizations.
CDC estimates that between about 8,720 and 18,050 2009 H1N1-related deaths occurred between April 2009 and March 13, 2010. The mid-level in this range is about 12,270 2009 H1N1-related deaths.

We may not know precisely, but we’re pretty sure it was big (relative to COVID19.

"We may not know precisely, but we’re pretty sure it was big (relative to COVID19."

Does that follow from those numbers?

9. Me in real life: I was not accepted to the Top Ten-Ivy League college to which I'd applied. So I presented myself at the admissions office in person to argue my case. I went day after day. I also showed up at the departmental office of my hoped for major. Made friends with the secretaries. After a week or so, gears began to turn. I was grudgingly accepted. Graduated with prizes and honors. Looking at photos of myself from back then, I can see: I was a pretty good looking young man.

I work with a well-regarded doctor. Had a similar story about being accepted to the top hospital training program in his specialty. Finished near the bottom of his class in medical school. Had is application rejected with scorn. Showed up at the specialty head's office. Made friends with the secretary. Met the boss. Got accepted and had a great career. He was, and is a very good looking fellow.

I can not imagine leaving in a country where education spots are assigned to people just because they are good-looking or smooth-talking. In Brazil, testing is king.

That's because almost everybody in Brazil is good looking. Pretty smooth talking, too, from my experience.

All 94 residents of NJ nursing home presumed positive...

Nursing homes, emergency rooms, homeless camps, stadiums, mass transit, classrooms are no go for quite a while.

If mass transit is, or is believed to be, toxic, how much impact does that have on NYC and other cities with high subway dependence?

2. Which states are practicing social distancing the most? (NYT)
Montana and Wyoming. I think they are the most sparse. The largest towns are 60k.

4. My Bloomberg column on how the macroeconomics of Covid-19 do and do not resemble WWII. Oops, correct link here.

I think an interesting resemblance might be the post WWII recession starting in Feb 45. It only lasted 8 months and was caused by the rapid drop in military spending coupled with GI's being discharged from the army and returning home.

It was a dramatic drop and a lot of capital lost nearly all value (you could buy cheap 'army surplus' goods in bulk, in the Pacific they offered jeeps for free to GIs if they just paid the shipping home but many didn't want them so they threw them into the ocean). What made for a "V" shaped recovery was a lot of pent up demand. People made money during the war but couldn't spend it. Rationing restricted purchases. No new cars were made as all the auto factories retooled to war production. People were shamed into buying war bonds rather than spending.

Now a lot of people may emerge from weeks of lockdown with money to burn but on what I'm not sure. You can go out to eat but you're not going out to eat 30 times in a day to make up for the month you missed. I suspect there will be a lot of home office purchases and people will emerge taking up some hobbies. Cruise ships may never recover but I think travel will. Where we go from here is hard to say.

#5 > Pooled tests for Covid-19 (random pooling or stratified pooling based on social networks) can improve testing efficiency when prevalence rates are lower.

Nobody had read the NY Health Commission COVID.19 test directive?? It instructed not to test up to moderately ill patients. Now which countries or areas have low prevalence rates?? If so is cost or test throughput the dominant factors?? Putting lipstick on a pig.


At my undergraduate alma mater, each individual college makes its own admission decisions.

While there, I noticed the women at the School of Hotel Administration were the most attractive, and the women at the School of Industrial and Labor Relations were a close second.

Later, I learned that the Hotel School requires every candidate to actually interview -- not meet with an alum, but interview with an admissions official who rates each candidate.

And Industrial and Labor Relations requires an interview with each transfer candidate.

Yes, hotel managers are pretty consistently good looking, and for reasonable commercial reasons. But industrial and labor relations? Huh? It sounds like a glorified human resources department.

#5: Horse racing has been doing group testing for years...
I just assumed it was de rigeur...

#14. A clear case of the need for economists to negotiate a truce. Which is not helped by the gratuitous attribution of bad motives to both sides.

Comments for this post are closed