Before I get to that, I must report that my claim that the worst is behind us is not looking as good today. For the U.S. as a whole, as of the afternoon update the 3DDRR was 1.52, essentially unchanged from yesterday. Excluding NY, it was 1.50, slightly higher than yesterday’s 1.47

A commenter points to a piece by Zvi Mowshowitz.

My prior at this point is that the difference between a low and high initial viral load of Covid-19 is large. The theory makes too much sense. But with high uncertainty.

…We should be studying with experiments how Covid-19 spreads, and how likely each method is to work, using controlled experiments. Yes, this involves infecting individuals. Considering how many lives are at stake and the ability to test using young healthy volunteers who are then isolated, I fail to see how anyone who objects on the basis of ‘ethics’ knows what that word means, or why we should listen to them.

. . .it does not seem so hard for most of us to avoid direct out-of-household interaction with highly symptomatic people.

That last point raises many questions.

1. The people most likely to give you a high-load infection are the people you live with. Almost by definition, you cannot avoid them. Are you ready to re-open schools and let your kids come home with infections?

2. If the trick to this whole thing is to avoid people who might give you a high-load infection, then can you do that on a normal rush hour on the NY subway?

3. What the heck do we do about nursing homes, where a lot of people with low defense factors live in the same place?

1. Roman Frydman and Edmund Phelps write,

The government’s approach is ill-suited to the crisis. The stimulus isn’t merely the wrong dose, but the wrong medicine altogether.

Instead of boosting public employment or seeking to stimulate demand, lawmakers should focus on mobilizing the private sector to combat the public-health crisis. We don’t dispute the need to improve America’s highways, airports and other infrastructure. But that would do nothing to address the specific causes of today’s crisis: self-imposed lockdowns to stop the spread of disease.

The CARES Act will live in infamy as one of the worst pieces of legislation every enacted. The portions of it that were supposed to address short-term liquidity needs have been nullified by complexity and bureaucratic snarls. But you can bet that the portions that reflect rent-seeking and irrelevant agendas will be implemented much more effectively.

And the politicians in Washington think that this is only the appetizer. They are preparing the next course.

2. Gary Cohn and Glen Hutchins write,

The Fed will be operating at an unprecedented scale, reportedly lending as much as $5 trillion, which is more than its entire balance sheet before the crisis. It will also be engaging in a practice in which it has little experience: targeting capital to individual companies in commercial industries. This is an important and complex task that requires great care and speed.

The Fed’s ability to identify and implement appropriate tools will be critical to its success in this new role. It is vital that the central bank succeeds in mitigating the pandemic’s damage to the economy. It is also important the Fed avoids the stigma that followed the 2008 “bailouts.” This will require a thoughtful approach to staffing, process and disclosure that ideally would be implemented at the outset.

They are cheerleading for the Fed, which I am sure is what you have to do if you want to stay in mainstream journalism or economics. We have become more Chinese than China, in that the centralization of capital allocation is more rigorous here and the Distributed Information Suppression Complex is more effective than China’s much cruder control over news.

3. A reader forwards a pointer to paper by Ke, et al.

Here, we argue that because death and the cause of death are usually recorded reliably and are less affected by surveillance intensity changes or delay in confirmation than case counts, the time series of death counts reflects the growth of an epidemic reliably, with a delay in onset determined by the time between infection to death. Based on this idea, we designed a simple 88 methodology to disentangle the epidemic growth from confounding factors, such as underreporting, delays in case confirmation and changes in surveillance intensity. We fit models to both case incidence data and death count data collected from eight European countries and the US in March 2020. We show that in most countries, the detection rate of infected individuals is in general low, and COVID-19 spreads very fast in these countries

In other words, they agree with me that the death rate gives you a way of looking at the spread rate through a rear-view mirror. I don’t know why they even bother to look at confirmed-case data at all.

4. Another reader sent me a link to an article behind an FT paywall (https://www.ft.com/content/9ee6f251-f3ee-4d42-8cac-e372f8564088) that says that Iran has adopted a credit-line approach similar to what I have suggested. Who says I have no influence?

Note to commenters: feel free to correct me on any of this, preferably with links to definitive sources. I will post corrections here.

There are two stages to the virus disease. In the first stage, one gets flu-like symptoms. In the second stage, if it occurs, an immune response coats the lungs, reducing oxygen flow do the blood. I believe this is called ARDS, although sometimes I see it referred to as pneumonia. The second stage is by far the most deadly.

UPDATE: This article makes me think perhaps in three stages: flu-like; pneumonia; ARDS. But it doesn’t change the main point.

On the defensive side, we know that risk of death goes up with both age and obesity. As I understand it, obesity is a big risk for the second stage. I do not know whether being young is protective for the first stage, the second stage, or both. Perhaps it does not matter which.

On the offensive side, it seems intuitively reasonable that a more aggressive virus attack, meaning you inhale more of the virus, is more likely to cause worse symptoms at the first stage. It seems intuitively reasonable that having worse symptoms at the first stage is more likely to trigger the second stage.

This reminds me of an Avalon Hill war game, in which you move counters, and when a battle occurs, the winner is determined by the attacker’s attack factor, the defender’s defense factor, and a roll of the dice.

I view Robin Hanson’ variolation idea through the lens of that metaphor. Our current approach is either to have a mass retreat or a mass advance, not bothering to look at attack factors or defense factors. It seems to make more sense to choose our battles where we can be confident of low attack factors and high defense factors.

[UPDATE, April 16. This post was a bad call. Sorry.]

I am posting this at 11 AM eastern time, and I am about to go on a bike ride. By the time I come back, I may regret saying this. But I think that the worst is over on the virus crisis. The trend of the 3DDRR is going to be down, perhaps even faster than my optimistic scenario.

If I am correct, then many questions remain. The big one is whether things would turn bad again if we opened up more sectors of the economy. I wish we had a better way to test that other than experimentally lifting lockdowns.

UPDATE, 2:20 PM. a commenter suggests that New York might be distorting the figures. I took totals for the U.S. and subtracted the numbers from New York. Then I calculated the 3DDRRxNY, that is, the rate for the U.S. excluding New York. It is lower than NY. And also declining. I’ll keep my eye on 3DDRRxNY, but so far I stick with my call.

One of the unknowns in the virus crisis is what explains differences in severity. Of the people who have been infected, it seems that more than 95 percent experience low severity. Also, we see wide differences in severity across countries. Is Taiwan doing better than Spain because fewer people have been infected in Taiwan, or the infections are less severe in Taiwan, or both?

It seems to me that the possible explanations for variations in severity include:

1. How you are attacked–how much of the virus you get and how far it goes initially into your respiratory system.
2. How well your individual body defends.
3. How you are treated by the health care system.

The conventional wisdom, as I understand it, is that (2) matters, and I believe this conventional wisdom. That is, we think that young people without underlying conditions defend better once infected than do old people or people with underlying conditions. Of course, it would be better to have knowledge of which underlying conditions affect the ability to defend.

The conventional wisdom, as I understand it, is that (3) matters, but I am skeptical about it. The conventional wisdom is that we need to keep the number of hospital beds and ventilators ahead of the spread of the virus, or otherwise people will die unnecessarily. The conventional wisdom seems consistent with the high death rates in Northern Italy, Spain, and New York City. But there could be other explanations. Perhaps the rate of infection was higher in those areas. Perhaps how you are attacked matters, and people in these areas were more likely to be attacked more severely.

Suppose that more ventilators and hospital beds had been available in these dire regions. Would that have produced more cures, or merely kept some people alive a few more weeks? I am getting the impression that a shortage of ventilators means that victims who are beyond hope might have to be denied a ventilator, but it is less clear that people who could survive if given a ventilator must be denied one. I am by no means committed to this point of view. It is just a guess. Any evidence to the contrary would be sufficient to get me to change my mind.

The conventional wisdom is relatively silent about (1). But I wish we knew more. For example, suppose that strong attacks only come from symptomatic spreaders, while getting the virus from an asymptomatic spreader means that you face a weak attack. That would imply that fears of asymptomatic spreaders are exaggerated, which would have some significant policy implications. It would imply that a focus on identifying and isolating the symptomatic individuals is the key to preventing deaths. It might mean that universal masks and scarves, while not preventing all infections, might do well at preventing severe infections, particularly if symptomatic individuals are identified and isolated.

1. A grim article on what ventilators can or cannot accomplish. (NYT)

2. Today’s figure puts the 3DDRR at 1.64, a further drop from yesterday. The doubling time has no extended from three days to five days. If this keeps up, in another week there will be a ferocious argument over whether it is time to lift the lockdowns.

[corrected at 12:40 PM–bad arithmetic error before. Thanks to David Henderson for spotting it]
Because we have not done random-sample testing, we have no idea of the true number of cases in the United States. Reported numbers are worthless. Below, when I refer to cases I mean the (unknown) true total number of people who have ever been infected, not the reported case numbers.

Suppose that we try to work backward. Suppose that we assume a lag of n days from the time of infection to the time of death. Then the number of cases as of n days ago is equal to the number of deaths as of today, divided by the true (unknown) case fatality rate.

The more cases (meaning actual cases, not reported cases) that it took to generate all of the deaths as of today, the happier we should be. A higher number means that there are more people who have already been infected, so we are closer to the peak of the curve.

For example, as of yesterday, there were 8314 deaths. If we assume a true case fatality rate per 1000 of 20, that means that as of n days ago there were 50 times 8314, or 41,570 415,700 cases. If both cases and deaths has doubled every three days (3DDRR = 2.0), then the number of cases today is 2^(n/3) times 415,700. If n is 9, then that means we had about 3.3 million cases as of yesterday. Is that a lower bound? Or is the cfr higher than 20 per 1000 (higher than 2 percent)?

If we raise n to 15 and keep the cfr at 20 in 1000, then the true number of cases as of yesterday was about 13 million. If we raise n to 15 but assume a cfr of only 2 in 1000, then the true number of cases as of yesterday was about 130 million. Is that the upper bound? If we are anywhere near that, we are close to a peak.

How will we know when and how to exit from the lockdown strategy? I think we need to conduct experiments.

I know this is a pipe dream. The only experiment we will ever see will be “Lift the lockdown and see what happens.”

But I would suggest that we recruit young, healthy people who are uninfected and have never had the virus as experimental subjects for the following experiments. Note that I would be willing to see the one child of mine who qualifies as healthy to participate as an experimental subject, because I think that these experiments would be safe. Note that for these experiments the “infected person” is either asymptomatic or has only very mild symptoms.

1. The doorknob effect. Have the infected person open a door, and have the experimental subjects follow at one-minute intervals and open that same door. Quarantine the experimental subjects for two weeks and meanwhile test them for the virus.

2. The classroom with masks and scarves. Have the infected person sit in a classroom with one hundred experimental subjects. Everyone, including the infected person (whose identity is known only to the investigator, not to the experimental subjects), wears a face covering of some sort. Try this experiment with the infected person having different types of face coverings. Quarantine and test.

3. The virus-in-the-air experiment. Have a person known to have the disease in mild form walk through a hallway. Have experimental subjects follow at one-minute intervals. Repeat this experiment with a different combinations of face covering. One version would have nobody wear face covering. Another version would have both the infected person and the experimental subjects wear face covering. etc. Quarantine and test.

These sorts of experiments would provide a better scientific basis for making decisions regarding modifying or lifting the lockdowns.