What kind of crowd are we now seeing? The 5 surprises in this pandemic.

There are 5 aspects of the covid-19 pandemic I really did not see coming, all pointing to a phenomenon that European sociologists of a century ago spent their whole lives describing, coming up with theories about crowds and their behaviour – theories now largely forgotten. Scholars like Norbert Elias, Theodor Adorno, and Elias Canetti, witnesses and survivors of the World Wars, wrote about crowds and used phrases few people now recognise, like “hunting packs”. Those sociologists witnessed how whole populations in villages, cities, and countries changed in a matter of weeks from docile citizens into fascists, communists, warring tribes, and mourning packs.Cartoon angry mob stick characters walking Vector Image

It seems crowds are back, eliciting individual behaviour not seen on this scale for nearly a century. Let me talk about these crowds in the context of the 5 elements that appeared over time that surprised me.


                        Contagion of fear and of policies throughout the world

The emotional interconnectedness of the whole world shone out in this crisis, as evidenced by the quick and ubiquitous contagion of mass hysteria through social media and the popular media in February-March 2020. Even the Chinese censors were unable to prevent mass panic from breaking out over covid-19 on Weibo and other social networks, and there was no stopping the mass hysteria that soon followed in the West and eventually appeared everywhere in the world.

Newspapers were lapping up the fear; “experts” were fanning the flames; the WHO and lots of other organisations were quick in pushing the “something terrible is coming” messages; normally rational scientists with a keen sense of perspective were writing mass petitions that begged governments for totalitarian responses; elites in poor countries followed suit; etc. This is all evidence of contagion of emotions and beliefs, turning individuals into fearful crowds. It happened essentially in no more than a few weeks. That’s all it took.

I really did not see coming that the world was so emotionally interconnected. What I witnessed with surprise was how the contagion also affected many of my friends and family members, making them look very differently at the same objective risks and possibilities within a space of just a few days in mid-March. I still remember how one week my students were fearless rationalists and the next week they were anxious and meek. Smart, talented young people with nothing to fear from this virus. The contagious wave of fear, affirmed by policies, transformed them.


                How the West followed China and thus respects China.

One could see the implicit respect in the West for China shown by the West following the Chinese policies of mass lock downs.

One does not follow the hysteria of the beggar but tells him to calm down, yet one does take serious the hysteria and subsequent actions of the respected citizen. So too with countries. Europe initially poo pooed the policies of China, but adopted them very quickly when fear spread. India copied the UK and Africa followed Europe, following the logic that if the countries one looks up to do this, it must make sense, right? The surprise in this is how Europe followed China.

I had not yet realised how respected China now actually is as a citizen in the world community and in Europe. Interestingly, the place that most sees itself as the superior of China, the US, is the place that most strongly resisted following its example, though even there large parts of the population were on board with the same policies.


                     The widespread and sustained appeal of the hysteria.

This is the deepest puzzle for me and the most interesting observation. Populations became overwhelmingly supportive of totalitarian responses to the threat of the virus and strongly resented smaller groups or countries that tried visibly different policies (like Sweden). That desire to see others do the same as oneself is one of the hallmarks of crowd behaviour: the crowd wants to feel one and does not tolerate dissent. Through media and social networks, people were egging each other on to stick to rules and punish dissenters. Dissenting voices in the media and academia were censored.

There was also other classic crowd behaviour: status, careers, beauty, and any other form of social position were momentarily irrelevant. Everyone got a break from their ambitions and other burdens carried in “normal life”. They lost their individuality. Instead, everyone was a potential victim of the virus and everyone was focused on the threat, with every “other” a potential contaminant. Countries and regions locked themselves against others, whether those other regions were ”cleaner” or the opposite. A large new set of rituals and beliefs very quickly got taken up: washing hands, keeping distance, wearing masks, and wiping surfaces. Such things were known as totems, taboos, and sacred rituals by social scientists of a century ago.

This behaviour is very reminiscent of the stories of crowds by writers of the 19th and 20th century who witnessed the rise of nationalism, puritanism, fascism, and communism, deeply impressed by how millions of individuals became united in crowds. Quintessential elements are the extreme fluidity of values, beliefs, and the negation of individualism and the pursuit of individual success. In crowds, groups dynamics are highly accelerated as individuals blur into one and want to blur into one: crowds are groups operating in a high-intensity mode that unlocks behaviour not otherwise seen. The high-intensity mode comes with a clear goal everyone obsesses about, whereas normally groups have fuzzier goals like the wellbeing of its members.

Yet, the crowds of Europe as depicted by Elias Canetti, Theodor Adorno, and Gustav Le Bon were dense, with people packed close together physically, either hunting, feasting, or lamenting. The images one then thinks of are of marching armies, stadiums full of fanatical supporters, jamborees, rave parties, and crowded funerals full of weeping widows. One thinks of people touching many others, becoming one entity in joint movement and thinking.

The covid-19 crowds were of a very different nature, neither feasting, nor hunting, nor lamenting. They were not made up of people physically close together, but by people keeping their distance. Only on the internet and in their minds were people acting together against a common threat that was perceived to threaten each of them as individuals.

Is this a new type of crowd, the “anxious internet crowd”, or some more well-known form of crowd that is somewhat disguised? I will speculate about that later, but the key observation is the sustained popularity of the totalitarian response to covid-19 in Europe, which indicates something is going on that really appeals to people: they want it to go on.

In March 2020, I truly thought the hysteria and the resulting support for draconian measures would soon fade as people realised the risks were not that high and the damage of the reactions was immense. From the extensive economic and wellbeing literatures, I knew mental health and happiness would go down a lot and that unemployment would go up a lot, which is indeed what actually happened. Yet, mistakenly, I expected this unhappiness and loss of jobs to translate into protests and resistance.

When the low death numbers started rolling in in April and the IMF started pumping out dire predictions of the collapsing economies, I thought: “here we go, this will at least wake up the economists and the governments so the madness will now soon end”. Well, the economists in central banks did start to jump up and down, but that was about it in terms of waking up. My mistake was to see it all as evidence of mass hysteria, and not of the transformation of populations into crowds. Mass hysteria is dangerous but very transitory. Crowds are much scarier things.

I expected large groups would realise within weeks the enormity of their loss of liberty and the violation of their personal sphere and privacy. I thought parents would rebel against the damage done to their children. I thought adults would rebel against the inhuman locking up and social isolation of their elderly parents in nursing homes. I thought the groups that lost their jobs and dignity would be rebelling in the streets. I thought scientists would regain their critical views of the mass experiments their societies had embarked upon.

None of this occurred anywhere. The closest to the protests I expected to see were the rallies in the US by small right-wing groups, apparently only saved from the lure of the hysteria in the rest of the world by the extreme partisanship of their politics and their disdain for the rest of the world. I am not normally on the side of groups like the Tea Party or the Second Amendment die-hards, but this was one of those occasions where I thought their behaviour was normal and rational whilst that of the rest was not.

Yet, on deeper reflection what might be a better explanation for these “hold-outs” in the US is that they were already crowds, but then with a very different goal that united them. Perhaps their behaviour is not due to the US being resistant to crowds, but due to the US having succumbed to them a while back already. Covid-19 was too late to turn the US into a single crowd. The unexpected benefits of extreme partisanship!

Whatever the reason for the protests in the US though, those protesters were the exception in the West, particularly here in the UK where only somewhat aristocratic intellectuals were speaking out against the totalitarian policies. Among the masses of scientists and the population, the crowd-state endures. Even now, mid-June, parents are protesting in large numbers against resuming the education and social interaction of their children; the locked down unemployed protest against measures to end the lock down; office workers protest against the prospect of going back to offices and work places; etc.

So the UK population is basically protesting against the return of normality and the chance of a decent life for themselves and their kids. This can’t just be ongoing fear. It’s the kind of behaviour we normally in social science associate with crowds as it is within crowds that people stop caring about their actual health, wealth, and all the other things they care about in normal life. Some deep need they do not normally get to satisfy is being met that makes them willing to damage their children, their own health and their own future. This is distinctly “crowd-like” behaviour, even if it’s a weird crowd.

These crowds are getting some kind of kick out of having an immediate joint goal that they don’t want to let go of. And particularly in the UK, which to me suggests the relief of the usual “keeping up appearances” culture that so typifies the UK must be immense. Whatever it is though, I have been taken aback by the abiding popularity of totalitarian responses to this minor threat and continue to wonder about what type of crowd I am watching.


                  Virtue signalling and hyper-rationality

The crisis saw a hyper-activity of social scientists and governments in coming up with ways to be seen to protect the population from the virus. Schools got hundreds of pieces of conflicting advice on what to do by the ministries; the health sector was flooded with advice from medics and others; the newspapers saw one petition after another with particular advice from particular groups, etc. In my line of work, social science, researchers have stopped doing almost anything else and are churning out maps of covid infections, analyses of who is more infected, the optimal way to prevent longer-term job-loss whilst still outlawing normal job activities, and the economic effects of different types of lock downs.

I see 99% of that work and advice as making the problems worse, ie totally counter-productive and essentially validating the goal of the crowd, with no regard as to actual health outcomes: it has largely been about seeming to reduce risks of infections and deaths from this one particular disease, to the exclusion of all other health risks or other life concerns. Social scientists have slotted in seamlessly with the program of how to help the seeming, taking the overall strategy as unquestionable. Of course they mumble something generic in their conclusions, but to truly strongly oppose the new orthodoxy? Not done much, though, to be fair, increasingly so.

It is this seeming aspect that makes me see this activity as a spectacular form of virtue signalling. They are falling over themselves to be part of the crowd, which is the essence of virtue signalling. The lack of interest in overall health and wealth effects makes it virtue signalling: ticking a very particular box but not making tough tradeoff calls that would openly offend the others obsessed only with covid-19 infections and deaths. The lack of interest in a long-run solution is also telling (except for the somewhat magical idea of a vaccine that would arrive in months rather than the usual decades that they take. It smacks of the belief in a wonder-weapon that will unexpectedly win the war for a losing army).

Whilst I knew in March that a hyper-focus was a natural aspect of fear, I didn’t expect what can be called this calm “hyper-rational” approach taken to the supposed threat. It is almost as if the virus quickly morphed from something that scientists and governments feared into something that they loved to hate and be seen to fight against using all their talents. Such a goal-transformation is also a typical crowd-attribute: crowds like being a crowd and the goal that unites them is somewhat immaterial to a crowd, so it is very willing to switch goal functions at the drop of a hat. From fleeing to guarding. From hunting to total war. As long as the crowd doesn’t end.

In the language of Elias Canetti, one can describe what happened as the formation of a fleeing crowd (lock down) turning into a hunting pack (track-trace), turning into a warring crowd (eliminate, safe zones). And now that the crowds have lost their goal, they are dissipating in many countries and normal concerns return. One can see the massive take-up of the BLM movement in various other countries than the original and focal point of that movement as, partly, due to the crowds in those other countries very eager to latch onto another goal before the crowd falls apart.

I still don’t fully understand the psychology behind this, but the March-June 2020 period really does remind me most of the enthousiasm among scientists and government departments in the second world war when they were supporting the war effort of their countries. Germany and the UK come to mind most of all. I am saddened that covid-19 is the cause they ended up with this time round rather than something more sustainable and useful (like perhaps the local environment or heritage): they have picked a fight they cant really win and a religion that can only disappoint, enslave, and impoverish. Just like the ideologies of the 1930s.

Yet, the analogy with a war effort is really what it looks like. There is the same unquestioning presumption that the cause is right, that the fight will be won, that naysayers and non-combatants are basically traitors, and that there are technical solutions that will quickly overcome any apparent problem or collateral damage. There is also the same disregard and disinterest on the part of individuals in the enormity of the collateral damage, either to their own kids, people in other countries, their own futures, etc. There is even the same fatalism about the inevitability of the path they are on. These are individuals somehow enjoying not being individuals.

It is coming apart now, but last so in London, the home of theater.


                  The medical establishment defending its policy pre-eminence

There has been, particularly the last few weeks, a doubling down of the medical hierarchy and scientists involved in the advice to governments. I expected the scientific advisers to the European governments (the UK and the Netherlands in particular), but also the journal the Lancet, and all the other key public health advisers and organisations in the rest of the West to switch stories round about mid-April. I expected them to start peddling back hard and openly calling for the lock downs to end, schools to reopen, the necessity for people to touch each other, etc. Whilst there have indeed been many medical groups shouting about the damage being done in their neck of the woods (cancer doctors, education groups, mental health practitioners, organisations for abused women, etc.), the group feeding directly into policy has not changed its tune. The health generals are refusing to admit their war is lost and was futile to begin with.

Indeed, the exact opposite is happening right now: an editor to the Lancet has just published a scathing book on how governments weren’t enslaved enough to medical “advice” and thus got thousands killed unnecessarily. I am afraid there is a counter-argument to be made that the Lancet has been complicit in the deaths of millions by refusing to adopt realistic and overall views of the total effects of the advice they published continuously in their journal, so pressing the “thousands unnecessary deaths” line when the damage runs into the equivalent of millions is just weird. Methinks this editor protesteth too much.

Though governments throughout Europe are essentially forced by the collapsing economy and social systems around them to open up their economies, even if the population doesn’t want this, the medical advisers here in the UK are right now still doubling-down on the scare-mongering. Top journals in medicine still publish stories on how important a 2-meter distancing is, how likely a devastating second wave, the dangers of children who could infect their parents, against the lack of absolute safety in offices, against less than 2 weeks quarantines, against using temporary nurses, etc. They remain willfully blind to the huge health and social costs of this advice, simply hiding behind the unforgivable excuse that that is not their problem and they are just warning against the risk of covid-19.

So whilst the prophesies of doom have been well and truly debunked by the low death rates everywhere, no matter what policies were pursued, the medical hierarchy maintains a steady glut of fear stories. They also maintain that their advice has saved millions and that society needs to progress with what they call “extreme caution” and I call “extreme recklessness”. Only the Norwegian health authorities were brave enough to say they over-reacted, but even they did not admit the scale of their over-reaction.

                         In conclusion

To my best current reading, what we’ve seen in March-June 2020 in Europe is the emergence of fleeing crowds turning into hunting packs turning into warring crowds. These crowds are now either slowly dissipating as they have run out of a joint goal, or they are still looking for other goals to keep the crowd intact.

This is behaviour not seen in Europe on this scale for almost a century. Perhaps the reason crowds are back are because the levels of inequality are back to those of that time: an inequality that puts immense psychological pressure on the majority of the population, which is thereby eager to adopt the numbing effects of becoming a crowd, relieving themselves of the burden of expectations and subjugation normal in highly unequal societies. Perhaps that is why the US was more immune to the mass hysteria: its high levels of inequality created crowds there a while back, some of which vehemently opposed to other groups they were suddenly asked to join with.

The crowd behaviour so openly on display now is making me re-evaluate some other recent phenomena too. Political correctness and virtue signalling now in hindsight seem like embryonic crowd-formation and crowd-enforcement. Ditto for re-emerging ultra-nationalist groups. If its true that crowd-formation is essentially how populations react to high continuous stresses, such as from increased inequality, then very unequal places like Australia are going to see a lot more of this type of phenomenon. More equal places have less to fear, though of course the huge depression that is now upon is, is an acute crowd-formation event itself.

I feel how I imagine opponents of the Nazis felt in the Third Reich: dazed amazement at what has “begeistered” (animated) the majority of the population, coupled with the determination to hold on to the hope that that population will eventually come to its senses, saddened at the huge destruction it seems to take. Like those who loved German culture lamented the debasement of the peoples of Goethe and Schiller, so too do I feel the Weltschmerz of the damage done in this crowd moment to all I love. Snap out of it! Please!

This entry was posted in Coronavirus crisis, Cultural Critique, Economics and public policy, History, Life, Philosophy, Politics - international, Politics - national, Science, Social, Society, Theatre. Bookmark the permalink.

120 Responses to What kind of crowd are we now seeing? The 5 surprises in this pandemic.

  1. Nicholas Gruen says:

    Without commenting on your wider thesis, domestic terrorism was a major driver of the crowd behaviour in Nazi Germany with brownshirts running around beating people up and terrorising Jews.

    • paul frijters says:

      absolutely. Driver and consequence at the same time: the brownshirts formed a crowd acting like a crowd and gaining energy off its activities. The Jews were not its only opponents though: socialists and intellectuals who disagreed (the “enemy within”) were also targets who ended up in camps. The smaller crowd successfully marginalizing all other crowds till nearly the whole country was as one.

      But I deliberately also talk about communism, nationalism, puritanism, and many other ideologies and groups that these sociologists identified as crowds in the 1850-1950 period. Le Bon wrote at the end of the 19th century, more than a generation before the brown shirts of Germany. The nazis were simply the most spectacular example we have in European consciousness of crowd behaviour fanatically held on to till a very bitter end. But they were not the first, nor alone in that period. Nor is the crowd behaviour I think I am seeing this time of the same level or longevity as some of those of the 1930s. But if I am right that the level of inequality is a big factor in the emergence and popularity of the “crowd-mode” in large groups, then we will get more crowds in the coming years. Possibly much nastier ones.

      I may of course be entirely wrong, but in a way I am really engaging here with trying to understand what happened, what the realistic alternatives were, and also the psychology and point of view of those who have been in the mainstream on this topic. And I am using all of it to look ahead and to understand people and time-periods. I am thus not opposing, but trying to understand, combining grand historical stories with basic data and very direct personal observations. I also iterate back and forth with predictions and analyses of why those predictions were wrong. This is how I do social science.

  2. Juergen Bracht says:

    Hi Paul,
    thank you for writing up your thoughts. I like the pointer to N Elias’ and E Canetti’s work. I had read up on their works while spending time in the Holy Land during the Second Intifada. So, there I watched the shutdown of social life and warnings in public transport, demonstrations of support for the young persons in the front line, and many of the facettes you have reminded me about. Especially, hostiliy to dissenters — on both sides — and aversion to pointers to detriments of terrorism and counter-terrorism — on both sides. I remember the alerts in the morning radio about the certain expectations of murder in broad dayline.
    In the current crisis, I understand people’s aversion to be Guinea pigs; imagine you are Patient 0 in institutionalized quarantine without medication. Perhaps they want to see how the infection rips through your young body. Ghastly, be assured.

    • Hi Juergen,

      thanks. Yes, I too have read Elias and Canettis work a long time ago. I very much admired Elias (a big name in Amsterdam) and was intruiged by Canetti. I recognised his work as somewhat brilliant but also childish and slightly mad, so I stored in my mind as “he’s got a point and is seeing some things very clearly that are probably very important, but I dont yet know how to slot these insights into the picture I am building up”. I feel like I finally do understand what they were on about, having witnessed the phenomenon they were going on and on about. The element that is new for me is the importance of this “numbing element” in forming a crowd: the peace of mind that comes with becoming part of a crowd and can thus let individual social expectations go for a moment. It is that what rationalises the emergence of it now, the emergence of it 100 years ago, and even the distribution of its emergence this time round. It even rationalises who is most keen to be a crowd and remain a crowd, which includes the poor. And it rationalise who is least part of this: the aristocrats who didnt feel burdened by social pressure because they saw themselves on top of it all anyway.
      As I said in my reply to Nick though, I still want to allow for the possibility that this is not the best way of seeing it and that something else is going on. Understanding crowds is the deepest level of social science I know, and thus also the easiest topic to be totally wrong about.

      • Juergen Bracht says:

        Personally, I can confirm the effectiveness of „numbing element“.

        Even if you have dissented for a long time and your lunch partners have moved to another table a few times, at a point you might not “see” the combat helikoptors on a fighting mission over the beach anymore, and you might move away from the man with the big bag under his bus seat unawares. I suspect it is similar on the other side, but I have not tried that out yet.

  3. Aaron says:

    A very well written piece. I, too, find myself looking on in disbelief at the brainwashing and wondering when (or if) this will all end. The common answer I often hear is “only when we have a vaccine”.

    So I decided to talk to a few people. To understand what’s going on in their minds and why they support lockdowns. A few patterns emerged. I found a few clear groups:
    1. Elderly people (often with comorbidities) who are egging on the hysteria – this is driven by the human desire for self-preservation. Can’t blame them really.
    2. Relatives of these people, often living with them – again, they want to ensure their loved ones don’t die, so they want to continue lockdowns.
    3. Out here in California, people who can work from home and feel little to no impact so far – their jobs are secure, their expenses have gone down and they feel they are playing their part in helping society fight the virus.
    4. Similarly, unemployed people here in California who are getting paid a lot of money (both from state and federal governments) to stay home. Some are making more than they normally would while employed.

    The last two groups puzzle me most because they seem happy to give up their liberties for a virus that poses little threat to them.

    There is one common thread that unites almost all the Americans pushing for lockdowns – hatred of Donald Trump. I see a similar pattern in Brazil where hatred of Bolsonaro is a common thread of unity. It’s a bit like the Jews were turned into scapegoats with Nazism and that hatred used to unite the masses, although Trump and Bolsonaro haven’t exactly covered themselves in glory here either.

    • Hi Aaron,

      thanks. I encourage you to read some of those sociologists of a century ago. What they say will strike you as slightly mad, but also very insightful. Its their observations of how communities they observe closely transform that are probably the most persuasive and important.
      Your response still treats the individuals in the groups you see as rational individuals. The point of crowds is that individuals no longer behave that way but very differently: different rules apply. In the language of regular economics, their goal function, discount rate, risk aversion, frameworks, self-image, self-awareness, and self-regulation radically change. That’s why appeals to how they were before do not work: one is in a sense not talking to the same people anymore, though afterwards they will probably become similar to whom they were before. During the crowd phase, you should see them more as a kind of blend between their former selves (particularly in terms of their skills) and a highly focussed group entity (think of it as a kind of “beast”: the transformation into something else is a topic you will see a lot written on by these ancient crowd theorists). If it helps, think of it as a cult, which is the closest one gets but does not quite cover what is happening here because a cult is much smaller and invariably has a glorious leader. This crowd has had no leader, nor has it needed one. That’s also why people Trump resisted: he saw it as competition.

      Your observation on the hatred of Trump leading many Americans to buy into the European example is of course true. Perhaps this partially signifies that the US is divided into somewhat crowd-like entities already, defined by their opposition to the other crowds. Perhaps many Americans know no other group mode, but we Europeans and Australians certainly do (we nearly all see this as strange times).

      It’s a bloody difficult topic and I dont yet feel totally on top of it either, though feel the excitement of learning about it.

      • Aaron says:

        Makes sense, Paul. I will add these to my reading list. Plenty of time to do some reading these days :-)

        How do you see all this ending, by the way? Are we basically going to go in with this insanity until the much anticipated vaccine appears? No government (except Sweden) has embraced herd immunity yet. Lots of places are opening up despite an increase in infections, but they’re also copying China by re-imposing lockdowns and therefore thwarting herd immunity. Basically it feels like we’re getting into cyclical lockdowns.

        • In Europe it seems to be the case that populations are slowly snapping out of it now, though it will wake a while. Competition is weaving its magic by getting countries to open up quicker to get the summer tourists. All the neglected causes are starting to complain. The institutions are starting to function more normally again. The recession is becoming the big talking point. Etc. I cant see them returning to lock-downs. Where do you see that happening outside of these collectivist places? Do you mean New Zealand?

          The US is snapping out of it big time with competition between states a big driver. Cant see them locking down again either. Can you?

          The lack of a second wave in many places in Europe strongly suggests many places basically have achieved herd immunity or are close to it. Since those serology tests suggest at most 25% of the population has anti-bodies specific to this virus, that strengthens the theories that a high proportion of the population already had some immunity via different diseases. That would also rationalise a lot of the differences in death rates between regions and countries: those countries with low death rates (like Germany) might have had a bigger wave of something else in the past. However, some countries will just get a second wave and then we’ll see how big their previous wave of some related disease really was.

          The rest of the world will follow Europe and the US, I think. It’s these collectivist countries I am least sure about. They might go for some highly totalitarian surveillance option to suppress this virus and any other (which is in the long run a very unhealthy thing to do).

          We’ll see vaccine announcement soon, I think. Though I doubt they’ll be all that effective, governments can hide behind them so have a major stake in rolling out big vaccination programs.

          NZ and Australia have an interesting choice to make. Impoverishment and increasing health problems till the magic vaccine comes, or simply accepting a wave of cases and opening up fully, or even deliberately creating a wave of cases to get it over with quick.

          • Aaron says:

            Lockdown is being re-imposed in Chennai (India). Of course it never made any sense there, nor does it now. I’ve also been reading about some US cities like Austin and Miami reimposing lockdowns, but it’s unclear if the mayors have the authority to impose them. Also read about about 400 households in Berlin being locked down – I guess some sort of neighborhood lockdown there. Lots of talk of opening each state in “phases” based on number of cases and deaths and if they rise then to roll back to an earlier phase, including going back to lockdown.

            The countries that are attempting eradication will continue with that approach. Impossible for them to just let things go at this point. NZ and Australia for sure.

            As for a vaccine, that worries me frankly. It could set a very bad precedent that we should simply lock down the world the next time there’s a virus and wait for a vaccine without bothering to look at the tradeoffs.

            What do you think?

            • paul frijters says:

              there is a nice saying on the long-run: states will do the sensible thing, but only after exploring all other options. Quite. So, no, I dont think we’ll get endless cycles of scares followed by lock downs until a vaccine emerges. It’s just too costly and the countries that dont do it will have a huge advantage over those that do, so competition is our friend here.

              Also, crowds dont last and often dont latch on to the same goal as a previous crowd. Some groups last (like nation states), but crowds dont.

              Interesting about these city lock downs. I hadnt heard of them. The Germans might of course well get a second wave given how small their first one was: successful repression is not success at all with a virus because its temporary.

  4. Conrad says:

    I assume you agree with me now that the cheapest Covid solution has to take into account the way people behave based on their fears and perceptions. I also don’t see the response as unexpected as the way you do. When did people ever add up multiple distil costs and behave rationally? How about people acting on a rule of thumb — stay out of harms way. This is a difference between what many psychology people think (that many things we think of as rational are epiphenoma) and what many economists love to think, that if you add up all the costs and quantify them people will do the sensible thing, even when the behavior you get isn’t quantifiable in this way (such as that which you describe here).

    It is also clearly a multilevel problem even in terms of rational behavior since the cost and risk to the individual can vary widely — and you assume people will do what is best in the long run for everyone (why I don’t know). If you’re obese and over 50, clearly stringent measures are worthwhile based on a stay-out-of-harms way heuristic, and that’s many of our politicians and loud-mouths with money and influence (who may or may not be right at the everybody level). So it is no surprise you get splintering easily.

    • The iconoclast crisis of 700 ad to about 800 ad is one of the more extreme examples of opposing mobs coming very close to pulling their entire society apart and it was all literally about ‘pictures ‘ nothing more substantial

    • Conrad so agree about the ‘unrealism’ of expecting people to behave rationally.
      In case it’s of interest the following is from a art historian on the Byzantine iconoclast civil war .

      “ In 730 CE, a longstanding and irresolvable controversy suddenly escalated in Constantinople and throughout the Eastern Empire such that, following a series of decrees by the Emperor Leo III, a bitter civil war broke out and was fought over a single issue for more than 100 years—art. The radical party in the conflict, the iconoclasts (including Leo), regarded graven images as by definition heretical, as in Judaism and Islam. Their form of orthodoxy demanded that all such images be destroyed, and they set about doing so. The iconoclastic party tended to be drawn from the Eastern, poorer, non-Greek peoples of the Byzantine Empire who had sporadic contact with the rapidly developing Islamic caliphate—the prophet Muhammad having died only as recently as 632 CE. Therefore, a purely political consideration in Constantinople was the urgent need to shore up the outermost eastern provinces lest whole swathes of the Byzantine Empire be lost to the caliphate. Art, all of it, would necessarily be permanent collateral damage. For a long time, the iconoclasts were in the ascendant, and their political objective was largely attained. Meanwhile, they lost no time and spared no energy and treasure in the systematic destruction of art, illuminated manuscripts, mosaics and so on, everything, which explains why we retain so little that predates 730.

      The opposing party, the iconodules, tended to be ethnically Greek, more prosperous, more literate, and concentrated into metropolitan Constantinople and the Balkan and South Italian provinces to the west. They maintained that images of the incarnate deity, the word made flesh, were not only a necessary component of Orthodox liturgy and church architecture—a visible microcosm of the earthly and heavenly realms within the decorative programme of a single church—but that images, icons, were more than that: They were an essential demonstration of the doctrine of the incarnation, and a kind of proof. A lot of that thinking was developed in some haste as a sort of rearguard action while the civil war raged. Following many setbacks and defeats, in 843 the iconodules finally emerged triumphant.

      I can think of few such disturbances, before or since, and lasting so long, in which art, both imagery and architecture, was not only a central but, rather, *the* single issue of dispute that precipitated violent civil strife and sweeping cultural and political change for more than a century. The period was coloured by the intricate machinations of bureaucrats in successive Byzantine courts; the sophisticated and distended theological speculation of bishops on either side, sitting for months and months in fractious, raucous council at the edge of the Bosphorus; the raids made by waves of imperial troops into town, villages and monasteries in order to scrape fresco and mosaic off the walls and ceilings of churches, and the riotous ejection of troublesome deposed Patriarchs into distant, inclement places of exile. The Puritan movement in sixteenth- and seventeenth-century England was a picnic by comparison, and of course spread over a far wider range of fundamental doctrinal and socio-political disagreements.

      My thesis, to the extent that I can remember it, was that whereas much emphasis had been placed on the all-or-nothing character of the conflict and its resolution, in the years following 843 there was instead a sort of extended transition, in which old approaches to important subjects, above all the crucifixion, jostled for supremacy with new ones, sometimes even in a single codex or work of art. The Chludov Psalter is an excellent example (ca. second half of the 10th century CE; State Historical Museum, Moscow: MS codex add. gr. D129).”

    • paul frijters says:


      “I assume you agree with me now that the cheapest Covid solution has to take into account the way people behave based on their fears and perceptions. ”

      you miaa the key point of the post: its not really about fear or perception, but about the formation of a crowd (which never lasts). It seems to me you are one the crowd members eager for the crowd to go on. You are one my data points.

      One does not control crowds from within and hence take “into account…fears and perception”. The crowd controls those within and they like it. The fear is now just the excuse for the crowd to exist.

      How crowds end is an important question though. I am reading up on it.

      • Conrad says:

        Glad to be a data point :), although probably not such a good one. I think the Korean style solution where you can do most things would have been fine months ago, especially after we found out that Covid was not SARs in terms of outcomes. If things are not too bad in Seoul (HK is similar), I think we can reasonably suspect things would be even better here with such a low population density. As far as I can tell, and you have reinforced my opinion with your observation from Euroland about China, the modal opinion is far more authoritarian. Also, if it wasn’t the case that I thought a vaccine was inevitable (I’m not sure why you don’t), I’d certainly have to reconsider my position, but rushing to herd immunity is going to look like a pretty bad move in hindsight if a vaccine comes along sooner rather than later (and people will forget that it was in hindsight and it was really a probability game with different options).

        As for groups, they can obviously last more or less indefinitely. My favourite example is the CCP. As I’m sure you know, they killed 40-60 million people, and they won’t even admit that was an especially bad thing. Indeed, they didn’t even change their name and they’re still here.

        This is also the problem with scientific debates — if they get politicized, then the the ideas will attach to political parties (I find economics especially bad for this), and thus they will last more or less indefinitely (think Laffer curve) — at least far longer than Covid will run if you assume a vaccine will be found in a few years or we will get to herd immunity. I really like Sweden in this respect. At least from casual observation of their evaluation of what they did or could have done better with Covid, they can just admit they were wrong or could have done things better, and, unlike here, no-one chops off their head and tries to take political advantage of it. So they really can change opinions if they want.

        • I have written extensively about groups and religions in the past, but this is a different phenomenon, though not entirely unrelated.

          As I said to Graham, I am making a distinction here between groups and crowds (though some of those sociologists I mention do not explicitly make such a clear distinction). Groups can be very long-lived things with joint interests, as you say, lasting generations. Like Australia. Or families. Crowds have this single-goal aspect, as in parties, hunts, and funerals. They can last hours or years, but not generations. I see the phenomenon more as groups running at an unsustainable level of intensity, which is both highly enjoyable but also potentially very destructive. So crowds are more like a state which groups can be in.

          You indeed are a data point to me now of a crowd member. A very informative one. Thank you.

        • Conrad
          Re authoritarian mode
          Postal voted yesterday for the Eden Monaro byelection. There are 14 candidates so I looked up some re who gets third fourth preferences.
          The Science Party lists this as one of its ambitions:
          “ Increase health research to end aging”
          Feel that many these days think that authorities can do anything, that they want.

        • Kien Choong says:

          It will take time for the CCP to come to publicly terms with the Great Famine; but I don’t think they deny it.

          Whereas the Australian PM recently denied that slavery ever existed in Australia. I suspect most Australians were unaware of that part of our history.

  5. Hasbeen says:

    The black lives matter riots & movement is showing a very similar trajectory

  6. Graham Young says:

    Thanks Paul. Interesting post. These are the factors that are generally left out of economic analysis, and why economists are so often clueless. What this analysis lacks is analysis of leadership. Some countries have behaved differently. Some leaders have behaved differently. Groups are a given, but why some groups do better than others is the more interesting question.

    • Hi Graham,

      I think in many places, leaders had no real choice given the institutions they had (though they should now think ahead and think of the institutions they need to prevent a recurrence). In the UK, the pressure was just so enormous mid-March that the government had no choice. In some other countries, like the Netherlands, I think leaders could have made a big difference. The Netherlands could have, I now think, been even more laisser-faire than Sweden (and it was initially), but it would have needed a more energetic prime-minister to quickly recognise what was going on and mobilise a media and scientific response against the fear stories. Instead, the PM totally gave up control of the narrative to institutional health authorities, which is about the worst bunch to give control to. Australia might also have gone differently, I think, but only if it had very quickly organised against the fear: one actively needs to push back by mobilising and amplifying the voices of those scientists capable of generating push-back stories.
      Those are the three countries I know best. I suspect there was also a real choice in New Zealand, Denmark, Norway, Germany, France, Finland, and some other European countries, but they had the misfortune in having the wrong leadership teams at that time.

      I am making a distinction here between groups and crowds. Groups can be very long-lived things, lasting generations. Like Australia. Or families. Crowds have this single-goal aspect, as in parties, hunts, and funerals. They can last hours or years, but not generations. I see the phenomenon more as groups running at an unsustainable level of intensity, which is both highly enjoyable but also potentially very destructive.

  7. Kien Choong says:

    Hi, I may be making an obvious point (apologies), but we tend to look to how other people behave when faced with uncertainty. That may be one reason why financial markets fluctuate so much in times of crises.

    Covid-19, the “novel” coronavirus, was truly novel. We’ve not previously experienced a disease that is so infections even before symptoms develop. We have little experience on what types of “social distancing” have most “bang for buck” in reducing infection rates.

    Perhaps if a future epidemic that is similarly infections arises, our experience with Covid-19 will make us less susceptible to “crowd effects” in how we respond. That said, every crisis will be different and unexpected in some fashion.

    • paul frijters says:

      “we tend to look to how other people behave when faced with uncertainty”

      yes, i agree with that, but I think in hindsight it is that which made it difficult to discern the crowd formation. For a long time I though it was “just” mass hysteria, blind copying of what others do. Whilst that was involved, it doesnt fit many other aspects. The continuing popularity. The disinterest in overall effects. The antipathy towards other countries who dont go along with the program. The hyper-rationality. The virtue signalling. The goal transformation. The behaviour of the health generals. Etc.

      And yes, I agree with you that many countries wont repeat the exercise. And who knows, maybe some countries will learn from this crowd moment how seductive and destructive it can be. The glass may not be half-full, but its never entirely empty :-)

      • Kien Choong says:

        Ah, I see the point you are making. You’re right, there is more to it than just crowd behaviour in face of uncertainty.

        Self-confirmation bias may explain its persistence and continuing popularity. Antipathy towards others seem to (unfortunately) reflect our “Us vs Them” tendencies. It doesn’t take much to divide ourselves into rival groups, even over arbitrary markers like “guards vs prisoners” (in replications of the famous Milgram experiment), or “cowboys vs Indians” (in teenage summer camps).

        On the “health generals”, you may be right that they have too much influence. (I feel the same way about the influence Australia’s intelligence/defence community has over our foreign policy.) But I think it is right that health professionals make their views known. They are the experts on public health matters and we need to know what they think. It’s up to the national cabinet to weigh all considerations vs letting public health considerations dominate.

  8. Paul it’s ages since I read Canetti, when I was about twenty he was nearly compulsory reading in my circle. The one I still really remember is “Ear Witness: Fifty Characters” and particularily “The Tablecloth Lunatic,” have you read it?

    The dispute re how to approach Covid19 ( like the long running dispute re ;all wild fire is bad , suppress all fires vs some fires are good) reminds me of Aesop’s version of the fox and the hedgehog, its not really about what works best, it’s more about personality types and identity and therefore much more intractable than disputes about what works.

    • I only read “crowds and power” by Canetti. Both brilliant and barking. It stayed with me as something with a point that I couldnt really place.
      But very interesting observations. He dug up a lot on the Aborigines. Mainly based on, I recall, a book by that German anthropologist Schneider who traveled all over the place in the mid 19th century and described a lot of the different cultures.

      • I read most of his books, agree about the strangeness of his vision . Agree about a point that’s hard to place.

        However after reading Nicholas Rothwells most recent books I feel a bit less at a loss to place it.
        The connections to Central European anthropologists and science is something that to a degree was , literally , blown to pieces between 1939 and 1945.

  9. KT2 says:

    Another data point;

    “Pollen Explains Flu-Like and COVID-19 Seasonality

    Martijn J Hoogeveen, Eric CM Van Gorp,Ellen K Hoogeveen

    doi: https://doi.org/10.1101/2020.06.05.20123133

    This article is a preprint and has not been peer-reviewed

    …” We conclude that pollen is a predictor for the inverse seasonality of flu-like epidemics including COVID-19, and solar radiation is a co-inhibitor. The observed seasonality of COVID-19 during Spring, suggests that COVID-19 may revive in The Netherlands after week 33, the start being preceded by the relative absence of pollen, and follows standard pollen-flu seasonality patterns.”


  10. Troubling aspect of , elimination, is there really is no exit strategy. If in say two years time we are still waiting for a cure, the preasure to , hang in, don’t write off all we have sacrificed, just one more push and perfection will be ours, will be enormous.

    And when elimination was adopted as a goal it was too early for it to be based on any evidence, it’s simply a preference of some regardless .

    • paul frijters says:

      I find it very interesting that you have not succumbed to this crowd-pull, John. Your reaction reminds me very much of the play by Ionesco called the Rhinoceros, which you might know.

      Its about a French village where all the inhabitants, one by one, turn into a Rhinoceros, which is clearly meant to describe this transformation of a normal citizen into a communist, fascist, puritan, vampire hunter, whatever. The whole point of the play is that all the supposedly upstanding smart citizens are the first to succumb and turn into a Rhinoceros. The one who turns last is someone who basically goes through life with no particular interest in status games or intellectual positions. Just someone who takes every day as it comes. You remind me a lot of the central character in the play. From the wikipedia page describing the play:

      “Over the course of three acts, the inhabitants of a small, provincial French town turn into rhinoceroses; ultimately the only human who does not succumb to this mass metamorphosis is the central character, Bérenger, a flustered everyman figure who is initially criticized in the play for his drinking, tardiness, and slovenly lifestyle and then, later, for his increasing paranoia and obsession with the rhinoceroses. The play is often read as a response and criticism to the sudden upsurge of Fascism and Nazism during the events preceding World War II, and explores the themes of conformity, culture, fascism, responsibility, logic, mass movements, mob mentality, philosophy and morality. ”


      • There was a rather good stage production of Rhinoceros with Zero Mostel and gene wilder that was filmed have you seen it?

        At best identity is a temporary vehicle for the great spirit ,at worst it’s mere strut and fret, signifying stuff all.

        Our revels now are ended. These our actors,
        As I foretold you, were all spirits and
        Are melted into air, into thin air:
        And, like the baseless fabric of this vision,
        The cloud-capp’d towers, the gorgeous palaces,
        The solemn temples, the great globe itself,
        Ye all which it inherit, shall dissolve
        And, like this insubstantial pageant faded,
        Leave not a rack behind. We are such stuff
        As dreams are made on, and our little life
        Is rounded with a sleep

  11. R. N. England says:

    Love of parents is one of the pillars of Chinese culture. Visiting them is the main reason for their high-speed rail network. East Asians travel for love; Westerners for entertainment. The high death toll of Covid-19 amongst the old is the reason for China’s unquestioning determination to fight the disease with greater determination than the West, and why ordinary people and their government wholeheartedly cooperate in a science-led, organised fight. In the West, as exemplified by this thread, the plight of the old is met with callous indifference because people have become the loveless tools of property (money) interests. Australia and New Zealand seem a little more like China than the rest. I suspect Nicholas is in more sympathy than most with the Chinese position because of his family background.

    • Aaron says:

      Love of parents should never be the basis of any public policy. Governments need an objective framework and should look at the cost vs. benefit of lockdowns in terms of years of life lost. China either miscalculated or failed to perform such a calculation. You might be confusing their determination with the totalitarian regime there which allows them to literally lock up their entire country by welding their doors shut. Other countries panicked and followed suit. We now have mass hysteria.

      “Science-led fight” is a rather meaningless statement because the science is not absolute here. There are differing opinions amongst scientists about this.

      • Conrad says:

        We don’t have mass hysteria. Most of the countries that had lock downs have opened up (Germany, Italy, Spain, Austria..), and people are presumably far more likely to go about their normal day-to-day existence than if the cases were still high and they were worried about catching it.

        The government also takes people’s feelings into account all the time incidentally. For example, one of Paul’s examples is IVF. The only reason you offer subsidized IVF to people is to make them happy or achieve some sort of abstract meaning to their lives. Its vastly expensive and it’s not like we are decreasing in population.

        Similarly, we spend vast amounts on late-life care for old people with intractable diseases, which often buys them a very small increase in life expectancy (of often a variable miserable end to life). We could achieve many more life years saved (let alone good vs. miserable life years) if we spent that money on the health of other groups. But we don’t. So clearly we don’t just add up life years for many things, even when, unlike coronavirus, the trade-off is pretty obvious.

        • Aaron says:

          We certainly do have mass hysteria because lots of countries are in various stages of lockdown despite very low death rates, the media is publishing doomsday headlines, people are still pushing lockdowns, scientists are continuing to defend their position on lockdowns, etc.

          The government can take feelings into account, but acting on emotion and making a decision to disproportionately benefit one group to the detriment of others is not their job. They’re supposed to act in the best interest of the population as a whole.

          When else have we made such a massively bad tradeoff as we have with lockdowns? 50 to 70x more years of life lost and not to mention putting the entire population into misery. This is rather unprecedented. And even if we do make bad tradeoffs as you claim we do, it doesn’t mean it’s a good thing. We should probably not be doing that.

          • Conrad says:

            Aaron, expensive end of life health-care is a clear cut example which we and most Western countries have been subsidizing for decades. So clearly the decision making process of the government includes factors not to do with adding and subtracting lives. You might not happen to like that, and you might not happen to think it is something governments should be in the business of, but they are.

            So even if people believed your and Paul’s rendition of costs (many don’t, which complicates things further compared to the previous example), the way people evaluate things is clearly multifactorial. In this case, it is a clear-cut example where people are willing to sacrifice things for a minority of the population (or haven’t bothered thinking about it), otherwise governments would be voted out for it given the size of the transfer.

            This reminds me of Douglas Adams and Vogons feeding their grandmother to the Ravenous Bugblatter Beast of Traal. Humans must like their grandmothers more than Vogons, otherwise they wouldn’t be willing to pay for them.

            • Aaron says:

              I think you’re missing the point. I’m talking about a decision of a very different magnitude that affects the entire population massively. There isn’t really any precedent for it. In either case, just because governments make certain wrong decisions doesn’t mean they should continue doing so.

              I think what’s happening in the current instance is a disregard of tradeoffs. There’s panic and hysteria. People and governments aren’t acting rationally. That’s Paul’s point.

            • Conrad says:

              I used the medical example because it is very similar — Trying to save largely older people from dieing at some expense. So it is not so different.

              The main area it differs, just in terms of lives, is that it is ongoing and has been for decades, so if anything the decision is probably more important than the coronavirus, which will pass one way or another no matter what.

              I also don’t see why the decision is wrong. You’ve just assumed that (thus going against most of the worlds democratically elected governments).

              • Aaron says:

                What’s an example of decision to save old people’s lives that is so expensive and costs so many more years of lives than it saves? Governments make such decisions all the time. No society saves lives at all cost.

                I’m basing my argument on Paul’s analysis which I find quite compelling. In this instance it appears several governments panicked based on a flawed model and didn’t look at the tradeoffs. If they have, why don’t they publish those anywhere? Other governments quickly copied them and here we are. There’s a strange hysteria that’s taken hold now.

              • Conrad says:

                Aaron, there any number of decisions as a whole (dementia care, cancer…), many of which are expensive — it’s a very well understood phenomena (basically, being old leads to vastly more health problems) and important due to the aging of the population which means it won’t be affordable everywhere. I did a random search and here is a paper from the US from some time ago US and places like Aus are likely to be even more skewed given our friendly health-care system. So the overall cost is enormous.

                • Conrad
                  None of the ‘excess’ treatments you reference have ever resulted in a tripling of unemployment ,the cancelation of almost all travel and the closure of all schools.
                  And none of these treatments to prolong life are compulsory either.

                  • Conrad says:

                    Sure, the underlying trade-offs are somewhat different. The cumulative cost will be far more for old-age health care, it will never be over, and it doesn’t directly cause a blip in unemployment (but it probably causes unemployment indirecty given the cost). Obviously they don’t close travel or schools for it. Some treatments are also effectively compulsary (e.g., drugs in old-age care facilities for people with dementia).

                    My point is that simple trade-offs in hypothetical life years lost are only part of the decision and thus decisions that use multiple measures are not necessarily “wrong”. They’re things based on what the population happens to want, presumably at least in part based on their morals. Ignoring this ignores political reality and hence what you do. Without taking this into account, you can propose “the best” solution all you want, and no-one will accept it.

                    Incidentally, I personally believe the current redistribution is a bad trade-off (it just costs too much with an aging population) and the government is just quiet about it because changing anything would offend too many people.

  12. murph the surf says:

    The comments which depict the institutional response as hysteria based on copying the Chinese government response are displaying a lack of knowledge of what could be considered the stock standard reaction to the outbreak of a new disease.
    These plans are all prepared in advance , they are also as far as I am aware accepted by governments around the globe.
    Stage one is actually more severe than lockdown , it is standstill.
    It is just that , no one moves.
    The authorised people then assess the situation and start to move to obtaining all the necessary information for tracebacks.
    Legislative powers exist to allow these actions and the disciplined forces assist as necessary.
    Once sufficient information is available rings of containment are announced.
    These delineate who can do what and where they can do what.
    At this point some readers will be wondering how such a system can operate and the explanation is based on these plans being more in response to exotic ( in this sense meaning not endemic locally) animal diseases that invade a country than human pandemics.
    Still the model is not unique to any one area of the world.
    WHO documents a multilayered response so communication keeps the population informed and especially combats misinformation spreading.
    The difficulty with Covid 19 was the poor institutional response in Wuhan when a cluster was detected.It is a bold statement to say that such an outcome would not occur in many other areas of the world.

    • Aaron says:

      The stock standard to an outbreak is not lockdown. We’ve never done this before for a virus of this type (extremely low fatality rate, very clear population group at risk, etc) Standstill is even less realistic. I’m not entirely sure the response in Wuhan was all that bad because it was a new virus and it spreads silently through asymptomatic people. Very hard to eradicate it for that reason.

      • Conrad says:

        “extremely low fatality rate”

        It seems likely to kill more people than most pandemics in history, so comparatively it isn’t low at all.

        • Nicholas Gruen says:

          And that’s after sacrificing $5 trillion and counting to get the death toll down.

          The death toll in the counterfactual being argued would presumably be in the millions. That still leaves it open to Paul, Aaron and John to argue what they’re arguing (and hoping the emerging chronic health issues from survivors aren’t too bad).

          • Conrad says:

            I imagine it will unfortunately be in the millions even in the best case (e.g., a vaccine arrives in a year) once the numbers from Africa, India, and Indonesia, start going up.

            In a scenario where people did nothing and a vaccine didn’t arrive, then you get to 70 million from just the current population assuming a death rate of 1%.

            • Aaron says:

              A death rate of 1% is simply not credible any more. Paul has written about this fairly extensively. Even NYC, where the virus was basically allowed to roam around freely ended up around 0.15%. That’s about as bad as it gets.

              The numbers in Africa and India are actually rather low, presumably because of the younger population. They’re doing far better than Europe was at this stage of the pandemic.

          • Nicholas in this example
            the counter factual is the null hypothesis

          • Aaron says:

            Nicholas – you mean sacrificing $5 trillion and simply delaying most of the deaths?

        • Aaron says:

          We’re talking an infection fatality rate of 0.2 to 0.3% and heavily skewed towards the elderly and people with existing conditions. The stock standard response should be to protect those people, quarantine the sick and get on with it.

          • Conrad says:

            Aaron, you have constantly given blatantly biased figures, some of which have been bordering on the ridiculous. Here are some estimates from the Lancet which are nicely stratified by age so you can update either your crapometer, your poor ability to understand numbers, or both: https://www.thelancet.com/action/showPdf?pii=S1473-3099%2820%2930243-7

            • Aaron says:

              Which of figures are ridiculous, I’m curious? Here’s one meta-analysis of IFRs:


              Also a rather long list of serological studies: https://docs.google.com/spreadsheets/d/1zC3kW1sMu0sjnT_vP1sh4zL0tF6fIHbA6fcG5RQdqSc/edit#gid=0

              I’d ask that we steer clear of personal attacks. I’m not interested in engaging in that manner.

              • Conrad says:

                Aaron you deliberately always give lowest bound figures and you argued black and blue that we could all catch the virus in a year (all 20 million of us in Australia lone). Since that was completely unbelievable, then you said it was a bad definition of herd immunity. You now dig up papers with no expert commentary and no other expert reviewing.

                • Aaron says:

                  I’m not doing that. The average was 0.36% wasn’t it from many studies?

                  As for herd immunity, getting to 60% or whatever infection rate isn’t the goal. It’s to get to a high enough number to make it manageable. We’ve been over this before. The growth is exponential, not linear. Taking 20 million or whatever and dividing by 12 months is not useful because the virus doesn’t grow linearly.

                  Very little is peer reviewed with covid, so that doesn’t strike me as a good standard to set. Unless of course you’re talking about papers published in Nature arguing lockdowns saved lives through circular reasoning. :-)

            • Conrad
              We simply dont know how many have had the virus and had no symptoms . Therefore all of the estimates re fatality rates are at best a work in progress.

              For example Japan has done virtually no testing its possible that very few have been infected or that large numbers have been infected. In the UK their attempts at large scale testing have been only marginally more successful than their attempt at a ‘DIY’ phone app.

              I expect that time goes on the % fatality rate will, globally shrink a lot , the Indian populations average age is I think less than 30…

              • Aaron says:

                Exactly. India seems to be doing way better than most of the developed world.

                • I suspect there is a lot of bad sh*t happening in India now that we wont get to hear about until quite a bit later. The journos who would normally tell us what is going on are nearly all locked up in the cities. God knows what is happening in those slums and villages. If they truly locked up that population and forced them to rely on charities and savings, you’d think millions must have died from starvation by now. Many of the poorest people wont be registered citizens so its not clear their deaths would show up in the official statistics. I fear a human catastrophe is underway in India right now that we will only get to know about much later. I hope I am wrong.

  13. Conrad
    If the sort of spikes and retightening of the screws we are seeing in Victoria keep happening every few months or even weeks, the public mood re trade offs around bread on the table and paying the mortgage Vs extending life from 80 to 83 ,could swing dramatically.

    Vaccine would be wonderful, but we need a plan b, we need an exit strategy.

    • Conrad says:

      I personally think the tightening is too much, but it highlights my point that the current modal mood is for harsher restrictions and not weaker ones, so you have a long time to go to get people to change their minds. To me it is senseless — people are worrying about 20 cases a day of which many were just people returning back to Aus. We also now know low case levels are controllable, and we have all the stuff to control it. As someone with two direct relatives who had heart operations, I know my most likely cause of death/sickness and the probability must be orders of magnitude higher. My risk of bowel cancer etc. is also vastly more.

      To me the preferable plan B is people just get cleaner like the Japanese, where sick people wear masks, preferably don’t go to work (unlike Japan), and wear masks on public transport. In a place like Aus which is largely hot and dry and where there are not massively dense cities, I don’t see why this wouldn’t just lead to low case levels permanently, which as far as I can tell we will have no-matter what unless plane travel has permanent restrictions.

      • Suspect that our governing class spends too much time listening to the mood on twitter i.e. looking in the mirror.
        Something for the Costs side of the equation.

        In the UK a lot of deaths Covid19 deaths are :

        Evidence has shown that dementia is the most common pre-existing condition found among deaths involving COVID-19. There has been a huge surge in the number of deaths of people with dementia who do not have the virus.

        Shielding and isolation initiatives were introduced to save lives, but there is a real worry that for people living with dementia, in care homes and in their community, such initiatives have contributed to an increase in deterioration with potentially devastating impacts.

        Lockdown easements and planning appear to be underway. However, the Government is yet to give guidance on what happens next for people affected by dementia who rely on social care.

        Given the devastating impact lockdown and coronavirus is having on people affected by dementia, plans to support them must be prioritised.

        • paul frijters says:

          that is just awful isn’t it. Those demented locked-away pining for human contact, denied “for their own safety” to the point that they just died. There is a banality about it.

          • Conrad says:

            It’s funny you guys should bring dementia up. This is because, as far as I can tell, if you were to calculate the cost of keeping people with later stage dementia alive in terms of cost per year of life added, and compared that to spending the money on many other things that would add years to people’s lives (e.g., bowel cancer screening, adding a new drug to the PBS etc.), the conclusion that you would have to come to is that we shouldn’t be spending the money on them. For example, in terms of the expense, we spend $4.6 billion in just hospital costs alone to help these people. So the fact we do it isn’t because it is the best trade-off in terms of life years between them and other groups (let alone quality life years). We do it for reasons like we happen to think they deserve it and that everyone should get reasonable health care. So we don’t optimize on life years but we do on level of service offered.

  14. Aaron says:

    An excellent article I read today about all the fear mongering going on in the media. It’s totally mind blowing how everyone is addicted to fear: https://outkick.com/media-ignores-90-coronavirus-death-collapse-in-country/

    • Well fear does get the clicks.
      Australia has just recorded its first Covid19 death in a month.
      The BLM protests did not result in a spike .
      The number in hospital continues to fall.
      In Victoria the spike in new cases seems to be down to all the official information only being available in English !

      Yet if i want to visit a elderly family member in Qld I’d first have to do 14 days in a hotel.

    • yes, the media almost everywhere jumped on the fear story early, hard, and won’t let go. It’s not a conspiracy either: they are not doing big business a favour and lock downs have been associated with strong drops in newspaper circulation. Their fear-mongering is a clue as to what is going on.

      • Conrad says:

        They fear-monger about everything in the public mind, so they are simply reflecting what their audience want to hear (I’m sure that must tell us something about people in general!). For online news, which is most news these days, presumably the click rate is much higher than slow news days. People love it — why else have hourly updates for things that go for months upon months. So it is demand driven as far as I can see. I suspect that for things like coronavirus, where any sensational aspect presumably wears off quickly, people must have some sort of internal mental model where aspects of it are prioritized in their thinking.

        I often use thisas a fun graph to show people don’t estimate health risks well but there are many others that show more less the same thing. This is the type of thing people are up against when trying to convince others of logical outcomes where you add up numbers to show some outcome.

  15. R. N. England says:

    People here sound as if the can’t, or couldn’t, wait to get their hands on their inheritance.

  16. R. N. England says:

    Chinese ancestor worship is a quaint reminder that gratitude to one’s parents does not necessarily cease with their death.
    In the haggling market place sacred to neoliberal economics, gratitude to the worker is the first moral casualty.

  17. KT2 says:

    Geez, R N, you copped it fear anger and emotion over a few words. Imagine what youd get if there was a pandemic! R N. Your statement was certainly a trigger. And many here wondering why fear and emotion are prevelant.

    John R Walker you said; 
    “Yet if i want to visit a elderly family member in Qld I’d first have to do 14 days in a hotel.” and a lot worse, blaming R N England for your psyche.

    My neighbor has done what you haven’t – left kids, work and family to do it. NSW to Vic. Don’t know where you are. She has done it savour the last welby’s of her father. And keep him out of a nursing home. Irrationally (ymmv), dumping a years worth of capital accumulation not including mortgage..

    If you are prevented from doing so, why please? 

    • Christ teaches me : do not be like the two-faced snakes that loudly twitter their virtue on the main-street. Instead when you do right thing do it quietly; just get on with it.
      So my lips are sealed.

  18. Nicholas Gruen says:

    Can someone explain to me how these case fatality rate estimates can be at 0.2-3% when the actual number of people having died in the hotspots of Northern Italy is about 0.6 odd percent of the total population with a similar – I think higher number in NYC.

    If you accept this challenge, please understand that I’m not trying to become a minor specialist in this area. I’m just trying to do some ground-truthing of what I’m hearing around the traps here.

    • Aaron says:

      NYC is 0.2% of the total population as of now. The IFR is much higher there than the rest of the US.

      Specifically which part of Northern Italy is 0.6%?

      NYC and Italy are likely done, they’ve got close to herd immunity as evidenced by the sharp drop. Antibodies are not the only form of immunity, that’s why serological studies don’t capture everything.

    • Nicholas
      In the long run because India represents : so much of the worlds total population, is crowded , lockdowns are simply not possible for any length of time and the vast majority of Indians are under 40 i.e. at low risk of serious consequences if infected , the global figure for Covid19s fatality rate will almost certainly trend down.

      • Conrad says:

        The ground truth isn’t that. That is the lowest bound estimate, and I suspect as with many lowest and highest bound estimates, may well be pushed for political reasons (c.f., “we must lock-down everything” vs. “we must open everything”). So you just end up seeing more of the extreme predictions. This is especially so because there hasn’t been much time for things to settle down and many of the tests are crappy (the drive-through test you can get even here has a false-negative rate of 30%…) so I imagine data sources need very high scrutiny.

        John — I’m less convinced about India. The figures are already not looking great despite them having only about 1/3 of the old people per head of population compared to places like Australia and anecdotal reports that the people simply may have more resistance to it — the latter of these reminds me of the early German data and had good their hospitals were.

        Part of the problem is they have a poorly functioning hospital system, so people will die who wouldn’t in other places (in this respect, China is not terrible, so the data would not be comparable), especially once it gets overloaded (which it appears to be becoming now). Not that I’d trust much data coming out of India — I doubt the government could collect accurate data — but I imagine it is going to be terrible because of this. I hope I’m wrong.

    • Hi Nick,

      let me try. As you know, I immediately estimated the fatality rate at 0.2% in mid-March and never had to deviate from that initial estimate, copping plenty of flack, btw.

      A nice fact-checking analysis of what is now being claimed is here: https://eu.usatoday.com/story/news/factcheck/2020/06/05/fact-check-cdc-estimates-covid-19-death-rate-0-26/5269331002/
      which to my mind says that the US Centres for Disease Control and Prevention is indeed putting the overall fatality rate just above 0.2%. Given how no country in the world is above 0.1% the CDC estimates will probably only decline.

      Of course at a very local level, fatality rates can be higher (someone who dies from it has a 100% fatality rate!), so Lombardy had higher rates in some areas for a variety of possible reasons: more frail old people or perhaps because those communities lacked prior immunity as previous related diseases passed those regions by unscathed, leaving them more vulnerable this time round. That latter explanation is now looking the more likely. It may also apply to whole countries.

      The issue is always what question you are really after. It is actually damned hard to give a good definition of what one is after.

      Loosely speaking, the “case fatality rate” is what a doctor in a hospital wants to know: given that someone with serious symptoms is in the hospital, how likely is that person to die? Those rates are thus estimated by researchers from hospital data and can be as high as 15%.

      Loosely speaking, the “infection fatality rate” is what a GP wants to know: given that the person in front of her has tested positive for this disease, how likely is that person going to die? That rate is estimated by researchers from dividing the deaths by the number of those tested positives in the most representative samples they can find, giving them an estimate between 0.4% and 1%.

      However, many people who were exposed to the virus will not test positive on the available tests. And of course, it is never a realistic thing to say 100% of the whole population will get exposed. So you have to factor in how many you think were exposed but did not get the disease sufficiently to test positive. And you have to think of how many in a population would get exposed in some kind of worst-case scenario (or best-case scenario depending on what your goal is). My initial judgment on those items in mid-march was that in the population of the West as a whole, probably for every person who is infected enough so that the tests tell you they were infected, there are three others mildly exposed who do not show up as positive in the tests. That’s what the cruise-ship data pointed to: only 700 out of 3,700 on that ship tested positive whereas I reckoned all of them must have had some level of exposure. Hence the 0.2% initial estimate. The CDC uses a lower infection-fatality-rate (test-based) but a lower ratio between the positive tests and the exposed non-positives, giving them a similar end number to mine.

      When thinking of scenarios, its handy to have a 100% exposure rate as an important one, but of course one in any disease tries to shield the most vulnerable (that’s why nurses have flu tests and jabs: to have herd immunity around their patients).

      • Conrad says:

        I disagree with that — especially in terms of what is likely to happen, because a lot of the data comes from places with very good hospital systems (Singapore the clear winner!) and small (manageable) numbers of cases. This is not going to be the future for most places. If you look at places with still excellent health systems but not super systems where serious break-outs have occurred, the rates are clearly much higher.

        For example, according to the coronavirus odometer, in New York there have been 413K cases and 31K deaths with very high testing. The population of New York is about 8.4 million. Assuming every single person in NY has had the coronavirus you already have about .025% death rate from those contacting the cirus. But what proportion of people have actually had the virus? Who knows as the test is so inaccurate as has been the sampling, but the latest testing I can find is less than 30%. So even if we massively over-estimate this with 50% you get to a .5% death rate.

        Now extrapolate this to big poor dirty cities (like those in Brazil and Africa). Unless the effect of hospitals is close to zero, it is going to be a disaster.

        • Conrad says:

          If you want another example, if you assume 20% of the entire UK has had coronavirus (12 million), then, with 43K deaths, you get to 0.36%.

        • Conrad says:

          A bit of random web searching gives me these further numbers for France and Spain (I wouldn’t vouch for the data):
          Spain Deaths 28,327, Antibodies 5%, pop 47 mill
          France Deaths 29731, Antibodies 4.4%, 67 million

          So both France and Spain get over 1% on these numbers

        • Conrad says:

          ok, I did further searching. Belgium and Sweden are high too.

          More importantly, I also looked at the sensitivity of the current tests to look at your 3-1 missing claim.

          The Doherty Institute tested some of these and the results are here:


          You can see these tests are not super until 14 days after getting symptoms (still not 3-1 though). After that they still could be better, but all of them reach at least 80% sensitivity after 14 days. So whilst I agree the results from the cruise ship don’t seem realistic, they may have been done by older tests. The newer tests are not too bad for answering this question.

      • Conrad says:

        And today the US CDC suggests there are likely to be around 20 million cases based on blood tests (methodology unknown as far as I can tell, apart from “blood tests” from samples previously collected). This gives a death rate of .5-.6%.

        So if the Euro antibody data is somewhat underestimating the number due to problems with the tests (which are clearly not as nearly bad as assumed to get to .1-.2%) and simply more deaths because Europe has an older population than the US, then this data all looks within the same ball park.

        • Conrad
          The data would be skewed towards those who were ill enough to attract attention, particularly in March through to early May.
          Another possibility is that some may have some degree of intrinsic non antibody ,resistance to becoming infected in the first place. Perhaps something analogous to the benefit of having ,unexpressed sickle cell anaemia gene.

          • Conrad says:

            I couldn’t find the methodology they used (perhaps it is somewhere) — that’s why I said that. However, the CDC has great statisticians and biomedical people, and they are sure to be aware of problems with the tests and how to do proper sampling and what you can conclude from it (unlike the WHO, who I think deserve to exist, but clearly have done a poor job with the coronavirus and health advice). I assume the methodology is living somewhere online so people can think about more (or will be). Given what we know now, imagine that the estimates are getting pretty reasonable (and they are certainly within the ballpark of everyone else). Perhaps most people are wrong, although I think the onus is on them to come up with the evidence now. For example, it’s clear the tests are not missing 2 from 3 cases after two weeks, and that you cannot generalize these to tests used with the Ruby Princess at the start of the pandemic here which were known to be crap.

            • Conrad the CDC is also the group that early on blocked a number of University groups from developing their own tests. The CDC believed that they could‘do it inHouse’ the labs the CDC authorised to develop at test failed rudimentary standards re cross contamination and as a result the US lost 6 critical weeks. The FDA also blocked hospitals on the west coast in those early six weeks, from treating early suspicious cases on the assumption that they were Covid19 cases .
              Am also told that the CDCs computer related systems are fairly arcane.

              On a local level I am told the NSW health department operating systems are actually DOS based i.e. date to the 70s!

  19. May be of interest :


    Prof Eleanor Riley, Professor of Immunology and Infectious Disease, University of Edinburgh, said:

    “This is a useful addition to the knowledge base around serological assays for COVID-19. It pulls all the publicly available data together in one place, will be periodically updated as new data emerge and it gives us a baseline to work from. The analysis only goes up until mid-April and thus only captures data from the early stage of the epidemic; many of the questions we would like answers to cannot yet be answered but regular updates should alleviate this problem.

    “However, it is already very clear from this analysis that antibody testing early in the course of disease is unreliable and thus antibody tests cannot (and should not) replace virus detection for diagnosis of acute cases except where the time-course of disease is already well advanced. This is a useful and important clarification.

    “The data also make it very clear that most, if not all, commercially available tests are not sufficiently accurate to warrant their use outside of the healthcare setting.

    “The assessment of the quality of the available data, and in particular the recommendations of how data should be presented in future publications, is also very useful and should lead to better quality of data presentation in the future, enabling some of our outstanding questions to be answered.”

    • Conrad says:

      Their results show pretty much what the Doherty ones show. They are not great clinically (where you want very high accuracy), but the sensitivity for answering question like what percentage of the population have been infected after weeks is enough (especially given the estimates of false negatives and positives). The actual review is readable and free.

      • Are the results good enough for the purposes of an , immunity passport?

        • Conrad says:

          That would be really cool (you should tell the government :)!) — If the different tests were differentially sensitive to different aspects of the virus you could also easily get the probability of error really low (e.g., “you must have a positive reading from 7 of 8 tests”).

            • Conrad says:

              Even better — I had a search around for kits, and it appears if bought in bulk they are only $30 or so each (The Aus government bought 1 million for 29 million dollars that didn’t work to their satisfaction, but it gives some indication of price). Even if they cost $50 and $100 to administer, that’s only $500 for 8 tests. I’d pay that so I could go overseas without having to be in quarantine on the way back (if I’d had coronavirus — which I doubt). So it’s very affordable. Even for tourists coming into Australia, it probably isn’t the worst cost they could have.

              • if my wife and I had to spend $500 each plus the costs of travel to Qld so as to spend a day with her father we would do it but we couldn’t afford to do it that often . And I’d imagine that if he was seriously ill it would be hard to do 8 tests quickly enough.
                On the other side if it was something bulk billed the costs to government could be fairly big?

                • Conrad says:

                  You’d only have to spend it once presuming you get immunity for many years.

                  The tests are also pretty fast — for influenza you already get a turn around in a few days. You could probably pay extra to get them even faster if there was enough demand.

  20. Conrad
    That thread is too squeezed
    Re “presuming you get immunity for many years.”
    Thought the jury is out on that, isn’t it?

  21. Conrad
    I gather that the CDC now estimates the total number of infections in the US to be about ten times the number of known cases. Combined with the UK figures re the large number of infections contracted in either hospitals or aged care facilities the mortality rate for the wider community looks relatively small when compared to earlier estimates. And it looks like very intense transmission clusters require special circumstances : hospitals , aged care facilities and meat works seem to be the main problem areas . ( Feel that hospitals will need revisit their barrier nursing skills)

    Have felt all along that the viral fire has behaved in some places like fire landing in a broad leaf deciduous forest with very little under-story growth i.e. trees whose leaves are literally flame retardant, and in other places the viral fire has behaved like it was landing in a eucalyptus forest with great piles of fuel on the ground i.e. a forest whose leaves are literally full of accelerants ,over-topping a great pile of very dry fuel.

    • Conrad says:

      Small compared to the initial suggestions — but I find that most of the population based studies are pretty consistent. The CDC death rate estimate is .5-.6% of people that catch it. The other studies which all come from Euroland as far as I’m aware are typically around 1%. They may have been somewhat overestimated due to properties of tests, but Euroland also has an older population, and as you note some of the places had very bad luck with things like old-age care facilities. The US also hasn’t got to the end of the lag (i.e., currently infected people dieing), whereas most of Euroland more or less has. So that number will increase in the US compared to Euroland even if everything stopped tomorrow.

      This also explains why places like New York, which have a death rate of 1615 people per million are still getting around 800-1000 cases a day, despite a very heavy lockdown and only a recent relaxing of the some of the rules. If the death was .2%, then they should have herd immunity by now (80% population resistance) but you are still seeing many cases despite much more difficult conditions for transmission.

  22. Michael Yugen says:

    Hi Paul, reading your excellent piece in late October feels like you were predicting the future – living in the digital age of cognitive indolent herds.

    • paul frijters says:

      thanks. Yeah, scanning possible futures is what I do. I dont always get them right though.

      • Conrad says:

        The real problem in terms of evaluation is that you will be judged in hindsight on the outcome, even if an obviously unlikely outcome happens to eventuate that you sensibly didn’t predict.

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.