The Speed Premium in an exponentially growing pandemic world

I’ve blogged a few times in the past about the importance of speed, and speed as an input into productivity and innovation. What many people do not realize is that “the speed premium” is vastly higher when a deadly virus is doubling in reach every five to seven days.

An economic or epidemiological plan from a week ago might be worthless or even misleading today.  For instance, some scientists have told me that at some point, if the virus is widespread enough, there is no choice but to let it burn its way through the population (not saying we are there yet, probably not according to the consensus of experts I am seeing).

Have you perused recent newspapers and mentally noted how many of the articles — such as reviews of art exhibitions — obviously were written and planned in The Time Before (can I call it that?).  Those articles are now largely worthless, though a few of them may have nostalgia value.

If you are writing commentary, the value is being there is the morning, not the evening.  The “commentary cycle” used to stretch at least a day or two, occasionally a full week.

The corporate value of being prepared early with a good telework plan has been especially high.

Ben Thompson writes: “…on January 23, the day that China locked down Wuhan, Taiwan had the capability of producing 2.44 million masks a day; this week Taiwan is expected to exceed 13 million masks a day, a sufficient number for not only medical workers but also the general public.”

If you are seeking to start a business, to deal with the third party vendors that Amazon is (temporarily) abandoning, you cannot just wait a month or two.  You have to start now.

The Chinese system has its flaws from an anti-pandemic point of view, most of all low transparency.  But their typical rapid speed of response has been astonishing — setting up that hospital in six days — and it is a big reason why they are on a (partial) rebound.

If you are giving philanthropic grants, you have to be ready to give them now.  If you give them three months from now, you may well miss the boat in terms of expected impact.

Are you ready for a world where the speed premium is so insanely high?

I wish to thank Daniel Gross for a conversation related to this blog post.  We spoke at 3.5x.


This pandemic has shown the relative strengths and weaknesses of each country's model.

China's authoritarian system is sensitive to any sign of dissent and thus suppressed early reporting (BAD). Once it identified a threat, especially one it regarded as existential, it marshalled all resources and imposed significant limitations on people to quell the outbreak (GOOD, in this instance). They could suspend habeas corpus, ramp up manufacturing, build hospitals, and move in health care workers from other provinces without having to go through checks and balances. Just like the Soviet system, the state directing output can achieve impressive results in a singular direction.

The US normally would allow more decentralized ground up reporting (GOOD), but there was a self inflicted injury by not being attuned to the problem and having test kits ready to go, which was probably our biggest failing. Whose fault that is depends on your political biases, so I won't speculate on that point.

However, the American worship of individualism makes it hard to perform mass quarantine in the way that China did. Also the political system leans towards paralysis and obstructionism than quick action (our weaknesses).

Our businesses however have performed well, as expected. We use price signals, not government directives, to direct resource allocation and production. Prices have risen providing a good market incentive. Using that 3M and others are ramping up production of masks. There is a lag in this though, as it takes time to observe price signals and then shift production.

Innovation in lab testing has also been quick and impressive. Our lab has FDA approved PCR testing for COVID in house ready to go live. Cepheid's system will allow point of care testing e.g. at airports. America relies on innovation to solve problems, and we're doing great on this front.

As Tyler wrote, America is usually slow to start but impressive once it gets going. You're going to see the tide of disease turning in the next few weeks.

We're likely seeing the stock market bottom here. I've started to buy into broad stock market index funds already.

I do not doubt that impressive efforts to tackle the disease are getting going in America. However, this may be a case in which "slow to start" imposed a handicap that cannot be overcome by later excellent efforts.

The early failures with testing, as will as the continued slowness to implement quarantine in much of the country, may facilitate a spread of the disease that overwhelms later good activity.

Hopefully this is not the case, but other places that imposed restrictions much earlier than the U.S. have had great trouble controlling the outbreak. Italy imposed a national lockdown when they had only 6,400 cases and less than 400 deaths. 16 days later, they have nearly 70,000 cases and over 6800 deaths, numbers that are still rising rapidly. What will the U.S. numbers look like, even if complete lockdown is imposed now?

The numbers are going to be very, very bad. That would have been true even if the U.S. had reacted better than France, a country where it is also estimated the number of unknown cases is much higher than captured in statistics, as the French problems are much closer to Spain's than they are to Germany's.

Define very, very bad in terms of US deaths. 10,000? 50,000? 100,000?

He can't, no one can.

Nobody wants to say for fear of being labeled insensitive. A normal flu season in the US is ~ 30 k deaths , a bad flu season ( 2017-18) is ~ 60 k deaths.
So ~ 30 k deaths this year should be acceptable. It would make this a bad flu season.
However, I am afraid that at this point, we’re so invested in fighting this that any number will seem unacceptable

OK, but if, in the end, we have 15,000 deaths, I would conclude "Holy crap, we're lucky we didn't dawdle longer."

The big unknown at this point is the second wave, and the race to build capacity and find treatments and maybe a vaccine.

We could have fewer deaths for another reason. Namely that it’s not that deadly and the number of infected was way underestimated

The testing from the last week doesn't support that. We've run over 400K tests and have a case count of 60K. Ergo, only 1 out of 6 people who are suspected of having it really have it. Surely that group has a much higher incidence than the general population?

I suppose it's still possible that there are 20x infected (1 million) hiding in the population, but it seems unlikely at this point.

However, we are currently running about 60K tests per day and that number is climbing. So, we'll probably have a good idea of the total numbers by next week.

Let me correct my assertion. It would surprise me for the CFR to drop from the current 1.4% but I'm doubtful it will drop below 0.5%.

Sure, if only you have 16% in the pool of individuals suspected enough to test, then surely it's higher than in the general population, but by what factor? Seems like if it's even 5% in the general population, then the death rate would be either *very* low if you trust the official tested deaths, or difficult to estimate if you assume that the vast majority deaths are missed, but likely not even 0.5% without very large increases in overall death rate?

How much below 0.5% do we have to be before "Moderate social distancing, reduced opening hours for bars, reduced staffing density for workplaces, shelter the elderly, massive investment to up respirator and ICU capacity" becomes much strongly preferred over "Lockdown until we get a vaccine"? Which seems to be the only other strategy being proffered.

E.g. we're back to "mitigation" again, when the hammer of containment becomes synonymous with King Canute.

" then surely it's higher than in the general population, but by what factor?"

Yes, that's certainly a trillion dollar question. We probably won't know for at least another week of high volume testing.

Although it is possible that the general population rate could be higher for those who have *had* the infection.

As Catinthehat raises previously, only seralogical (antibody) tests will tell you if someone has had it.

Antigen tests tell you if someone does have it right now. And sure, those who have are selected for being suspected of having it will probably hit antigen tests at a higher % than the gpop, but the antibody test is less certain.

That would be great news, no second wave. Every day of new deaths clarifies the picture. Let's give it a few more days at least before we declare that we have overreacted.

We are on track for 30K deaths by mid-April. That assumes we consider the current lock down. (We'll pass 1K tomorrow.) Death isn't the only endpoint though. We'll be seeing a rise in lung capacity based disability among survivors over the next few years.

The fact that essentially no one (other than island or near-island nations) has been able to slow the spread with testing and quarantine tends to suggest that the speed of the US response as compared to other countries has been irrelevant. It's just a question now, here and everywhere, of flattening the curve.

What about South Korea? No country has been able to stop this but some have done better at mitigating it than others.

When you consider that they have one land border, and it is strictly controlled, I think South Korea would qualify as a 'near island' nation.

Why does land border matter?

If you achieve Wuhan-style lockdown then I see how porous border prevents you from getting to 0, but I see it as orthogonal to the challenge of slowing new cases from the 3-5 day doubling rate that most countries follow pretty closely to something slower.

Incidentally, I think Iran has had better success than the West & achieving slowdown in cases and deaths, and I've seen a decent amount of speculation that temperature and humidity impact R0, and so Europeans and New York have so far had a big disadvantage.

New York City also has a lot of extremely old people. The usual pattern for New Yorkers is to retire to Florida or some other sunny place, but when their health starts to fail and they need an urban environment - no driving - and better social and medical services, they move back to the city. City life lets one live quite nicely after one's golden years have tarnished, but something like COVID19 is extremely dangerous.

Norway looks like they are doing fine. Also Germany looks like the only large European country with its shit together, as usual.

"Italy imposed a national lockdown when they had only 6,400 cases and less than 400 deaths. 16 days later, they have nearly 70,000 cases and over 6800 deaths, numbers that are still rising rapidly."

The good news is that the acceleration of coronavirus cases each day in Italy has fallen to an average of 9% over the past three day average and the increase in deaths has declined to 12% over the past three day. Italy has turned the corner.

Nate Silver threw cold water on that, said it's down in Lombardy while up in other regions.

We all know Q1 2020 is going to be bad for several business. Sell-side analysts and hobbyists are already working on estimates of this impact. But ,we still don't know: (i) the whole and impact on Q1 2020 revenues, (ii) which big business may be insolvent in the followings weeks, and (iii) the reaction of the crowd to low revenue and insolvencies.

Be patient, if I remember well the bottom came after 6 months on 2009.After reading the newsletter of Matt Levine yesterday, I'll keep the cash several weeks more.

@Rick Zhang - excellent "center mass" analysis, +1.

@dan1111 - excellent analysis, +1, the USA = Italy, "it's all over but the shouting"

@Very bad - don't be so pessimistic. According to the "Super Forecasters" site that TC linked to the other day (keep in mind a prediction market where real money is at stake would be more accurate than proven super-forecasters, but for stupid legal reasons in the USA this is not done) the USA might "only", say the consensus, have a 70% chance of less than 23M Americans infected by Covid-19 and a 80% chance of less than 350k deaths in the USA. See:

My contribution to this thread? Emergency situations tend to give a 'crude solution' to a problem, hence the 'lack of fluff' that TC cites (no museum reviews), but in fact fluff is what is the real meaning of life. As any doctor will tell you, an emergency room physician's standard of care is much lower than an ordinary physician. The job now is to 'patch up' America as best one can, and deal with the fallout and finger pointing later. What finger pointing? That Trump is indirectly responsible for 100s of 1000s of American deaths. That --and this is pretty much a foregone conclusion--the Covid-19 virus was designed as a chimeric virus in 2015 in the U of NC (Chapel Hill) and later perfected in a Wuhan bioweapons lab. On this last point, not even a conspiracy theory if you look at the evidence, there's even an NIH paper about this virus, called SARS-CoV, a first generation version of SARS-CoV-2, this years more slick, more 'natural' looking chimeric virus, listing one of the very same two Chinese nationals on the 2015 paper who this year helped sequence this year's virus, See: (and CNTRL-F "chimeric' if pressed for time, note the at least one of the two Chinese researchers listed in this paper are prominent researchers today in Wuhan)

As for stocks, I'm also buying *soon* but not yet. A good source of when to buy is Ben Carlson's posts at Forbes, "Wealth of Common Sense". Ben and I occasionally also trade emails, as is befitting for somebody as distinguished as I.

You realize that my very, very bad numbers involve around 100,000 confirmed infection cases in the NYC area, with just 10,000 requiring breathing assistance for days or weeks. Numbers spread over a month, not even a week or two, unlike in Madrid. Meaning that with even less preparation, NYC will have done better than Madrid as of today, where bodies are being put into an Olympic sized ice rink.

Well, yeah, China's top specialists in SARS research will be all over this latest outbreak.

A quick way to debunk/confirm this is to compare samples from SARS-CoV-2 with SHC014. With sequencing technology it's hard to hide bioweapons without blowing your cover.

For the sake of conspiracy theory, you think this is a lab leak?

@ long nym - I've done a couple of biotech inventions but it's not my field; maybe it's yours. Here is my sir rebuttal to your rebuttal, based on a rebuttal of the alleged conspiracy theory (but not on point with my point about the 2015 chimeric virus, but close enough): (incompetent rebuttal, essentially says "SARS-CoV-2 is too sophisticated to have been done by off-the-shelf rDNA (recombinant DNA) so it must be natural evolution."

@long nym - you may be right. But, don't underestimate the Wuhan bioweapons lab's ability to cover their tracks? They would be a pretty poor bioweapons lab if they were using off-the-shelf ligase / lyase and standard entry points on a DNA sequence when manufacturing chimeric organisms (I speculate)? A topic for a Reddit thread but I've been banned there And the above post got pulled from a popular finance site. Essentially, if the above is right, both the Chinese and the Americans are correct in saying the Wuhan virus was made in America / made in Wuhan.

I can't see why they bothered engineering this coronavirus when the whole point of their research was to demonstrate that it was highly likely to evolve naturally. SARS and MERS revealed a family of viruses capable, with suitable mutations, of attacking the human respiratory system. The paper discusses how a deadly epidemic virus evolves from the wild pool. Then they built a chimera to show that their tinkertoy analysis could work in the wild. They probably used this paper warning of the risk of such danger to get funded in China. China couldn't have better people working on the problem.

Sure, someone could have engineered COVID-19, but COVID-19 was an epidemic waiting to happen.

P.S. I have a farmer friend who raises meat animals and sometimes bristles at some of the weird ass regulations. For example, she couldn't get a USDA stamp for some buffalo since it was slaughtered over dirt and grass, not over a concrete platform. In light of the problems China has been having with cross over viruses, I'll agree that some of these regulations are weird ass, but there is a good chance they are there to prevent certain types of contamination.

While the US response has markedly improved in recent days, I disagree that it was transparent and allowed ground up reporting at the beginning. Although there were cases in Washington by January, regulators did not allow testing for community spread until Helen Chu went against the regulators and found community spread at the end of February. Worse, Senator Burr was receiving classified intelligence briefings that a pandemic was likely in early February and decided to sell stocks while telling the public that everything was fine. Much of the US government’s knowledge about the coronavirus at the beginning was considered classified and any leaker would have been sent to jail just like a Chinese whistleblower. The sudden crash of the stock market in late February shows that people were caught unaware of the magnitude of this problem because the government was telling the public that everything was fine while high-level government officials were offloading their own stocks. I do not think what little ground-up reporting existed (the only people who were right were fringe prepper types who always predict the worst) helped America prepare in the early stages any better than China. In fact, I would argue America was worse prepared at the beginning than China—people in China had informal knowledge and didn’t trust the government so many did in fact leave Wuhan before the government acknowledged how bad it was, but Americans trust the government so there was no mass movement of ordinary people to sell stocks or leave Seattle/New York until now when problems have gotten worse than they were in China when China’s lockdown began.

Once it became clear that this was the Virus of Kings, closing all the airports at once would have been an appropriately speedy response.

By the time this became clear (which going by the stock market was the end of February at the earliest), closing the airports would have accomplished nothing.

So the paper of record assured us - travel bans were "unjust and wouldn't work anyway" - on February 5th.

Yesterday I read an amusing Reader's Digest-ish quote by a Walter Kirn, on twitter: "I used to just read a major mainstream newspaper and watch a cable channel or two and get my information that way ... Now I check a thousand lunatic sources a day and average the results."

I am unwilling to entirely surrender whatever small information I can glean - admittedly off-book, and perhaps as often leading me into error, as you would undoubtedly say - from observing what is "on the table" and what must never be "on the table" in not only this but other matters of national interest.

Oh, credit where it is due, since the media keeps telling us to listen only to this or that expert: it was Rosie Spinks, a "global tourism reporter", to whom the NYT entrusted that determination.

No. Every little bit helps. Every little bit hurts.

"closing the airports would have accomplished nothing."

At a minimum they could have required masks on everyone at the airport and checked temperatures on arriving and departing passengers. It would have significantly lowered the number of spreaders in airports and probably caught some percentage of infected.

"closing the airports would have accomplished nothing."

No, closing travel did slow things down for us. I don't think anyone still believes it didn't help. Everyone wearing masks would be better (in conjunction), but that wasn't an option.

Nice to see Taiwan’s increased mask production.

See below about some bad news, as global medical glove production has been cut significantly by Malaysia, not increased.

Another demonstration of how the U.S. is truly a paradise for capitalism, against a background of a frantic rush to buy medical supplies. "That includes a more coordinated process for buying and distributing supplies so that states and the federal government can stop trying to outbid one another, he said." The Taiwanese won't even have to price gouge, they will just let states and the federal government set the price.

And always reassuring to see how time tested vendor methods continue to function in a pandemic, at least in America. "In conversations with ventilator makers, one company “told me I was competing with FEMA to get ventilators,” Pritzker said. “I called another manufacturer of ventilators, and he pointed out to me that I would be competing with countries other than the United States. … I better put in as big an order as possible in order to put myself higher on the list of priority.” Anyone here cynical enough to think that even with a bigger order, they will only receive a fraction of that expanded order in the next couple of weeks, though the company will hold them to that bigger order regardless?

The Federal government through the FDA and CDC were slow to the problem but the state governments of WA, CA, and NY were able to respond much faster and find resources. Washington state probably Federal law but in service of the public good they'll get away with it, maybe get a medal. Sometimes the answer to a dysfunctional big government is a functional big government. Hooray for Federalism.

Kentucky mayor calling people dipsh*ts for not staying at home. My kind of mayor.

At this point, the only effective method of handling the first wave is a still stand, though speed in preparations beforehand is critical. Pandemics are not wars.

The Chinese have demonstrated that a still stand works to get past the first wave. The Spanish, French, and Italians continue to tighten their still stand in the hope that the first wave won't crest much higher, with the Germans starting their still stand from a somewhat better position. The South Korean model is clearly not available to use by the U.S. or in Europe, but that has been clear since the first couple of days of the SK response (SK is probably one of the last countries with active Cold War ABC planning, training, and supplies).

An epidemiological plan from a week ago is going to look pretty much identical to one from a month ago. Covid-19 has not changed in such a short time frame.

Telework does not replace the people actually driving the trucks that come from the factories where people are working to manufacture medical supplies, who are eating food that is handled throughout the food industry chain. More plainly, it does not replace a single person able to insert a breathing tube.

And don't sell the Spanish or Germans or other democratic societies short. Spain is likely already using at least part of this facility, which began being set up on March 20 - 'Some 5,500 hospital beds, including intensive care units, will be set up inside the 240,000 square metre IFEMA conference centre on the capital’s outskirts to cope with surging demand expected in the coming days, the Madrid region said in a tweet.' The Germans have likely completed a 1000 bed facility in Berlin, which they started working on around March 18. It is quite possible to match the Chinese, particularly since the West has had months of lead time to plan and prepare.

The time for speed is during preparation, both to handle the sick and to prepare for a still stand, in the weeks and months available before the pandemic first starts spreading. Sadly, it just may be that the U.S. will demonstrate another model, to contrast with the totalitarian lock down, the democratic shut down, and whatever the U.S. ends up doing. The Spanish, however, are clearly thinking about the problem they are facing, meaning the U.S. still has a bit of time to plan along practical similar lines as this example of 'zoning' shows - "Madrid converts Olympic-sized ice rink to a morgue, near a conference centre kitted out with hospital beds" What is still happening in Madrid is an absolute best case scenario for NYC, one fears.

Each country, apart from China, that has faced this rising wave has tried its best to warn everyone else what will happen, with an urgency that has nothing to do with speed.

A still stand?

What in the name of autism....

Sorry prior but this was funny

He's a human who aspires to fail a Turing test.

Speed to public health? Go to veterans Mitch McConnell and Rand Paul perhaps. Kentucky life expectancy already runs less than Moscow, China and Cuba.

Here's where speed is important, because you're behind the times. It was a few weeks ago that pundits a few weeks ago were gleefully predicting that Trump supporters would be hardest hit. Now that the disease is bursting forth in NYC, Seattle, and California cities, in the wake of bar crowds in Brooklyn and beachgoers in CA defying social distancing, we're supposed to forget that those predictions were ever made.

I don't think we're supposed to forget they were made. It's always been clear that the virus would hit cities harder than rural areas.

I think that we're still in the very early days of this crisis and shouldn't yet judge end results, as proportionally, states/areas that are not taking this as seriously as they need to WILL be the most impacted.

This is true, but of course those people " gleefully predicting that Trump supporters would be hardest hit" were judging the end results far too early. So there's nothing wrong with calling them out on it.

It's always been clear that the virus would hit cities harder than rural areas.

Typical mental image of the US. No large metro areas in the sticks, except for, let's see Dallas-Ft. Worth, #7; Houston, #9; Atlanta, #11; Minneapolis-St. Paul, #14; Cleveland, #15; Denver, #16; etc. definition a large metro area isn't the sticks...

Tom, about those predictions: yesterday, detected cases increased by 31% in Trump states as compared to 21% in Clinton states. …

Republicans speedily agreed to another taxpayer funded bailout. At $2T Trump's stimulus is more than double Obama's. $500 billion for corporations. $367 billion for small business loans. $150 billion for state and local governments. $100 billion to hospitals. $1200 to every American with less than $75,000 in income. Airlines get their bailout. What a wild ride these last few weeks. Only thing rarer than a libertarian is a fiscal conservative.

$500 billion for corporations beats 2019's $480 billion stock buyback total...sweeet!

Why do small businesses get loans while big businesses get direct handouts? This might be a good topic for a future Bloomberg thinkpiece.

Particularly as defending those hand outs will require a lot of hard work on the part of very well paid people.

To add to that thinkpiece, small business is forced to pay wages during shutdown while big business is exempted. Why are we penalizing small business?

An even better question:

Why are those of us who have - since the 2008 debacle - busted our ass and done everything possible to get out of debt and now work paycheck-to-paycheck to remain debt free (and build a very small nest egg) having to fund the "bailouts" for those who instead chose to remain in debt and/or continue pushing and leveraging up to their eyeballs and beyond?

You guys are too optimistic. All your questions will be ignored because this think tank is funded by big business and pushes that angle hard. I mean they even published a 'love letter' for big business. Who does that? It would be an Onion parody if it weren't real life. So they don't care about small business and even less for line workers that bust their ass. If you're lucky, they maybe they call you a socialist.

I'm happy to receive my share of the bailouts of $1200. I feel like a winner like Goldman Sachs, General Motors, General Electric, AIG, and the airline industry. No wait, they are all losers. I have the wrong feelings.

Published 1990

You're late.

As does having at least some ability to not rely on global supply chains that would predictably seize up in event of a major war or pandemic. This is actually news from March 18, not today.

"Malaysian factories supply as many as three of every four medical gloves on the global market. Day in and day out health care workers around the world rely on millions of these gloves, likely without knowing the source, as demand soars and supplies dwindle.

But Malaysia’s glove factories are operating at half capacity as a result of recent government restrictions on businesses aimed at stemming the coronavirus pandemic in the country, the Associated Press reported.

Malaysia, Southeast Asia’s hardest-hit country, has banned foreigners and deployed the military to enforce a two-week lockdown. As part of the latest restrictions, factories had to cease production as of March 18.

Manufacturers considered essential have been allowed to apply for an exemption to remain open. Still, they can only operate with half their staff to reduce the chance of spreading the virus in a crowded workplace. Factories must also meet domestic demand before being allowed to export their products, according to the AP."

This not relying on others is clearly a perspective supported by President Trump when he said just yesterday “We should never be reliant on a foreign country for the means of our own survival. America will never be a supplicant nation.”

Prior_approval’s weird fetish for autarky continues.

It is the freaking pandemic we've been waiting for, for a hundred years, and you think the most important thing to talk about is who is which anonymous commentator at Marginal Revolution.

Dude, you have problems.

With heavy guilt and shame, I fap to prior's comments.

I have read that this former GMU employee was fired 30 years ago for either sexual harassment or sexual assault.

In the spirit of the metoo movement, I think that should be noted every single time you identify this person. Personal courage is important - people who have personally experienced sexual harassment or sexual assault, possibly including yourself in such a traumatic situation, need to be supported unconditionally.

Factories in the US are shutting down too though. Domestic supply chains right now are less secure than buying products from countries where factories are open like Taiwan and even China.

Note that the amount produced in Malaysia, assuming the reporting is accurate, is gigantic. This may be less about supply chains per se, but concentration.

Apparently forty years of U.S. tax cut "investment" was not enough to automate U.S. glove manufacturing. Who knew glove automation could be so complicated.

From WSJ. Suggests the fatality rate of COVID-19 much, much lower than estimated.

The risk is not to the general population, but to hospitals being overwhelmed, primarily by elderly with co-morbidities.

Throw money at hospitals, build tent hospitals, requisition space nearby, cancel all elective surgery, put everyone on outpatient if possible. Spend hundreds of billions call out the Army Corps of Engineers, all hands on deck.

But save a $4 trillion, $5 trillion or who knows how much in lost GDP....$10 trillion over three years?

The morgues are overwhelmed by the elderly with co-morbidities. The hospitals are overwhelmed with non-elderly people unable to breathe on their own.

I am not happy about it. I am pointing out the least bad option.

The current strategy, fantastically expensive, is not about saving lives so much as reducing peak loads on hospitals.

The current strategy is about reducing peak loads on hospitals to save as many lives as possible. A certain group of people is going to have a high mortality rate, that being the elderly with co-morbidities. Another much larger group than the elderly with co-morbidities will require assistance breathing for days or weeks before recovering (leaving aside how many in that group will never breathe on their own again, after the infection is over). Though they are not quite making it obvious, in Italy or Spain, the elderly with co-morbidities (and reading between the lines, being elderly is considered a co-mobid factor by itself) are not really being provided any breathing assistance at all, as all available capacity has been consumed by the non-elderly.

Please update your view of the current strategy because most of what you suggest is already or in the middle of being implemented.

Another oddity: Veterans Administration hospitals not reporting being overwhelmed. They keep putting out press releases they are ready. Google your heart out, that's what is happening. VA website.

If they are not overwhelmed, can they help?

We can't let socialized medicine leave the quarantine.

Very strange. I'm guessing there is a lag for some reason.

Ben, Not every part of the country has even had their public system overwhelmed yet. You know that. You also know they are reserve capacity for when the public system gets overwhelmed. If anything, the most efficient thing to do in the future is to expand the VA system as the place to have surge capacity and let the private system be the place that runs tighter, because they will anyway, as there is no money to be made in having excess beds.

Even NYC is not at the point Spain was a week ago. It is odd to see how after being able to watch the Italian or Spanish catastrophe continue to unfold, people are unable to make any meaningful comparisons in terms of what is coming, within certain broad parameters. Parameters that exclude the disease going away like a miracle, or in April.

in a viral pandemic
the media wants to stop broadcasting
the president's media conferences to the country?

They should have stopped that a couple of weeks ago, though Fox never will stop broadcasting Trump saying whatever comes off the top of his head. “Easter’s a very special day for me.”

“Wouldn’t it be great to have all the churches full?” Trump asked. “You’ll have packed churches all over our country … I think it’ll be a beautiful time.”

Give him points for consistency, as he did say that Covid19 will go away like a miracle, or be over by April. The president is a problem in terms of actual public health guidance, certainly from the perspective of the Catholic Church.

"The Vatican department for liturgy published guidelines on Friday for bishops and priests on the celebration of Holy Week, the Triduum, and Easter liturgies during the coronavirus pandemic.

The document recommends that bishops postpone those liturgies which may be postponed. It also indicates how priests and bishops can offer those celebrations which cannot be moved, such as Easter, in places where public liturgies are suspended. ... The decree orders that, in the places where there are restrictions from civil and Church authorities, the bishop, in agreement with the bishops’ conference, may offer the liturgies of the Easter Triduum in the cathedral, and priests of the diocese may offer the liturgies in their parishes, without the physical presence of the faithful.

“The faithful should be informed of the times of the celebration so that they can prayerfully unite themselves in their homes,” it indicates, adding that live television or internet broadcasts are helpful in this situation."

Obviously, this will not be the first time the president and the pope have had a fundamental disagreement involving matters of Christian behavior.

There are probably a number of neo-Savonarolas pointing out that Covid-19 is God's retribution for ignoring Him and warping His instructions. It's easy to see a huge religious resurgence in the days ahead.

CNN: Mr President, what do you want to say to comfort the millions of people that are scared right now?

POTUS: Well I want to see this beat and America return to normal before Easter.


The head of the Netherlands’ public health institute has told the Dutch parliament that measures to control the spread of coronavirus appear to be working, Dutch News reports.

Jaap van Dissel said the country was seeing a “positive trend”, 10 days after authorities introduced a ban on mass gatherings and closed restaurants, bars, schools and cannabis-selling coffee shops.

Van Dissel said initial estimates showed the infection rate per carrier was on or below one, rather than the expected two or three, meaning that a graph of the number of infections should rise in a straight line, rather than a curve.

The number of new confirmed cases in the Netherlands grew by 17% on Tuesday from a day earlier to 5,560, according to official data.

Those figures likely reflect infections from early March, before the government banned public gatherings and closed schools, according to a report by the Reuters news agency.

The president and his advisors are effectively giving out a lotta good information resulting in the "media"
silly attempt to suppress his speech

The big unambiguous speed premium here would be the speed on testing, modelling and trace and contact. Getting the data, analysing the data well. Those countries that do will make good decisions, those that do not will make poor decisions, unless they're lucky.

We may find that the speed premium on "Action!" - building stuff and closing stuff down (maybe even air travel, though that least so, if so) - has been less positive. We'll see.

At least that seems to be the case in the two non-totalitarian countries that have done the best in terms of testing and tracing to date, Germany and South Korea.

For instance, some scientists have told me that at some point, if the virus is widespread enough, there is no choice but to let it burn its way through the population (not saying we are there yet, probably not according to the consensus of experts I am seeing).

And yet, your link to Tom Inglesby who represents the "consensus of experts" doesn't as much as consider the assertion that there exists a point beyond which we will need to have the virus burn through, let alone make a case that we aren't at that point yet.

Speed is only needed because of the lack of preparation. That’s like the guy starting to do his homework, an hour before it’s due. He is in crisis mode.
With preparation and training, you’re naturally fast anyway.
A firefighting squad, comes to a fire and they don’t become a debate team. They already know what to do.

And the amount of time squandered in the U.S. is likely to be a lesson taught to several generations of public health professionals, as compared to Italy, and to a lesser extent Spain, there is absolutely no way to plead ignorance about this pandemic and its effects.

Europe has weeks to prepare and did nothing.

Prior_approval, your autism is showing.

I have read that this former GMU employee was fired 30 years ago for either sexual harassment or sexual assault.

In the spirit of the metoo movement, I think that should be noted every single time you identify this person. Personal courage is important - people who have personally experienced sexual harassment or sexual assault, possibly including yourself in such a traumatic situation, need to be supported unconditionally.

-What happens when you are on a wet road, driving around a corner, and you speed up? [You skid off the road.]
-What happens when you are in orbit, trying to do a rendezvous, and you speed up? [Your orbit gets bigger, and you end up farther away.]
-Did AlphaStar play faster than its human opponent? (No.)
-If you fly faster in a dogfight, your turn circle gets bigger, and you lose.

Why do we think *speed* is the answer? Lots of people have been talking about faster OODA loops for a long time, so @tylercowen
isn't the first one to suggest this. I just think we should consider a different mental model.

Thanks, I was going to mention decision cycles aka OODA loops (observe - orient - decide - act). The notion of a cycle of observation and action and feedback is commonplace now, but it was put into a very good framework by an Air Force colonel, John Boyd. The applications for business and pretty much all parts of life are obvious. Faster decision cycles mean that you can outmaneuver your opponent, even if you're physically not that fast (e.g. German blitzkrieg operations during WW II, even though much of their army was still horse-drawn).

Your points about over-obsessing about speed are well taken. Through WW II, faster planes usually did have the advantage, but for decades speed has been secondary and as you mention maneuverability has been more important -- and in recent years stealth has been even more important.

All of this still fits within the OODA framework. But you are correct to point out that it's not just how quickly you can move through the decision cycle that matters -- it's also how WELL you can do each of those four steps.

E.g. if you have great stealth, then your opponent's ability to perform the very first step -- observe -- is heavily compromised. Fast decision cycles don't do him any good if his decisions are wrong or are based on faulty observations.

So quality matters as well as speed.

Tyler's call for speed still has some validity. But it has to be "quality speed".

Kevin Drum wrote yestwrday about another aspect of Taiwan's massive mask output- why hasn't the Trump admin made a deal to get some for the US??? Taiwan pulled off a PR coup by opening negotions for 100k of their current 88.2 Million weekly mask output (12.6 Million × 7) or 4.6 billion per year. It is literally 1/10 of 1 % of their output. And we don't even have them yet - negotiations are in progress while New York is going to turn into a morgue.
The speed premium means we need more than 100k, and now. China makes 116 Million/day so Taiwan's per capita output is 6X that - number 1 on earth.

This should be a huge international issue: The highest per capita mask maker on earth - is holding out on the earth - even their greatest ally.

(Drum's Taiwan prod. Figures are a few days out of date- they rise fast!)

I think everything Tyler says it's true, and my take away (whether or not it was the Straussian message) is that we are screwed.

Someone of my age group has to just stay well, because someone in my age group would not be given a ventilator, as we go into the Italian scenario.

So isolate as much as you can, wear masks and gloves if you got them, and stay well.

If you are old and in a position where you still have to work, with people, I am really sorry that our society puts you there.

Actually that's a pretty good argument for a 30-year-old to stay isolated as well. At 30 you aren't bulletproof, and those ventilators will be taken.

In terms of "speed" and changes in priority, two weeks ago we started isolating for others, altruistically. We are entering the phase where we isolate for ourselves, selfishly.

I agree that speed is important. In this circumstance, that speed requires excess infrastructure and knowledge that is a liability in our economy. If a private equity firm invested in a company that had 3 factories producing masks at 33% capacity and double the number of engineering/QA staff as needed, they would close facilities and fire that staff to boost financial results.

If the US wants to be a country with the ability to respond quickly to crises, we need to find a way to maintain that human and physical capital or develop more robust supply chains that can withstand expected shocks.

Identifying cases is less important in the short run than isolating.

Luxembourg, a nation of 620,000, has 1,099 cases, 0.18% of the population, highest measured rate in the world. They have suffered just 8 deaths through yesterday.

This tells us that actual case rates are much higher than reported in the rest of the world. This has been obvious for a long time. So, increases in "cases" or "death/case" numbers tell you more about the aggressiveness of testing than anything else. See Japan.

Anyway, New York was late to react, is cosmopolitan, and has the densest population in the USA. A few days delayed response to an exponential process is deadly.

I still don't see how the US experience ends up being close to Italy. Or Spain. People who understand how exponential processes work have been watching Spain (first death March 3rd) destroy Italy's pace of deaths in the first three weeks.

To calculate your "Italian odds," you don't look at case percentage. You don't even look at hospitalized percentage. You look at percent utilization of ICU and especially ventilators.

A rural county maybe fortunate in a less dense population and a naturally flatter curve, but they also don't have terribly many ICU beds or ventilators.

Why do you say the latter? If they're geared up to help car crash victims on rural roads and a more locally elderly population, they may have more.

(Is this "Cities have money and 'advanced healthcare'" presumption or based on something more real?)

There is a little state-level graphic of ventilators per 10,000 people here.

New York, which now says they need like four times more than they have, was better than most.

I think population density is a huge factor in the spread of the disease. We'll see. New York may need a lot more ventilators than most.

Does anyone have doubling times by US state?

My sense, comparing Los Angeles and Orange counties, is that their rates are about the same. Los Angeles is as they say "a week ahead." LA is pretty steady at 4-5x the OC cases. LA is more dense, and that might have something to do with making an early leader. That and rate of international travel.

I'd like to be wrong, but the national maps show cases pretty much everywhere, and the exponential is in effect.

It’s just the testing ramping up. The real number of infected is not known.

Obviously it cannot be "just" that or there would not be any emergency room impacts.

‘It looks like a war zone’: 2 docs say hospitals are under siege, supplies running low

Yes, the testing results are a lower bound , if everyone was tested the results would be the true number or use a large enough random sample. Perhaps 2% of New York is already infected 400 k people. How do you know ?

"It’s just the testing ramping up. The real number of infected is not known."

And early indications are that we don't have 20 times the number of cases that are reported. In every state that's tested more than 1000 people, the majority of people tested are coming up negative. (That assumes there's no significant problem with the tests.)

Even in New York state, only about 30% of tested people have been confirmed positive. My state has tested 11,000 with under 700 confirmed. So, 6% of suspected cases have been confirmed as real cases.

6% , 30 % are huge numbers, if the rest of the population is infected at that rate , then the death rate is tiny and the true number of cases much larger
Iceland found 0.86% infected in a random sample of 1870 , 5 days ago.
If 1 % of the US is infected already that 3.3 M people.

if the rest of the population is infected at that rate , "

That's almost certainly not true. The person testing positive is clearly not a random sample from the population. It's a highly biased group.

I'll grant you, if coronavirus is far less deadly than we thought and it's very mild among the younger than 40 cohort, we might have very large numbers of unconfirmed cases.

We should test close over 500K in the next 7-8 days. By that point, we should have far more confidence. Also, it would probably be well worth the effort if the US could perform 10K random tests over the next week or two.

yes, of course, they're only testing people with symptoms. Still from maybe 5 different sources ( one of them the 0.86% from Iceland), there's some evidence pointing to a large number of asymptomatic people. See the WSJ for a summary.
It's not proof but it's worth investigating. The Harvard paper cited by Tyler ( #13 in the Wednesday assorted links) makes the same point. We're making decisions without knowing that number.

+1, excellent link.

I really do hope that this is true. I'm curious if the Spanish, Italian and French new daily numbers (which declined over the last day) continue the trend downward. If that's the case, then they are past the inflection point of the sigmoid function. That would be very good news.

Death data are more reliable. The USA first death was February 29. Through yesterday (24 days later), the total US death toll was 778.

Italy's first death was February 21st. 24 days later, total Italian deaths were 2,158, almost three times the US number.

US deaths have been increasing 30% per day, which means a doubling in under three days. At the same juncture, Italy was seeing daily increases above 20%, but from a much higher base. A week later, Italian deaths are increasing a bit more than 10% per day (doubling weekly).

Iran is a couple days ahead of Italy, but if their data can be trusted, experience has been better than Italy, but not as good as China.

Spain didn't see its first death until March 3rd, but 21 days later, 2,991 Spaniards are dead. This breaks all records. Spain could surpass Italy in total deaths before the end of the month.

It's doubtful if those first benchmarks are accurate. Plenty of people were dying of the flu in February. Enough to mask a significant number of deaths without extensive testing, which wasn't happening.

You'd need a county level one to back up the statement.

The entire article backs my statement.

I have no idea why you want to be stubborn about this but whatever.

I'm not going to read an entire article to discuss one point (everyone who isn't prior approval would know this, I imagine), but there's nothing in there with a quick CTRL+F that says rural counties systematically have fewer respirators than urban ones.

If there's a specific point that supports "A rural county maybe fortunate in a less dense population and a naturally flatter curve, but they also don't have terribly many ICU beds or ventilators" please cite rather than point at a state level map.

I guess you'll just have to live without knowing ..

"While the nation’s medical system may have about 160,000 ventilators overall, many of them may already be in use. In Illinois, for example, there are 1,467 ventilators in the state, health officials told WGN-TV in Chicago, but 1,093 of them are already being used. That leaves only 374 available."

Data are suggesting that "dosing" might be important to the severity of disease. If you live with an infected person for a week, you're gonna get a big dose of virus. If you pick it up on the street, maybe not so bad.

This seems to have important implications for health care professionals, who are likely exposed to higher doses. Unlike masks or ventilators, health care professional production can't be ramped up quickly. We should be testing these people closely, they will likely be the limiting resource if we have multiple waves of virus, and an immune army of functional health care professionals would be a huge asset.

Viral load may determine how bad the infection gets. If you get a really large amounts within a few hours, your immune system has little time to react. Get a tiny amount via public exposure and a healthy immune system may eliminated with minor effects.

in a viral pandemic
"just in time" business production theory is overrated

+5, this may be one of the biggest take aways from the event.

Is it cost effective to maintain larger stockpiles to defend against black swan events? This may have significantly effected the math in favor of larger stockpiles.

They may not want to give up all the efficiency of it, but can also diversify their resources to add to resiliency.

While I agree with the thesis of the speed premium, I think this pandemic is exposing the fact that we systematically underinvest in slow, decade long projects like research and risk management.

"the consensus of experts ": it's a novel virus so there are no experts.

Of course there are legions of people who know far more about other viruses than I do. But that's not the same thing at all.

In areas where the virus has not achieved exponential growth, testing is very important now and may keep a pocket of persons safe as well as letting those folks have more distanced but nevertheless relatively safe interaction.

Two week or so closing, stop. Sick persons identified. Presumably others free of it. Test and test and test and trace. etc.

But, it is an empirical question, and needs to be demonstrated by testing...continuous testing

In areas where it has gone to widespread exponential growth, different measures have to be applied.

Maybe in areas where there isn't widespread covid you can frame 2 week shutdown as giving you the reward of more interaction later, along with more testing.

South Korea

As I'm sure you know, testing is not sufficient. You need three things:

1) testing

2) contact tracing of all positive results

3) quarantine of all positive results and all contacts

What we're seeing about the American system is that we have no staff for #2, we do not have the national character for #3.

Let's all go to the beach or the megachurch, am I right?

I’m wondering how effective the tracing will work. If I tested positive it would be easy to narrow down who I could have potentially infected as I’m pretty much a hermit. One of my closest friends is never in the same location for longer than he has to be and wouldn’t be able to remember where he was this morning much less who he was in contact with 72 hours ago. I have other friends and family who would consider social distancing to be not sharing the lap of someone. And if a stranger wanted to document where they were over the last 5 days I’m not sure they would tell the truth even if they could remember. It doesn’t mean we shouldn’t try it but I’m not sure how effective tracing will be.


I guess you didn't read carefully because I said: "Test and test and test and trace. etc."

As I'm sure you know, trace etc. doesn't just appear out of thin air.

Who is hiring 100,000 tracers right now?

"The Chinese system has its flaws from an anti-pandemic point of view, most of all low transparency. But their typical rapid speed of response has been astonishing — setting up that hospital in six days — and it is a big reason why they are on a (partial) rebound."

The building a hospital thing is ridiculous, they only have to build one because they did not have enough. They also had to do all those draconian measures exactly because they were the worst prepared country. They have no real health care system for their big peasant population. Most of their health care is some herbs and tiger penis. It's laughable.

But speed increases the risk of error, of making things worse. Why not use a tried and tested "takings" model instead? See:

I'm amazed by the the number of Western scientists who won't believe the numbers from Wuhan until we replicate the experiment in NYC.

I believe that the key is that do not trust what the Chinese government says. The WHO seemed to give misleading information. Since Taiwan is not a member of WHO, we had to fight the virus on our own and prepare in advance.

Comments for this post are closed