On April 15, I did a post arguing that Sweden is not the right Covid-19 model for libertarians, rather Taiwan is the model. Now that we are in September, it’s time to revisit some of the arguments.
One argument is that countries trying to control Covid-19 were merely delaying the inevitable. You hear people saying “we’re all going to get it eventually”.
But are we? Russia and China are already beginning to roll out vaccines, and Western countries are expected to begin doing so relatively soon. Back in mid-April, Taiwan had suffered 6 deaths in a population of 24 million—today its death toll is 7. Sweden has suffered 5813 deaths in a country of only 10 million. It’s not obvious to me that everyone in Taiwan will get Covid-19 before vaccines are available.
Sweden’s GDP has fallen by 7.7% over the past year, which is better than the European average but worse than the average performance of its Nordic neighbors (as I predicted). Taiwan’s GDP is only down about 0.1% over the past 12 months. (Down 2.4% over the past 6 months, but even there it does much better than Sweden.) Thus while both countries refused to shut down schools and restaurants; Taiwan did far better in terms of both health outcomes and GDP growth.
The so-called “sophisticated” argument against my Taiwan/Sweden comparison focuses on structural differences between the two countries. And indeed there are many cultural, political and technological differences that do help to explain the differing outcomes. So (it’s claimed) perhaps there was no chance that the Swedes would undertake the Taiwanese approach in early 2020. I agree.
But I’d actually call that a “pseudo-sophisticated” argument, as on close examination it’s not very helpful. I do have a deterministic view of how history plays out, so I don’t disagree with the claim that Sweden was unlikely to adopt the Taiwanese model in early 2020. But that misses the whole point of analysis and criticism. The point is not to rerun history—that cannot be done—rather the point is to learn lessons so that we are better prepared next time around.
Let’s suppose that in 10 years another dangerous virus emerges in China. I am confident that people will recall that the Taiwanese approach worked far better than the Swedish approach. Thus a key “pre-existing condition” that led to the Swedish failure in 2020 will no longer be operative. The public and politicians will understand that with an aggressive system of mask wearing plus test/trace/isolate it will be possible to prevent the disease from becoming widespread in any country with reasonable state capacity and civic virtue (which describes Sweden.) People will know that Taiwan was able to keep its schools and restaurants open, its economy running, and also avoid thousands of pointless deaths. That’s a powerful example.
Or at least they’ll understand that mask wearing plus test/trace/isolate work if I have any say in the matter, which is why I continue to post on this issue.
READER COMMENTS
Jose Pablo
Sep 2 2020 at 4:33pm
Mask wearing + rapid-test (rapid is key) availability seem to be a sensible, doable, effective approach.
I don´t understand, though, the insistence on “tracing” (meaning, I guess, “institutional” tracing)
In my personal experience, the first thing a normal person does after testing positive is inform all the people he/she has been in contact with for the last days. No “government appointed” tracer required. What for?
Real case: a tracer called this friend of mine (that had tested positive) 5 days after the positive test. After being informed that my friend has spent time with his sister and brother in law, asked if he should call them to make them aware of that. My friend, politely and calmly, informed him that they knew since just a couple of minutes after my friend had got the result himself.
Speed
Sep 2 2020 at 7:18pm
Your real case might also be called an ideal case. It also doesn’t trace back to the find the source of your friend’s infection.
“Although contact tracing can be enhanced by letting patients provide information, medication, and referrals to their contacts, evidence demonstrates that direct public health involvement in notification is most effective.”
https://en.wikipedia.org/wiki/Contact_tracing
More at the link.
Jose Pablo
Sep 3 2020 at 11:34am
The conclusion you mentioned refers to STDs which are a very different kind of animal. I will be very careful sending a WhatsApp to people I have had sexual intercourse with, even to remind them of this fact, much less to let them know about the STD.
I doubt the conclusions can be extrapolated to covid. Doing the tracing yourself has, at the very least, two significant advantages:
1.- Speed; you can inform people of your positive test in minutes. Going thru the process of informing the tracer, going thru your contacts, the tracers (most busy when most needed) calling your friends …
2.- Information value: the “anonymous” information provided by the tracer will not give you any clue on the amount and type of contact with the infected person.
I really cannot see any reason why “institutional tracing” could be superior in the Covid19 case.
Scott Sumner
Sep 2 2020 at 10:22pm
I did not say the tracing needed to be done by the government.
Jens
Sep 3 2020 at 9:35am
If I were infected, I would probably tell people I care about. But I would not inform people who I do not care about and especially people who I do not want to know that I am infected. I would expect a tracing service provider or agency that does this job to keep my identity secret. That won’t always be enough to keep the information about my infection secret, but it is the minimum.
Jose Pablo
Sep 3 2020 at 11:19am
Keeping your identity secret makes the whole tracing much more inefficient since it provides no information on the amount and type of interaction with the infected person. It is, definitely, not the same the covid19 positive testing person you have been in contact with being your girlfriend or your butcher.
I don´t get the “people who I don´t care about” part, either. This is not like inviting someone to a party. Do you wish to increase the possibility of the “people who you don´t care about” dying? This is serious hate there. And, at the end of the day, by letting them know you are not protecting them you are mostly protecting the rest of your community. Very likely, the people who you care about and that has been in contact with the people who you don´t care about.
Jens
Sep 3 2020 at 11:31am
Not everyone you meet is in your own community. This can also have something to do with hatred. But it doesn’t have to be like that. An intermediary who delivers the information only needs to convey the information to the endangered that they have been endangered. This can e.g. also be his piece of software. Anyone who was at risk should get tested. I think the efficiency increases due to the fact that there was contact, compared to random tests. And the (halfway) anonymity of the procedure increases the willingness to participate.
Warren Platts
Sep 3 2020 at 11:36am
Sir, that is morally reprehensible and tantamount to attempted murder.
Jens
Sep 3 2020 at 1:18pm
You have a point. So is it not just a moral but should also be a legal obligation? (possible)
I still prefer an intermediate. There are probably people I don’t want to tell that I’m infected.
In a sense, it’s attempted murder to drive a car.
JayT
Sep 5 2020 at 12:00am
Someone you “don’t care about” could include the person that was sitting next to you on the bus. How would you suggest I track that complete stranger down to let them know about my positive test? At least in theory, a tracing program could at least let everyone on the bus that used a refillable bus pass know that they were on a bus with an infected person.
Alan Goldhammer
Sep 2 2020 at 5:55pm
Taiwan are also developing their own COVID-19 vaccine that is in clinical trials. It looks like they are not necessarily going to rely on other countries for a vaccine.
DeservingPorcupine
Sep 2 2020 at 8:38pm
Sweden will have a normal number of deaths by year’s end. Short-term temporal substitution. They have lots of especially frail old people in care homes that died.
Sweden’s approach is better because it resulted in no excess deaths over non-trivial time periods and is *robust*. It could be applied in lots of countries, even though with legal systems like ours.
Scott Sumner
Sep 2 2020 at 10:22pm
This is completely incorrect. People die from Covid about 10 years before their time. And note that Sweden’s economy is also far worse than Taiwan’s—it was far from a sensible strategy.
DeservingPorcupine
Sep 3 2020 at 10:17pm
I don’t believe this is correct. When well over half the deaths are in care homes where life expectancy is less than a year, how can the average loss of years be 10? Not possible. Just look at their total, population-adjusted mortality for the first half of 2020 compared to previous year’s on euromomo. It was a slight bump from the last few years but is about the same as 2015.
DeservingPorcupine
Sep 4 2020 at 10:39am
Here’s one source. This is just for Stockholm, but of course that’s also where the bulk of the deaths are
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3609493
Scott Sumner
Sep 4 2020 at 12:54pm
That’s only for nursing homes. The studies I cite look at all deaths. Many people die 20 years before their time.
DeservingPorcupine
Sep 4 2020 at 2:18pm
That doesn’t seem plausible. If 70% of deaths are in care homes, and they have a life expectancy of, say, seven months, then to get to a weighted average of 10 years lost, the remaining 30% of deaths would have to be cut short by about 30 years, which is impossible to believe given that how rarely young people die (even accounting for conditional life expectancy of those who make it to, say, 50).
Scott Sumner
Sep 5 2020 at 5:49pm
There are a lot of “ifs” in your comment. I don’t think either you or I know exactly how much Covid shortens lifespans on average, I’m just citing a study by experts. I’ll defer to them unless another study shows otherwise.
robc
Sep 2 2020 at 9:28pm
The first question is if Taiwan’s test/track/isolate system is moral? If yes, then your analysis may be right. If no, the end results dont matter.
I dont think it passes the categorical imperative.
(Not sure Sweden does either, but its closer)
Warren Platts
Sep 2 2020 at 10:32pm
This is leaping to conclusions imho. It could be that for whatever reason (genetic or previous immunity to similar viruses or some other “immunological dark matter”), Taiwanese are generally less susceptible to the virus than Swedes. In effect, it could be the case that Taiwan had a pre-existing herd immunity before the pandemic even started. Both countries could have followed identical protocols, and the outcome might not have been much different than what we observed anyway.
Alan Goldhammer
Sep 3 2020 at 9:08am
There is no shred of evidence in any paper that I’ve read to date that argues this point and plenty that point to exposure to circulating coronaviruses not providing protection against SARS-CoV-2. Taiwan took steps that some might consider extreme because of the close call during the outbreak of SARS-CoV in 2003. There is a lot of magical thinking going on right now from those who have no knowledge of the contemporary literature on COVID-19. Ignorance of common public health practices has already contributed a lot of economic damage to the US. University of South Carolina just reported 1000 COVID-19 cases among the returning student body. Is this acceptable?
Warren Platts
Sep 3 2020 at 12:59pm
Correlation does not equal causation.
Moreover, it is not just Taiwan. All the east Asian countries have extremely low mortality rates. The highest is Hong Kong with 12/million; Japan, next with 10/million. None of the European countries come close (with the strange exception of Slovenia–that also happens to border northern Italy, one of the worst hit regions). In Germany, who is touted as a good example, the mortality rate was 112/million–about 10X worse than the worst hit Asian countries.
The difference is apparently not genetic. I just did a BOTE of Asians in USA who have died: their national mortality rate is on the order of 385 deaths/million. Indeed, more Asians have died of the virus in USA than in all of Asia combined (assuming the Chinese numbers are not too far off–and I am inclined to believe them now). So nothing obvious there.
Zeke5123
Sep 3 2020 at 6:26pm
Given the data coming out re T cell immunity and the history of coronaviruses in East Asia coupled with the numbers you cite suggest maybe they already had herd immunity.
It really is sloppy thinking by Sumner to think all other things equal.
Todd Kreider
Sep 4 2020 at 1:59pm
This.
I also don’t see how we can trust China’s case and death numbers in Wuhan or anywhere else. The virus got into Sweden when a bunch of infected skiers returned from Italy, which did not happen in Taiwan, Italy and Finland.
Jon Murphy
Sep 3 2020 at 9:28pm
True, but you at the very least need correlation to suggest causation. Alan’s point is you don’t even have that.
Mark Z
Sep 3 2020 at 8:55pm
This has been studied to some extent. Here’s a link to a pretty interesting paper: https://jvi.asm.org/content/94/13/e00510-20.
There is considerable variation between ethnicities in the HLA genes (HLA-A, HLA-B, and HLA-C), which code for proteins on the surface of T-cells and are responsible for activating T-cells when they (the HLA proteins) bind to bits of virus proteins. Different HLA variants likely do confer different levels of protection against the virus, so people certain ethnicities may be more likely to be susceptible to it than others. Figure 5 of the linked paper shows a map of the three ‘best’ HLA allele frequencies around the world and the three ‘worst’ alleles. According to this paper, it looks like east Asias may be slightly more likely to have the worst alleles and less likely to have the best ones than Europe and the US. These particular alleles may only be part of the story but HLA variation is unlikely to explain differences of the magnitude of Taiwan vs. Europe and North America.
Daniel Klein
Sep 3 2020 at 8:09am
Dear Scott,
I’ve been following you on Sweden Covid, and been a bit puzzled. In a reply above you write apparently with Sweden in mind: “People die from Covid about 10 years before their time.” Why do you believe that?
But more importantly:
Have you addressed the question about the reliability of data coming out of Taiwan? I think that one reason that people focus on Sweden rather than, say, Taiwan, is that people just don’t always trust governments and their data, and with good reason.
Scott Sumner
Sep 3 2020 at 2:48pm
Here are some studies:
https://thehill.com/changing-america/well-being/longevity/497097-those-who-died-from-covid-19-lost-more-than-a-decade-of
https://www.sciencetimes.com/articles/25448/20200424/covid-19-patients-losing-13-years-lives-average-study.htm
I was responding to a claim that Covid deaths reduce life spans by less than one year.
On the data, even if the death toll in Taiwan were 10 times higher than reported, it would not change my conclusions. And I see no reason to be less trustful of the Taiwan data than the Swedish data. Indeed I suspect the Taiwan data is more accurate, and that there are more unreported deaths from Covid-19 in Sweden than in Taiwan.
Daniel Klein
Sep 3 2020 at 6:01pm
Regarding, “People die from Covid about 10 years before their time,” you provide two links, both of which are based on a single study out of Scotland.
Here is a sentence from the abstract of a paper by Stern & Klein: “Those who died of COVID-19 in Stockholm’s nursing homes had a life-remaining median somewhere in the range of 5 to 9 months.”
Now read the short “Methods” section of the Scotland paper, and search the paper for the following terms:
“nursing,” “carehome,” “care home,” “elderly care,” “sheltered housing,” “home.”
DeservingPorcupine
Sep 4 2020 at 12:44pm
Ah, thanks. I just posted this same thing above because I didn’t see it here. It’s very clear that Sweden’s deaths didn’t knock off anywhere near a median of 10 years of life.
Scott Sumner
Sep 4 2020 at 12:55pm
I’d put more weight on a study that looked at all deaths over one that just looked at nursing home deaths.
Daniel Klein
Sep 4 2020 at 4:09pm
C’mon Scott. The point is that any responsible scientific calculation should make use of important intel about the reality, and one piece of important intel is the high rate of nursing-home deaths (50% in Sweden; plus another 20% with home-care services). Rather than building that intel into their calculation they seem to suppress that intel entirely, as the words I listed do not appear in their paper.
Scott Sumner
Sep 5 2020 at 5:52pm
You may be right that there are flaws in their research, which just means that further studies are necessary. Nonetheless, I very much doubt that their estimate is off by more than a factor of 10, which is what’s required to justify the comment above that I disagreed with.
Zeke5123
Sep 3 2020 at 6:14pm
Both are from April and only about a month and a half after this plague really started in earnest. Is that info really robust?
Jon Murphy
Sep 3 2020 at 9:17pm
I’m with Prof. Klein here.
In the US, approximately 58% of the COVID deaths have occurred in the age range of 75+. If we expand it to the 64+ rage, that number jumps to ~80%. Given the US average age is 78, I find it hard to believe that COVID chops on average 10 years off a person’s life.*
*The fact the mortality data isn’t broken down by exact year makes my point a little muddled, I admit that.
Scott Sumner
Sep 4 2020 at 12:57pm
The life expectancy of Americans is 78 at birth. If they survive to age 75 then life expectancy is another 12 years.
Jon Murphy
Sep 4 2020 at 1:23pm
Ah. Foolish me. I didn’t think of that. That’s a good point
Capt. J Parker
Sep 3 2020 at 3:32pm
When talking about coronavirus containment, thou shalt not mention Tiawan’s February 6 travel ban with mainland China and Hong Kong. Trump had the right idea but failed to take is far enough. The “Trump is a xenophobe” progressives at the New York Times, on the other hand, had it dead wrong.
Jon Murphy
Sep 3 2020 at 4:00pm
There are some 160 countries with travel bans. Why did the ban only work in Taiwan and not the other 159?
robc
Sep 3 2020 at 8:12pm
How many did it as early as Feb 6.
Jon Murphy
Sep 3 2020 at 8:59pm
I am working on that, but the data is less centralized. At least 40.
Mark Z
Sep 3 2020 at 8:33pm
It was more that he didn’t do it soon enough. The idea that travel bans are xenophobic is imbecile, but the Trump admin vigorously slammed the barn door shut after the cows were already out.
Scott Sumner
Sep 4 2020 at 1:01pm
The Chinese travel ban would have been a good idea if done earlier and followed up with other measures. As it is, it was too late to do any good. By that time there were already Covid cases in America, and the US government had a policy of discouraging every single thing that would reduce the epidemic (masks, testing, social distancing, etc.) By the end of January there were virtually zero Chinese Covid cases traveling to America, even without a travel ban. China was in virtual lockdown.
robc
Sep 4 2020 at 1:37pm
I know France has verified a case as early as late Nov/ early Dec. I dont know if the US has any verified that early, but probably. It was too late to slam the door closed by the time anyone had heard of the virus.
Todd Kreider
Sep 4 2020 at 12:18am
Scott still doesn’t understand that something radically different is happening in Asia with respect to Covid-19. According to American media, NY Times and friends, Japan was the laughing stock of Covid-19 “management” yet Japan has 60 times fewer deaths than the U.S. per capita. Oh, it must be that they wore masks… (eye roll)
Oleg
Sep 4 2020 at 11:10am
Indeed. In Taiwan, at least, I struggle to see the evidence that mask wearing or not had any impact on the result. They did very effective track and trace. I see no basis to assume that that is not responsible for 100% of Taiwan’s success at suppressing the virus (along with being an island, of course, and putting strict quarantine controls on those that entered).
Scott Sumner
Sep 4 2020 at 1:06pm
Actually, I was one of the very first bloggers to focus on how results differed dramatically by region. So your claim about me not understanding this fact is completely false. Go back to my early posts about it being a mostly white country disease. I’m not sure why you continue to make inaccurate comments about my views.
Todd Kreider
Sep 4 2020 at 2:12pm
I haven’t made any inaccurate statements about your views. For example, at first you didn’t think there would be more than 30,000 Covid-19 deaths but then changed that. Your claim among differences was that different regions had different humidity levels, and I don’t think that is why there are such huge differences between the East and West.
Scott Sumner
Sep 5 2020 at 5:55pm
The fact that you completely change the subject here is telling. I’ll take that as a tacit admission that you now understand that you misrepresented my views. Is it so hard to say you are sorry?
Jose Pablo
Sep 4 2020 at 10:06am
I don´t think “countries” are a relevant “data aggregation unit” for analyzing Covid19 policies. Despite its global reach it is a very “local pandemic”.
South Korea has 6.4 death per million residents compare with 566 for USA. But if you try to draw conclusions on policies from this fact, you will oversee that in Daegu there has been 65 deaths per million vs 2.5 in Seoul (20 times less with the same Covid19 policies) and that New York has had 1.670 deaths per million and Oregon 122 (14 times less).
Imagine the following game: You throw a dice. Then if you get 1 you multiple the number times 10, if you get 2, times 100 and so on. After 2 rounds one player has got 2 and the other has got 2,000,000. Looking at this score, I would not draw any conclusions regarding the skills of the players throwing the dice.
I do believe that, like in so many cases, government policies for Covid19 are much less relevant that we tend to think (and some orders of magnitude less relevant that our politicians seem to believe)
Scott Sumner
Sep 4 2020 at 1:04pm
Keep in mind that differences within countries such as China and South Korea are partly due to government policies. The very bad Wuhan result occurred before the lockdown, the better result in other provinces partly reflects the lockdown.
But you make a good point—it can’t be just genetics as the results within China and South Korea vary so much by region.
robc
Sep 4 2020 at 1:34pm
I dont know much about Korea, but I would expect lots of genetic differences between regions in China.
Warren Platts
Sep 5 2020 at 12:34pm
It is hard to say what the immunological secret sauce is: a simple genetic explanation, however, doesn’t explain why people of Asian descent in USA death rates are measured in a few hundreds per million compared to single digits/per million in Asia. Similarly, African Americans are dying of SARS2 at rates double their proportion of the population, but death rates in Africa itself are pretty low, at least for now.
I suspect it is probably a function of diet (like 35% of Americans are clinically obese compared to 5% of Japanese), pre-existing immunity, and local, superspreading events in stuffy environments with high viral loads that might overwhelm any pre-existing immunity. (E.g., Wuhan, welding people inside their homes may have exacerbated the epidemic; I have heard in Japan there is a tradition of households opening all doors and windows for 20 minutes a day to let out the evil spirits..)
Matthias Görgens
Sep 7 2020 at 2:21am
Germans are also really keen on airing out their residences.
Scott Sumner
Sep 5 2020 at 5:58pm
robc, China population is more than 90% Han (including Wuhan), so there’s far less genetic variation than in other big countries like the US, India and Brazil. It’s not even close.
Jose Pablo
Sep 5 2020 at 1:17pm
I do think that drawing conclusions on government policies from the number of deaths is a very risky business:
1.- Deaths per million is a variable with a “probability distribution”
2.- The shape of this distribution very likely depends on several drivers that can be clustered in two main categories:
“Non-actionable drivers”: like average age, genetics, prevalence of pre-existing conditions, level of people interaction (very likely bigger in Puerto Rico than in Stockholm), etc.
“Actionable drivers”: like lockdowns, mask wearing, social distancing, etc…
3.- This drivers “define” the shape of the distribution, but the distribution, in any case, follows a “power law” pattern, which means you can have very extreme differences in “observed values of the variable” even within the same “distribution” (meaning “for the same set of policies”).
If this is the case:
a) The real shape of the distribution(s) is very difficult (or imposible) to know from the available sampling(s)
b) Even if the distribution could be derived, the “average” value is meaningless to draw any conclusions with practical implications. The “political effort” to “manage the distribution” must focus on limiting the risk of “very fat tails”, not in managing the “average” or even higher order statistics.
c) Even for the optimal “politically managed distribution” very high “observed values” of the variable can happen and the overlapping with the “very poorly managed distribution observed values” could be very significant.
d) This makes very difficult to form robust judgments on the real effectiveness of the measures adopted by the government, since they affect the “shape” of the distribution which is almost imposible to observe.
e) The “politically managed” part of the “distribution shape” only affects marginally the “actionable drivers”, since even in the absence of “mandatory” lock downs or facial coverings, people would be taking measures by themselves and this voluntary measures will be self-balancing: they would be more radical the more deathly the outbreak.
If this is the case, the government policies would only marginally affect the “actionable drivers” which, themselves only partially affect the shape of the distribution (difficult to know how relevant they are compare with the non-actionable. My guess would be “less relevant”) and, still, even for the same “shape of the distribution”, very significant differences in the “observed values” could be obtained.
A couple of articles on this line:
https://www.santafe.edu/news-center/news/transmission-t-024-cristopher-moore-on-the-heavy-tail-of-outbreaks
https://nassimtaleb.org/2020/05/tail-risk-contagious-diseases/
Warren Platts
Sep 5 2020 at 1:08pm
If there is one lesson to be learned in this fiasco, it is that the question of the correct policy response is profoundly uncertain. All we have are educated guesses, and nobody, let alone so-called “experts,” can agree what is for the best.
Given such profound uncertainty, we ought to stop trying to make the “right” recommendation tbqh. If future history judges that we hit upon the correct response (in that the hoped for outcome was realized), (a) it would be largely a matter of luck; and (b) it is hard to say whether the same outcome would not have happened without the policy response.
Therefore, the goal should be, imho, to make the “right mistake”, if that makes any sense. That is, our goal should be to choose the path that will result in the least bad consequences in case we make a mistake.
In the beginning, there were projections that 2 million or more Americans would die from SARS2. Thus incurring $14 trillion in economic costs & losses worth $7 million per life saved would have been worth it–I guess. Now with 2020 hindsight, it appears those fears were overblown. Nonetheless, even though we arguably made a mistake, we probably made the right mistake: to do nothing & then incur millions of deaths would have been a disaster.
Going forward, we have two broad families of choices: open up or more lockdowns. Given that it appears we are over the hump (cases are trending down even though tests are up; and death rates from 2nd wave are trending down finally), erring on the side of caution this time means opening up more. If there is no 3rd wave, then opening up is the right move. However, if a 3rd wave does develop, opening up would be the right right mistake to make.
The lives saved by another nationwide lockdown (as Biden has said he is open to if the “science” tells him to) will not save millions of lives, even if a 3rd wave does happen to develop. And if no 3rd wave develops, trillions more in economic losses would be the wrong mistake at this time. Again, imho, ymmv..
Jon Murphy
Sep 5 2020 at 1:32pm
You don’t see how this is just two sides of the same coin? Minimizing costs is the same as profit maximization. If you are trying to minimize mistakes, you are necessarily trying to do the “right” recommendation.
The problem of experts you appropriately highlight in your opening paragraph does not disappear just because you re-phrase the question.
Jose Pablo
Sep 5 2020 at 3:00pm
Yes, “luck” (meaning the value of the “death rate” actually observed within the actual probability distribution of this variable) is, very likely, the main driver behind the impact of the pandemic at a local level.
Yes, it is impossible to know the real impact of government measures (I guess very limited)
Yes, government actions should focus (should had in March and should still now) on managing the risk of “fat tails”. Criticism or praise should focus on their success (or lack of) on this aspect.
Not so sure about the “economic implications of the measures” part. The “economic implications” of the government measures are related with their “marginal” impact on the economic interactions. Even in the absence of “mandatory” measures these interactions would be voluntarily reduced. And this voluntarily reduction will be related with the judgment of the people on how risky this interactions are. As you imply in your comment, this “marginal impact” is very difficult (¿impossible?) to observed.
Comments are closed.