samwise

So it coincides with the data from the countries who have been in lockdown? Interesting. As I originally said, probably to the delight of many, Trump fucked up and panicked because it came time to either make a decision, or pass the buck. Allowing the shutdown was catastrophically stupid. Just use common sense. "Every 1% increase in unemployment means 40,000 people die". We just increased unemployment, 100% willingly, by a gazillion million percent, because a low 5 figure number of old people and folks with conditions might die... Donald Trump.... "I'm not responsible for that" To be fair, Sweden is a low density country with a fairly homogenous citizenry who believes in individual responsibility to society. So they're doing some social distancing themselves without the draconian lockdown orders in place. The US saw what happened to NYC and extrapolated that to everywhere else in the country which is an assumption that is showing to be misguided with growing evidence. Every country of size did a lockdown. The idea the US had a choice to avoid recession is misleading. We had no data and global demand contracted. We don't control China, France, Spain, Italy, ect. I don't get this point because who are we blaming? Every local and state leader? It feels like a political point more than a helpful one. During Spanish Flu, some countries had minimal economic and health impact. Some got beat up in both. Luck plays a role. Preparedness plays a role. Sweden is definitely interesting. Thanks for posting frank87. It’s one of the control cases in a global study on proper policy response. The proper comparison is with its Scandinavian neighbours. If they all get off lightly, then something about the region (like population density) saved them all. But if Denmark is ok and many more Swedes die, then that would show that the lockdown helps. Or the opposite. I am also interested in an economic comparison between them. As Schwab mentioned, it’s not as if everything is normal in Sweden currently. Yes schools and hair-dressers and restaurants are not closed, but people are definitely going out less (almost comparable to Denmark or Norway) according to google’s data. So has Sweden done better economically than Denmark or Norway? Or is the net effect on the economy just as bad. Another interesting case will be Brazil, perhaps compared to Ecuador. https://www.theguardian.com/world/2020/apr/17/lula-bolsonaro-brazil-to-slaughterhouse-covid-19-coronavirus https://time.com/5822926/bolsonaro-fires-health-minister-coronavirus/ Unfortunately I don’t know much about these regions of the world. But this is a global board. If anyone can post about economic and health effects in these regions, then we can figure out what the trade off actually looks like, without depending much on models. Best case is the disease can be managed without too much economic damage. I would hope for that. Back from natural social experiments to models. Models are all we have for things never seen before, whether those models are mental or excel sheet based, mathematical or common sense based. There are now models looking at reopening. BC has talked about some partial reopening where contact between people is reduced, i.e give the virus fewer chances to transmit and reduce R0. They estimate BC is currently at 30% of normal, but could be increased to 60% of normal and things can be kept in control. Not sure yet what that means in terms of the economy. It’s still bad for airlines and hotels as discretionary travel will still be restricted. https://bc.ctvnews.ca/b-c-health-officials-say-some-covid-19-restrictions-could-be-lifted-in-mid-may-1.4901026

What happens when all the banks go full WFH and have to start paying/reimbursing employee internet access? Will they attempt to cut costs by taking on the Telcos? Good question. I don't know the answer. But can't you get unlimited plans decently cheap in most places?

Thanks for clarifying. Since you mention expertise on an internet forum, I assume you are a bio-statistician and run/consult for these kinds of trials? First, my apologies, I was not trying to be confrontational, in-fact was trying supportive of what you were trying to explain and add to it. But it came out very different. Full disclosure, we consult on clinical trials and other aspects of pharma/biotech business but not in the capacity of card carrying clinical design statistician. But sitting in some of the meetings where they discuss the trial design really helps me understand the complexity, limited information and very aggressive deadlines people in this field have to work with. Hence my deference to them. They surely get it wrong sometimes but their batting average will put anyone else to shame if they tried to do it a few times without the expertise. Thanks. Yes those people are good, and they charge big bucks. Glad to have someone here with the relevant experience to talk about trials. I think it will be a recurring topic.

When does speculation become investment? When do you decide that conviction is sufficient to put 10% of your net worth in a specific investment? ---)You need to define a threshold. I fully realize that comparing an investment decision to a life or death situation is different but you (we) need a framework. It has been suggested (from various observational and rational reasons) that remdesivir could result in better outcomes than doing nothing for CV. We don't know the answer to this question. In order to answer that question, trials that involve randomized blind trials with control groups will be essential unless several trials done in different centers with different patients clearly show an advantage (pre-defined targets). When patients are recruited for the studies now, they have a choice to accept (and potentially get drugs that are potentially useful {or not and with negative side effects}) or not. Given the difficult situation, patients may get access to remdesivir through a company-sponsored expanded access program or through a compassionate program. The results do look promising and that should speed up the process. The approach is obviously not perfect but has provided the shoulders on which future generations could see. How could you know if you don't know? The pharma cemetery is full of ideas that looked good at some point. +1 As usual you said it better than I did CB.

Thanks for clarifying. Since you mention expertise on an internet forum, I assume you are a bio-statistician and run/consult for these kinds of trials?

https://www.bloomberg.com/news/articles/2020-04-16/gorman-sees-morgan-stanley-future-with-much-less-real-estate Yes says the CEO of Morgan Staney, Covid survivor. Bad for office REITs I guess, good for slack and MSFT.

Cherzeca, its an important question, but this already happens with other life threatening illnesses. I don't think its unethical to give a placebo if you *don't know* that the drug works. If it doesn't work then both people getting the drug and getting the placebo are going to have statistically similar outcomes. Thats the point of the trial: to see *if it works*. It would be unethical to charge for a drug which doesn't work, or has bad side-effects/interactions. Thats what doctors did for centuries, and it has only recently improved. Lets keep the improvement. Empirical evidence is very important, probably more than theory, as proved by the story of the doctor who advocated hand washing and ended up in a mental asylum since no one would believe him. https://www.npr.org/sections/health-shots/2015/01/12/375663920/the-doctor-who-championed-hand-washing-and-saved-women-s-lives Other have considered the ethics, and the current scenario is the first place where they say a placebo trial is justified. From: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3844122/ Ethical analysis and international ethical guidance permit the use of placebo controls in randomized trials when scientifically indicated in four cases: (1) when there is no proven effective treatment for the condition under study;

Is 3% much above the false positive rate for these anti-body tests? Someone posted above that the rate of false positives might be 5-10%. lets assume it is down to 3% now. If thats true, you would see 3% positive results in absolutely clean samples with no infections. Thats noise, not signal. Bottomline is: When you have an inaccurate test, the numbers need to be much higher than the test's accuracy to have much significance. This is a known issue in disease tests, which I remember reading about as an undergrad. See the example below, which can seem uninutitive until you go through it. They actually use known disease incidence as an input, but unfortunately thats what we are trying to figure out here. http://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/BS704_Probability/BS704_Probability6.html "A patient goes to see a doctor. The doctor performs a test with 99 percent reliability--that is, 99 percent of people who are sick test positive and 99 percent of the healthy people test negative. The doctor knows that only 1 percent of the people in the country are sick. Now the question is: if the patient tests positive, what are the chances the patient is sick?" The intuitive answer is 99 percent, but the correct answer is 50 percent...." Because the 1 person who is sick will test positive. 1 out of the other 99 will also test positive (false positive). Of those 2 who test positive, only 1 has the disease. So 50% chance.

I was hoping Sweden would be the "control" here, so we would see what the effects of no lockdown could be. But I don't see much material difference in Sweden's mobility versus Denmark or Norway. People everywhere are going out less. Yes there is some differences, but they all are minor when compared to Italy, where you can see an extreme drop in people going out. As an effect of this, their economic numbers should also be pretty bad. Anyone from the region who can shed any light on the movement of people, shops and economic impacts in Sweden? Thanks in advance. Sweden: https://www.gstatic.com/covid19/mobility/2020-04-11_SE_Mobility_Report_en.pdf Other Nordic Countries: https://www.gstatic.com/covid19/mobility/2020-04-11_DK_Mobility_Report_en.pdf https://www.gstatic.com/covid19/mobility/2020-04-11_NO_Mobility_Report_en.pdf Italy: https://www.gstatic.com/covid19/mobility/2020-04-11_IT_Mobility_Report_en.pdf

China true friend of Canada? You send them the money and they send you the goods. What does the USA do?

Ontario Premier: Why do I get the feeling that 3M mask market share is about to take a dive in Canada through no fault of their own? How much you want to bet that this same conversation is happening in many other industrialized countries around the world? How much do you want to be the conversations aren't just about 3M masks but encompass a whole range of products made by American companies? +1 Fair weather friends are not friends at all.

Thanks for sharing. They have it at individual level, but can only present aggregated results for privacy reasons. I will repeat that this is useful information for contact tracing at an individual level. This data is really good. Both NY and Ontario are doing a terrible job compared to Italy, as of March 29th. Italy park visits are down 90%, but in NY only 47%, and Ontario only 14%... as of March 29th!!!! I know we just came out of the freezer and everyone wants to go out.. but only a 14% drop in park visits ? India is down 57%,, probably because the government has scared people much more. Italy retail and recreation visits are down 94%, but NY is at 62%, and Ontario at 59%. Who are these 40% of people still going to "places like restaurants, cafes, shopping centers, theme parks, museums, libraries, and movie theaters."

Thanks for sharing. They have it at individual level, but can only present aggregated results for privacy reasons. I will repeat that this is useful information for contact tracing at an individual level.

time to short life insurers? Or worth buying a put as a hedge if the virus can’t be contained? Possibly health insurers too? https://www.ft.com/content/b2fb2326-6d13-11ea-89df-41bea055720b I did the calculations below a couple of weeks ago, tried to buy more life insurance and failed. Seems insurance policies are mispriced for the current environment but can’t be changed now. Unfortunately I understand nothing about life insurers, e.g. is there pandemic reinsurance they bought from BRK? are the reduced costs of LTC care and annuity payouts hedges for them? Maybe it’s already priced into the put volatility. Actuaries are definitely watching this: https://www.soa.org/globalassets/assets/files/research/projects/resrch-li-mortality-delphi-study.pdf Life insurance is a long investment now! You don’t even pay a premium like you usually would for tail risk hedges. The ask from a reputable life insurer was 0.16% per year for a term policy I could cancel anytime and get without any medical checkups etc, ( probably hard while socially distant). This matches usual mortality risk for my age/sex https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1310071001&pickMembers%5B0%5D=1.1&pickMembers%5B1%5D=3.2 The current “consensus “ numbers for payoff are 80% chance of getting COVID-19 before herd immunity 0.4% chance of dying from it at my age range. Estimated chance of “payoff” = 80% of 0.4%= 0.32%, I.e. double the price. I’ve tried to hit the ask, but the insurance company won’t get back to me. This is a pretty illiquid market and price hasn’t moved yet, but it seems there are no policies on offer. Has anyone else had the same experience?

Neat idea. It could be improved. Doesn’t google know where everyone is and has been? They can even tell me how busy a restaurant or road is right now! Let them contact trace every person who tests positive and send a message to everyone they were within 6 feet of in the last 3 weeks “ you were in contact with a Coronavirus case, please come to the nearest test centre at once. We know who you are and where you are. If you don’t report in 1 hour, the police will come get you under the SAVE act!” SAVE : Saving Americans from Virus Epidemics Yes can’t be done politically, but technology is already here. You would still need to trace the old fashioned way those people google can’t find. Then maybe there is a better chance to keep a lid on infections after case numbers have been reduced. So instead of the lives versus GDP, we can now have the liberty/privacy versus lives versus GDP debate. I know New Hampshire’s position on this “Live Free or Die”. But what about everyone else? [in accordance with thread protocol and 300 pages of tradition, throw in gratuitous insults aimed at anyone who disagrees here]

Question for those who follow refiners. Does the crack spread get affected by crude prices? I would have thought it shouldn’t. Why are refiners down so much more than the market then ? Is it just an anticipation of lower demand and capacity utilization. Or is something else . Thanks in advance.

Aurelius, what you are saying is that it is possible things are much better than people fear. It is definitely possible since we know so little yet about this three month old virus. The opposite of Iceland is Italy. It bothers me how little of the reporting ever mentions confidence intervals or any measure of uncertainty around their estimates. Those ranges in terms of total lives lost are enormous, from a few hundred thousand to millions. Unfortunately decisions are required now (even the decision to wait and watch), before a full study can be done. They will be proved right or wrong with hindsight, after lots of economic and possibly human damage.

Isn’t that healthcare risk unique to the USA among most developed nations. Most OECD have socialized healthcare. So yes the USA may have to calculate its costs differently than other countries.

Can’t we move the political discussion somewhere else? Meanwhile it’s true that eventually we’ll have to relax restrictions in every country, but this article estimates 7 weeks when it’s a very strict lockdown. And then an “open society” like South Korea, with lots of testing, contact tracing (forget your privacy) , isolation of cases, no large crowds, face masks etc. That’s not exactly a return to normal. This article posted by Liberty is still the best description I’ve seen. If you have those controls in the first place, you never have to go through a lockdown, hopefully. But there is only 90 days(!) worth of data on this virus, so there is huge uncertainty around any of these statements. https://medium.com/@tomaspueyo/coronavirus-the-hammer-and-the-dance-be9337092b56

+1 printing money is one way to share the costs among society, but it won’t be equally shared. Those with fixed claims(cash or bonds) will suffer as the value of money goes down. Those with debt will have it easier. So it’s not quite a free pass with no redistribution of wealth. But I don’t see a better way either. This is possibly the best available option, unless you accept mass bankruptcy in people and businesses with its long term effects.

Liberty, thanks for sharing. First article I’ve seen which describes a plausible long term strategy.

Given how sporadic COVID testing has been implemented globally, this will probably be the measure-of-choice in 2020 as well, now almost 100 years later. Look how far we've come... :-\ Yes, unfortunately. https://www.latimes.com/california/story/2020-03-20/coronavirus-county-doctors-containment-testing?_amp=true&__twitter_impression=true

Agree with posters who say that a death is a death, regardless of cause. Instead of measuring deaths from COVID19, the relevant statistic is excess deaths. That is how mortality from 1918 influenza is commonly measured, since there were no tests for influenza at that time. This is the first statistic on excess death which I have seen. In the town of Bergamo in North Italy, deaths in the first two weeks of March went from 56 last year to 164 this year. Not all are attributed to COVID19, since many were not even tested. This is not a scientific study, just the mayors anecdotes. But it’s scary, and accounts for heart attacks, traffic accidents and all other causes of deaths in that town. Maybe they had more heart attacks from anxiety, or distracted driving, or undetected COVID19. Every one is free to form their hypothesis, I know what I would pick. https://www.reuters.com/article/us-health-coronavirus-italy-homes-insigh/uncounted-among-coronavirus-victims-deaths-sweep-through-italys-nursing-homes-idUSKBN2152V0 A few more towns that react late and we will gather enough data for a more solid conclusion which can help resolve the debate. Of course, their may be ethical considerations which might override the need for better clean data. People are on either side of that debate depending on their view of the ultimate excess death and life years lost.

I am also interested in this. Specially the prices of future tickets. I’ve heard anecdotally that current ticket prices are depressed but prices in the fall are around normal. This is a good indicator to track people’s future expectations.

Agreed. From my first hand experience. That’s exactly what every Canadian doctor has told me when I had a viral fever. Sure beats enemas, tree bark, mercury and bloodletting which respected doctors did during 1918. https://blogs.scientificamerican.com/observations/the-1918-influenza-pandemic-how-far-have-we-come/ Medicine has advanced by knowing their enemy and knowing their tools to treat the symptoms, but not found a cure yet. So testing may have limited clinical advantages. I don’t know if it helps limit the spread now. Sounds like Whack-a-mole. Testing does help gather data about your enemy. Data is the basis of all science, medical to cure a single case, or epidemiology to understand the spread of a disease in a population. We need both sciences to advance. We had 5(?) viruses emerge in the last 20 years (SARS, MERS, H1N1, Avian flu, Covid). I don’t think this stops with this battle. Physicians may be the troops, but epidemiologists are our scouts and spies. If we don’t listen to our scouts, we could have issues. We either have bad scouts or bad judgment.