Trading off consumption vs. covid-19 deaths

Somehow I missed this April 6 paper by Hall, Jones, and Klenow:

This short note develops a framework for thinking about the following question: What is the maximum amount of consumption that a utilitarian welfare function would be willing to trade off to avoid the deaths associated with COVID-19? Our baseline answer is 26%, or around 1/4 of one year’s consumption.

So what does that imply for optimal policy?  Will we manage to lose both?  Via Ivan Werning.


Ryan Avent just now:

"I do worry that some people on the right are considering the state capacity it would take to reopen safely and the government aid needed to maintain social distancing and concluding the only realistic way to get through this is for everyone to just...get it"

That might be what you mean by "lose both."

I am indeed worried that given our state incapacity we are essentially throwing a hail Mary pass and hoping Remdesivir passes phase three trials with flying colors in the next 6-12 months.

Losing both: we wait under partial lockdowns for enough people get the virus to push R-naught < 1 (12-24 months?) and we collapse the economy and financial system while doing so

I have yet to see an even remotely coherent response to this concern. What are we actually hoping for? Vaccine in mid 2021?

It's clear at this point that the "Experts" are flying blind and the politicos are loving their new found power.

Their hope? That they can scam even more money and consolidate even more power.

But it is distressing seeing brilliant people who know how to reason playing the "useful idiot" roll by refusing to even attempt to articulate a strategy. They don't even acknowledge the question.

The dog breeders?

That only applies in 5% of cases.

Right. A labradoodle breeder would be perfect.

In Australia our politicians set the country on the path to become another Italy or United States, but the "experts" -- medical professionals -- forced the government to change track by pointing out they were going to get a lot of people killed. Because of this public health measure were put into effect that are over 100 years old. The "experts" or indeed anyone who paid attention in school, knew what was required to control a respiratory infection with droplet transmission.

Thank you for illustrating Andrew's phrase about people who "don't even acknowledge the question." So let me ask it again, more bluntly. Why should we want to "control the infection"?

While not everyone who contracts the virus has symptoms, COVID-19 can hurt. It can hurt a lot. It can even kill you. While there is no good evidence that dead people can feel pain, when people die it does tend to upset people who aren't dead yet.

So, if I understand you, you want to "control the infection" to save lives and keep people from being hurt. Is that correct?

Yes. It's on account of having a small particle of basic decency that keeps bothering me.

"Yes" would have been sufficient. Anyway, how do you know that the net effect of "controlling the infection" by means of compulsory lock-downs will be to save life and avoid pain, rather than to destroy lives and cause pain ?

Only 20 identified new cases in Australia yesterday. That's a big drop from the 3 to 4 hundred a day at the peak and represents a lot of avoided suffering compared to if we hadn't gotten the rate of transmission down.

The numbers will bounce around a fair bit, but the average over the past 3 days has been 20.

But my state ended its new infection free streak and one new case was identified. Hopefully that will be the last, but we'll just have to wait and see.

Or examples too. This is not the place for pointing to the real world, or the world outside of the U.S.

It is just virtue signalling in this brave new world which has such ignorant people in it, bless Ford.

The EU is just as bad as the US. So why don't you stop cherry picking examples and try to actually engage in the discussion.

Exactly what did Australia do that was much better than say what Germany did?

Crikey keeps ignoring the details.

Germany is now reducing new infections at about the same rate Australia did, so in terms of virus transmission what Australia and Germany are doing is roughly the same. Australia just started doing it when new infections were about 15 per day per million people and Germany started when they were at about 84 per day per million people. If Germany had started two weeks earlier then they'd be in around the same position Australia is now on a per capita basis. It was really dumb that they didn't start two weeks earlier. It cost them 5,000+ lives and it will take them longer to eliminate the virus from the general population.

When many people are getting sick and dying, people stay home whether ordered to or not. The economy would crash in either case. What’s hard to understand about that?

The point of an order is homogeneity.

If we let this thing rip in March, there would have been many NYCs, the economy still would have crashed... there would just be more dead.

The shutdowns were meant to buy time for the production of PPEs, masks, testing supplies, and to learn how to social distance so that when we do go back out, we hopefully won’t be spreading it so easily.

We still aren’t ready PPE and testing wise, but we are in a much better place than we would have been. Which is why the top down shutdown is ending real soon.

Either way though, we are going to have a deep recession. There is no way out of it.

Not expecting a deep recession here in Australia. Business is picking up as good news comes in about the virus and people who still have jobs, which is most of them who had them at the start of the year, are starting to spend. They tend to be sitting on a lot of money now they are no longer eating out or going to the pub or spending much on gasoline. Fiscal stimulus payments to pensioners and low income earners and businesses are also boosting business.

My sister's doing 50 hours a week and getting paid for the overtime. I know that doesn't sound like much work to Americans but she normally does 37.5.

Unfortunately, pointing out that many places are able to deal with a pandemic is just a waste of time here.

Starting with the fact that people like TC and AT are deeply invested over a generation in creating an environment where government is seen as the source of all that is wrong in the world. The rest of the modern world ignores them, and oddly enough, the rest of the modern world can be seen handling a pandemic better than the U.S. A reality that is getting increasingly harder for Americans to ignore, though they continue to work diligently at it.

...where exactly is the US response worse than other countries?

This is asinine.

Yep. Better than middle of the pack, and if you live outside of NYC, doing very well.

"This is asinine."

This is prior with his normal shtick.

> knew what was required to control a respiratory infection with droplet transmission.

But every country knows this. How do you know it's not that you live on an island where borders are more readily closed? What do you think has made Australia so unique to date? What actions were taken that other countries failed to take?

My guess would be the fact that you are an island with non-porous borders and you closed borders agressively is the #1 reason. Ditto with NZ and Taiwan.

The US has thousands of people per day sneaking in, and they sneak into massive population centers undetected. And even if we close borders aggressively, as Taiwan and Australia have done, it's met with cries of "RACISM!" by the media. Our media really believes that Taiwan's border is open 24x7. They want what Taiwan has, but they dont' understand how Taiwan has achieved it.

There's a concerted and growing effort in the US to absolutely destroy the economy over this. Common sense things like stopping all international travel and sealing the borders are viewed as very bad.

People that are out of work are earning more money on benefits than did when working. A large % of the population is demanding benefits for people here illegally.

"state incapacity"

There is a reason I have been pointing you to The Fifth Risk since that book first appeared.

Special Report: HHS chief Azar had aide, former dog breeder, steer pandemic task force

It turns out Michael Lewis' criticisms were not just partisan.

I’m sure you think you have a point. So make it and speak clearly.

State your hypothesis.

How is it not obvious? Lewis wrote a book length treatment on the importance of both risk assessment and then execution in government.

At HHS, Harrison was initially deputy chief of staff before being promoted, in the summer of 2019, to replace Azar’s first chief of staff, Peter Urbanowicz, an experienced hospital executive with decades of experience in public health.

This January, Harrison became a key manager of the HHS virus response. “Everyone had to report up through him,” said one HHS official.

This is literally what the book is about, a rejection of institutional knowledge within government.

New York is the worst hit state in the US.

Can you point to explicit examples where the State and City governments rejected institutional knowledge?

And of course you better value all of that institutional knowledge if you actually want to have state capacity!

Why would it take 6-12 months to bring Rt <1? The moment serious social distancing happens, Rt falls and it probably falls by a lot. At least in respect to this shut down which I think has plausibly reduced average interactions by 90%.

Rt<1 but greater than 0 means there is spreading happening but instead of 1 infected person spreading it to 1 or more uninfected, it takes 2, 3 or 4 infected people to spread to a new infected person. Each 'round' cuts the population of people with it and it can fall pretty fast.

Serious shutdown brings Rt pretty close to 0. At 0 it takes maybe 14 days for the virus to go extinct. Otherwise it spreads and Rt falls on its own as we approach herd immunity.

6-12 months IMO is possible only if we achieve the worse policy possible. Don't rip the band aid off but go back and forth trying to open then close keeping Rt close to 1. Slowing it whenever it accelerates but then losing patience before it starts to seriously fall.

Even we will probably not achieve that level of policy stupidity, although we have selected the best people to take a shot at it.

On top of this a game changer would be real testing capacity. If we could do serious testing, enough testing so that for every positive we get 9 negatives or more and do it often we can zero in on those with the virus and isolate.

Other game changers like miracle drugs, fast vaccine development are possible, even somewhat probable but should not be counted on.

Bad game changer, it's discovered having the virus doesn't give you immunity or only a very short period of immunity. Even then a serious policy of driving Rt<1 will be the solution. I don't think anyone knows for sure if ebola survivors have immunity. I don't think they would be eager to test themselves. You don't need a vaccine or immunity to defeat a virus.

Here's what has been happening in Los Angeles, from comparing case counts by neighborhood on April 1 and April 13. Granted, L.A. has highly inadequate testing ... But on April 1, cases per capita tended to be concentrated in affluent areas around the Hollywood Hills. Some of that is influential people being better at getting tested, but a lot of it was Tom Hanks Disease spreading through world travelers and the popular who get invited to a lot of get-togethers.

By April 13, case counts in entertainment industry neighborhoods had slowed a lot, but they were now growing rapidly in working class neighborhoods.

So while R0 went in sharp decline among the wealthy and/or trendy, among people who had go to work it wasn't slowing that fast.

There's still a knock on effect. Rt may not go down as much among grocery store workers who show up every day but it is going down because the wealthy retreated to bunkers and everyone else has cut back their interactions.

BTW, so what I've picked up so far is there's two R's to care about:

R0 - the 'natural' transmission rate (which I suspect is culturally sensitive so a disease like measles might be a 10 in the pre-Covid US but maybe a 9.5 in Japan with mask wearing culture)

Rt - the transmission rate at time t in a outbreak. An infection will start at R0 but begin falling since each new infection makes it harder for a virus to find a new victim. Social distancing, vaccines, strategically surrounding vulnerable people with recovered people, etc. all make Rt lower than what it would have been without intervention.

> If we could do serious testing, enough testing so that for every positive we get 9 negatives or more and do it often we can zero in on those with the virus and isolate.

I don't follow. In WA state since the beginning of all this 92% of tests are negative, with about 150K tests given so far.

Some interesting data in a NYT story yesterday...they said patient zero in WA state had 65 contacts they found via tracing. Every single one was identified, followed up with and none developed covid. But what is interesting is that it gives you a sense of the effort require to trace.

If I have you a piece of paper with 65 people I had contact with--no names, just things like "Cashier at home depot, he had a beard, about 30 and kind of balding" how long would you estimate it takes to trace them all down?

I'd venture it would take about an 1 hour per contact to find an indentify the contact. Some could be done with just a phone call, others would be more involved. Round numbers, contact tracing might take 50-100 hours of labor per illness. To do things in a timely manner, you need 4-5 people doing contact tracing on a sick person so they can close it in 2-3 days.

If you have 100 people that need tracing in a big city, your are on the edge of losing the contract tracing game I'll bet.

If your society was canniballstic, what percentage of people would you eat to stay utilitarian. This is a Abstract Tree problem. You are looking for the stable N, numbers of people. There are two groups, the eatable and the hungry. You are trying to match the two groups, keep the dinner dates spread around.

What if people are very active, do not always have time to ead when dinner dates are independent events? Independent firings is the stability as that means dinner dates will all be little surprises. The independence of dinner dates is stability.

But if people are moving around too much (this is temperature) then they will partition the diner dates between rare banquets and short fast foods. This is a standard result, not limited to physics.

Hence, I can say. The ratio of fermions to bosons remains the same as if physics were at the lowest energy.

What would Marx say? We can never get the 6.023e23 people we need and go through continual regime change.

Constants matter:

“ the death rate for all ages from COVID-19 would be 0.81% without mitigation efforts.”

Haha. No.

“ Based on life expectancy tables, the life expectancy of victims would average 14.47 years.“

Haha, no.

I’m sure that there are more glaring errors.

Yeah the death rates for C-19 seem to be about 2% of infected, not 0.81%. The life expectancy value does not seem absurd however, and commentator Catinthehat might agree with it.

Sorry Ray while it is true that IFR is constantly being updated, the data from the most recent anti-body tests are resulting in IFRs less than 0.2%.

Oh and the death rate in the paper is of the total population, not of infected. They assume 75% of the population will get it. I know that number corresponds with your suspicions. I have my doubts, but it is plausible.

The antibody tests studies have issues, many issues. Clearly, Covid 19 is more widespread than the testing would indicate, but I think 2-3% prevalence looks reasonable not a much higher number.

IFR below 0.2% looks delusional. If 100% of New York city is infected, the implied IFR would be over 0.1%. I don't believe two thirds of NYC has the disease (the number required to get to 0.2%).

Their main issues being they conflict with your priors.

Dumb answer and hardly true. They are not random samples. The false positive rate is too high. We are gathering evidence, but nothing is confirmed. Certainly nothing that would lead us to suspect an IFR below 0.2%.

Both Stanford-Santa Clara and USC-LA County conclude IFR could be <0.2%. Also, As I understand it the first study vetted the test kit and it yielded zero false positives.

In the Bronx with 100% infected, the IFR is 0.175 %

@Catinthehat - you overreached and lost credibility. Noting in statistics is "100%". Fake science and fake scientist noted. Have you figured out how the Transpose function works in Excel yet, Cat? LOL keep trying you'll get it. Show your work my fiend!

You're missing her point completely. She's saying that the IFR in the Bronx would be 0.175% if 100% were infected, not that 100% is infected.

But we know from cruise ships the NYC death rates are specious.

If you have failing kidneys, diabetes and a bad ticker and have been in a nursing home for the last 6 months on oxygen and you die from what looks like pnumonia, NYC consider that a covid death. Why? Because they get money from the feds.

That is how NYC is getting these insane death rates--twice that of Spain. Money.

Yes, they find 14.47 years for the average number of years lost per fatality. I had 13.5.
The 0.86 IFR is certainly plausible in the Bronx. It's may be a little less for the rest of the US.
-they use 3 years of consumption per year lost so one year lost = 132 k
-- I think 75% infection is plausible but that would be over a few months
-- they don't address the problem of shutdown then reopen, which shifts the deaths to the future because the susceptible are still numerous ( not 75% are infected), so maybe the 0.35 year of consumption is for repeated shutdowns ?
Otherwise the assumption is that the virus is extinguished by the shutdown, or the deaths are super spread-out because contact and trace is in place and effective and new infections are low and/or a magic bullet treatment arrives soon after the lockdown is lifted.

Most published life tables are based on underwritten lives (better than average mortality) or population lives (average mortality). Given the impact of co-morbidities, taking a mortality table at face value does not make sense since most will have worse mortality than average excluding the impact of covid. Co-morbidities can very significantly change mortality patterns and in a way that interacts with age (being a 90 year old diabetic might mean your mortality is only 1.5x an average 90 year old but that factor could be 5x for a 30 year old).

No one has the data to do this such that I know for Covid, so it's all estimates. 14.47 "feels" too high to me.

Likely far too high.

The data is there. There are CDC deaths reports with demographics. They also exist in other countries like Italy. I did this 4 weeks ago and got 13.5 Their result 14.47 is not outlandish. You have to remember it’s an average.

Yes but did you did it just based on the age of the victims of Covid-19, or also on their pre-existing health conditions (the victims of Covid-19 have an higher rate of pre-existing conditions than the general population, even after controlling for age) ?

Yes, I assumed a life expectancy factor based on the risk factor for 3 conditions : hypertension, diabetes, heart disease for people over 55

There's also an unknown factor of increased post-covid mortality due to long term lung or other organ damage. Anecdotally, a previously very fit and healthy friend reports she can't breathe properly a month after the disease and has been told it may take a year to recover. That's uncertain, of course. It would be interesting to know the impact of infection on future disablement and morbidity if we are doing these calculations. It's probably not insignificant.

I personally know a person who tested negative and she feels just awful.
Proving that it's worse than we thought.

There will be a tradeoff in consumption, so the question is what are ways re can increase production without needlessly reducing it with minimal benefit.

I really like Prof. Jackson's comment at Stanford:

"This lack of coordination may end up being very costly for the world. The difficulty is that while one area is slowing the virus down and getting it under control, it is growing somewhere else. Once it is under control in the first area, it can return from another area. An analogy is trying to get rid of termites in a house. There is a reason they fumigate a whole house all at once, even if only a few parts of it show large infestations. If you just fumigate one room, by the time you clear it all out, the termites will likely be multiplying in other rooms. Once you reopen the first room, the termites will return from the other rooms where they have been growing. You could keep fumigating rooms forever.

The lack of coordination between states within the U.S. and across countries and continents in efforts to slow this virus means that we are dealing with things one room at a time – being reactive and not proactive. It is impossible to completely seal borders – both physically and economically. Thus, without coordination within and across countries we will end up endlessly reacting to resurgences of the virus." Here is the link: He also discusses this as it applies to network industries.

So, let's take an example and look at inefficient tradeoffs in a networked market:

1. Let's assume a manufacturer depends on an input from a Georgian manufacturer, and, let's say its even a small but necessary component. Georgia decides to open restaurants and large scale events. Well, there is a local, temporary increase in revenue for the local restaurants, but, two weeks later, the Georgia manufacturing plant, like the plant in South Dakota, becomes a hot spot and has to be shut down. Balance the effect on the economy in Georgia for restaurants versus the costs to the Georgia manufacturing plant AND the plants outside of Georgia that relied on that product. If you had to choose between the restaurant and protecting the plant, which would you choose. Would you choose it only from the perspective of Georgia, or would optimally choose it from the perspective of the United States.

Sure, CDC can take care of it, centrally.

You aren't answering the questions because you cannot.

Dismal, We specifically need a national plan on national testing.

Funny how it's never the NYC subways that get used in these examples....

Sorry, but that is not true. Yesterday I did comment on how subways probably.

Your comment doesn't follow. If you talk about networked industries and networked states, you are not obligated to talk about subways, or, Trumps malfeasance and poor judgment. I could if you would like me to.

Didn't UN just talk about a 20% reduction in consumption causing 400-600 million people to go into poverty? What effect does that have on life expectancy?

Putting the 0.1% in poverty would deliver more than 20% less consumption with only 7 million in poverty.

Who would have thought my Starbucks habit was literally holding thousands of people out of poverty. Never did I think my consumption was so noble nor my holding back consumption so evil.

Surely the economic system has to be a bit more than simply a hamster wheel of 1st world consumption.....

Where exactly do you think Starbucks gets their input materials from?

Surely you can follow and understand what global supply chains mean, right?

Yes every $7 I spend at Starbucks is adding pennies of income around the world. Nonetheless there is something a bit backward with the idea that I must consume like a glutton to prevent someone on the other side of the world from starving.

On another thread I asked the question what if after this is over we discover a lot of us like the 'shut down lifestyle' and pull back a lot on consumption? The economic system has to be rational and sensible enough to adjust to such shifts without throwing the whole world upside down.

Tyler: would love to hear your thoughts along these lines with reference to Stubborn Attachments to growth and compounding. Isn't the compounded loss from these six months or so a disaster for undiscounted future growth? I've just re-read the section titled "Our obligations to the elderly", and the discussion is very relevant. There may be a moral hazard argument about obligations to those who have raised us and protected us, but I think the most relevant quote is: "while we should do a good deal to help the elderly, the logic of sustainable growth places limits on these obligations, too".

" Our baseline answer is 26%, or around 1/4 of one year’s consumption." - well on a crude level that means 365 days/yr * 0.26 = 95 days of lockdown would be optimal for most people. Sounds reasonable. Greece will finish a six week lockdown (45 days) and C-19 is about burned out. If the USA did a three month lockdown it would, if done nationwide, burn out the disease. But ignorant Americans will never allow that to happen, since it's all about jobs x3 (never mind most of these jobs are used to buy things never used, i.e. conspicuous consumption)

That's also assuming it would 100% burn out and never come back after 95-day lockdown. If we do that and then we go back to pre-COVID normal, but it comes back, we end up as Tyler suggests: losing both consumption and lives.

Coronavirus is largely spread by people without symptoms. Here’s what that could mean for reopening the economy.
This is equivalent to using live virus for vaccination.
So we essentially have the regular problem, some folks have a severe allergy to the vaccination.

The problem then is finding those who have the allergy, the vaccine will not help them they are likely to be just a allergic to the dead strain. We need an allergy test for corona.

Age is an excellent predictor of death by coronavirus. In Italy, the raw number of deaths increases more than three orders of magnitude with age.
(The population skews a bit old in Italy. The real infection rate probably skews young (schools). These skews should roughly cancel out).

So, you could vaccinate the young (less than 40 years of age) to protect the old.

But--Is it fair to ask one group to assume risks in order to benefit a different group?

That's basically what we've been doing this past month.

Avoiding the deaths from pandemic, meteor strike, climate will increased consumption because paying millions of workers to build the capital to counteract the expected dangers of mass death will increase the total incomes and thus consumption.

The cold war era from 1945 to 1975 building nukes, bombers, rockets, subs, carriers, battleships, tanks, plus millions of military on active duty created the middle class. The "peace dividend" slowed GDP growth and slowed household income growth, and increased income inequality.

Lots of fatalities in long term care facilities. It’s not clear how well the shutdown is helping them.

I think this is more tied to the previous comment on networks. Shutdown stopped visitors to nursing homes, but their isolation was only as good as the knowledge and commitment of their staff. And then the surrounding community.

No nursing home bunkered in with staff sleeping on site, right?

Yes, it belongs to the previous TC post.

"The "peace dividend" slowed GDP growth and slowed household income growth, and increased income inequality"

Income inequality in the U.S. started from 1981/82 whereas the peace dividend is usually thought of as having started from 1991. Also, the cause of the rise in income inequality was a major shift in less skilled women and immigrants entering the workforce in greater numbers in the 1980s and 1990s. Income inequality hasn't increased much from 1997 when the gini coefficient was 41, the same as in 2016.

The death-consumption tradeoff seems pretty simplistic, as long-term health consequences among those who recover are not included. See, for example, (German) and For all we know, trying to "save" GDP could, because of productivity-sapping health outcomes, cost GDP for some time to come. Not to mention direct costs to care for those who recover but remain chronically ill.

Good point. Nor does it consider the externality of a dead parent on the raising and support of children. Long term Losses accrue to others as well, and that is not included in the model.

Nor do they consider the healthcare costs of those infected who develop problems later.

Yes, the game plan is to destroy the economy on the way to herd immunity.

The recession would have happened anyway, just with people distancing themselves socially. So the choice has already been made by people.

I think there is a lot to your statement. Although the governments pretend they have 'shut down' there jusrisdictions and can therefore 'reopen' them at will, they did not and cannot. A good example is the mayor of Las Vegas saying restaurants and hotels have to reopen so people can make a living. Problem is, income does not some simply from a restaurant being 'open'... it comes from serving customers. No doubt some would show up, but the normal Las Vegas crowds will not reappear until individuals make the decision to go there... and a 'death cluster' in Vegas similar to that in NYC might influence those decisions, mayoral dictate notwithstanding.

It is very likely that a lockdown followed by a reopening would be better for the economy than a long period of latent fear, uncontrolled pandemia and voluntary social distancing. Confidence would be restored more quickly. Moreover, in the first scenario more private losses are transfered to the state which is what you want for a quicker recovery.

Well hey! If I wasn't worried about the gov't forcing me to stay in my home, I guess now I need to worry about the gov't forcing me to go to a steakhouse in Las Vegas.

Dammit, you'll eat steak in Vegas and you'll like it.

+1 to both above.

Say 70% of people return but spend only 70% of the time and money that they used to. That 'reopening' is 50% less revenue. That will probably destroy most businesses who do not have the margins to keep going very long at that level. Imagine trying to do bailouts in such environments where there is no way to tell which places are closing because the market is simply moving against them or because fear is holding the public back. Imagine the states divided between some open and thriving, others closed and others open but starving. If you thought the country was divided before imagine it at that point.

Exactly. One sensible policy response may be to introduce a new, special, temporary bankruptcy statute that lets small businesspeople file without affecting their future credit. To kick-start the economy after this thing runs its course, and to shore it up while we're still in limbo, letting loose thousands of entrepreneurial small businesspeople, free of debt, stigma, and bad credit, can only help. Encouraging weak players to leave the market would also help "right-size" restaurant markets and the like, leaving more customers for to shore up the remaining businesses.

How much would it cost to *not* shut down? The answer would have rather large error bars given the unknowns with this virus. It's possible that the virus spreading through our communities may cause significant demand destruction itself, and for what length of time?

The Fatality Rate of one year of life is 0.9%.

I fear that most smart people have honed their smarts in stable environments, leading them to think we can, with this imagined stability, quickly figure out what's going on here. Nothing wrong with trying to figure out what's going on. I'm 100% for it. We need more it. And we especially need more Emergent Ventures.

But while we're still in the thick of this, and especially while governors think through a partial lifting of lockdown, it seems more appropriate for them to ask, "What do you do when you don't know what's going on?" In that world, there aren't "tradeoffs," because tradeoffs imply knowledge we don't have. There are, instead, only bets, hunches, and hedges.

If there were a science of the day, then, it wouldn't be economics, medicine, or even epidemiology. It would, it seems to me, be a science of robustness under uncertainty.


You don't wait for the car accident to about to happen to determine if it merits wearing a seat belt.

> leading them to think we can, with this imagined stability, quickly figure out what's going on here.

Very salient point. It's good to cross check by warping back in time and asking where along the way different decisions should have been made. What is clear is that as soon as Wuhan went on lock down, the action should have triggered:

1) Has this every happened before? No, it's extremely unique. It's a "tell" that this is serious beyond words.
2) Is there a reasonable chance what they are seeing has traveled here? Yes, it's very likely. It's probably here now.
3) Is there a good chance we might see what they are seeing in our larger population centers? Yes, there's a good chance.

And the "perfect" response would have been to slam borders shut on Jan 23 AND mandate 1-2 weeks inside for 90% of workers while we got a handle on this. Countries that went through SARS know how to "press pause" on their economy. We do not in the US--full speed or nothing. But the ability to slip an economy into various stages of a coma for X days seems a needed feature moving forward.

Increasingly, however, it because clear that because Trump picked an action, it caused many govs to pick the opposite reaction, do to nothing more than the logic "If A is bad, then B must be good"

Take the subway, enjoy tightly packed events. Those that were political in their decision making have a lot to think about.

Surprisingly I don't disagree with much of this. An early shutdown would have nipped this in the bud with much less cost. If we go back a bit further, serious stockpiles of PPE that would have enabled hospitals to go into 'lock down mode' (every patient is seen with fresh PPE no exceptions) would have also made things much easier.

As I said before Trump's travel ban was a gimmick, as is his 'immigration ban' and one simply has to compare out cases with every other country in the world. Although defenders have probably discovered it was a magical success if you count infections per 100 lefthanded ginger people or something like that. Beyond that he did very little except when pushed and prodded.

In a war, most bullets will miss. There are a lot of things that should have been done better in NYC like more rather than less subway rides. On all levels of gov't the 'masks don't work' line should have been challenged and people encouraged to make their own. We aren't going to know for years whether wearing cloth masks has a dramatic impact on transmission when walking down the street or thru grocery stores but there's no plausible logic for why it would hurt.

What's perplexing about the response from Trump is that, with his commercial real estate background, you think he'd morph readily into a style of thought that gives primacy to ignorance and uncertainty. In commercial real estate, especially transactions, one recurring question is "What do you do when everybody is lying to you?" Real estate deals have high information costs, and market participants exploit that. The standard response is habitual searching for robustness under mendacity.

Perhaps you're misjudging his background. I remember reading an interview with Woody Allen where he noted that he actually made almost as much movies from real estate as films. As he sold one Manhattan residence for another he booked huge profits. Perhaps what success Trump had in real estate was more luck than skill. My understanding is he was not taken seriously by NYC commercial developers. Perhaps his skill and background is truly entertainment and reality TV where it is usually easy to drop and old role or move beyond a mistake by changing the subject to something new. That's fine for a tv show you can turn off but the virus is relentless in that it doesn't care about our boredom or impatience.

"Will we manage to lose both?"

In a sense yes. Because regardless of government action or inaction, we were going to have a pandemic and economic recession, with people either curtailing activities voluntarily or simply being too sick to work.

Imagine a Production Possibilities Frontier with two goods, "economic goods" and "people not dying". Is there a tradeoff?

Economists recognize that there are always such tradeoffs. What some might not be recognizing (and even more commenters and politicians failing to recognize) is that the novel coronavirus shifted that whole PPF inwards.

So we do face a tradeoff between the economy and lives, but it's a new unknown PPF with unknown slope (unknown tradeoff costs). They've given us an estimate of that new slope -- but needless to say there are a lot of models out there; we an't estimate the PPF without reliable estimates from the epidemiological models and we're not there yet.

What's the optimal policy? Nobody knows.

I agree with your use of the production possibilities curve but "people not dying" as a good produced is a bit awkward....

I would consider it a simple shock to the PPF curve that draws it in. Our mix of various goods is available but at reduced rates because if we continued in our previous production of, say, concerts, workplaces would start getting disrupted as people get sick. Or if we maintained production but added a lot of ad hoc mask production and sanitizing procedures to keep people healthy, that too pulls the PPF inwards.

The shutdown then represents exactly what it sounds like. Going deep inside the PPF by drastically reducing consumption. It is essentially a type of savings. When your body is sick you lay in bed all day, same principle.

Medium term, a few helpful economic policies that avoid optimizing around the disease but encourage finding a new, higher PPF may include:

--For small businesses, a simplified bankruptcy process that, by law, would not show up on their credit. We need to encourage small businesspeople to be flexible. Propping up their old business is not the way to do it. Neither is penalizing them with a bad credit history.

--A move away from need-dependent unemployment and welfare benefits and toward a universal (or near-universal) basic (but tiny) income. For economic recovery, we will need everyone's best efforts. Providing disincentives to work is not the way to get them.

--Giving up on using the Fed to prop up asset prices. Guaranteeing investor returns discourages the investment acumen we'll need to pivot economically and get out of this mess.

Except for maybe UBI, these sorts of policies don't imply much specific knowledge. None of them require any knowledge of Coronavirus.

Make a personal estimate: A year of lockdown life equivalent to how many months of free life? Assume you are 70 - you can live another 10 years if you go to a nursing home with no visitors ever, or 5 years and lead a normal life. Then do the math for each age group.

Blackboard economics, anyone?

Pretty shocked the authors don't consider DALY/QALY approaches, and just use expected years of life. And the value of a life year being 3*consumption (v = 3) is a fairly ridiculous assumption, seeing as the UK values an effective intervention at £20,000-£30,000 per QALY.

Yes, this is the correct way to think about it. But unfortunately, we have jackass govs making statements like "I will not let a single person die that could be saved" and to a large % of the population that sounds great. "If it saves just one life, this will have been worth it..." is said so often about everything. They are clueless.

That's because you don't understand human motivation. I suppose you think a football coach should tell the team "we shouldn't loose to teams that we are better than while teams that are equal to us we should loose to 50% of the time, now go achieve PAR!"

We specifically need a national plan on national testing.

Comments for this post are closed