A seminar/workshop on wellbeing cost-benefit analysis applied to covid

Find below the video of a seminar for the Australian Institute for Progress done a few weeks ago detailing the basic cost-benefit view of the current pandemic.

The slides of this presentation are here: Presentation CBA Covid May 2020_V2

Most of the basic numbers in this slide-deck and workshop were discussed earlier on troppo, such as here.

This entry was posted in Coronavirus crisis, Employment, Health, History, Politics - international, Politics - national, Science, Social Policy, Society. Bookmark the permalink.

79 Responses to A seminar/workshop on wellbeing cost-benefit analysis applied to covid

  1. Robert Walter says:

    I confess I didn’t watch the presentation, but I did look through the slides. Why aren’t incidental benefits of the lockdown considered, only incidental costs? For instance, the flu season in Australia looks to be considerably reduced compared to previous years. Shouldn’t we take that into account?

    • paul frijters says:

      because there is an extremely large list of further small costs and benefits involved and those who have attempted to look at those have generally found the further costs far outweigh the further benefit, but that they are smaller items. Air quality, abuse in the home, severe mental health problems, family reunion problems, loss of exercise, rotting harvests, famines in poor countries, interrupted innoculation programs, etc.
      The slides pick out the big-ticket items which we know from a wellbeing point of view to matter the most. If I’d add the smaller stuff (which is not always easy to do), the cost benefit calculation looks even worse for the lock downs and the general reaction to the virus.

      But you are welcome to give it a try yourself, that is the whole point of the slides and my post: equip you with the basic to do some of these calculations yourself. Have a go!

  2. Conrad says:

    I may have missed it, but I don’t think the control group which you seem to be using implicitly is life as we know it. If you have large numbers of people dying, people will choose do to less anyway, and so you’ll get more unemployment anyway. That was certainly the case in HK in the first version of SARs when GDP took a large hit (even though the big money spinners in HK, like banking, were not really affected) and unemployment hit a record, even though there wern’t that many deaths. Just lots of paranoia. In addition, if, like Sweden, the numbers of cases are still going up, presumably the paranoia associated with it will continue and hence lower levels of damage may simply be drawn out for longer.

    Apart from that, unless you can convince other countries not to lock down, you will also get a big hit because all your trading partners are not going to be buying as much stuff, and some may simply not allow tourism and other services from/to your country. So you may pay potentially get left out when other countries are back in.

    So whilst I agree the costs of lock-down are gargantuan, the difference is really what we need and this depends on your assumptions on how bad countries that don’t lock down will be hit.

    • If the virus can’t be eliminated and the odds against achieving that are large , the question surely becomes , how can we protect the most vulnerable those over sixty five and those a bit younger with diabetes, without all the economic and health harm done by nation wide lockdowns ?

      • Nicholas Gruen says:

        Well the virus is on the way to elimination in Australia and perhaps already eliminated in New Zealand. So the question is what the cost of ongoing interdiction of travel is. At the moment the cost is very low because if we didn’t interdict travel, other countries will interdict us. And with success in eliminating the virus or having it at extraordinarily low levels (where we are now) maximises our chances of opening up to other countries within the ‘bubble’.

        This idea that we can’t eliminate the virus has dominated discussion fro the get-go and was an important reason we didn’t take action when we could have. Elimination was the advice from very early on of people like Yaneer Bar-Yam (a complexity scientist of some note) and Nassim Taleb (who’s ideas are inspired by actual risk taking in a market rather than the intellectual simulacra of the same thing in the economic academy which imagines us all to be in a casino where we know the odds). It worked for SARS at vastly lower cost than muddling through.

        • And, if it turns out that it cannot be eliminated what then?

        • Aaron says:

          Very few countries have eliminated the virus. The general trend I’ve seen is:
          1. Very early action – it’s frankly too late for most of the world. China might be an exception here, but it’s also a dictatorship.

          2. Rather draconian tracking measures imposed on the population. I read that NZ will require everyone to have QR codes: https://www.nytimes.com/2020/06/08/world/australia/new-zealand-coronavirus-ardern.html

          “Ms. Ardern also announced that QR codes would be appearing wherever people gather. She asked businesses to remind people to scan the codes into the government’s contact tracing app to make any future outbreak easier to track and isolate.”
          I don’t know if Australia has this, but it’s a common feature of every country that claims elimination or anywhere near it. This won’t be tolerated in places like the US and isn’t feasible in poor countries where lots of people do not have smartphones.

          The ship has sailed for most countries. Even the WHO has admitted the virus is endemic in most countries. Trying to harp on this seems pretty futile to me.

          It worked for SARS because it’s a vastly different virus. It doesn’t transmit as easily.

          • Conrad says:

            That’s not entirely correct. SARs was found relatively early because it got to HK very early and so there was some reasonable chance of publicising it because historically, unlike mainland China, messangers didn’t get shot. It also appeared in Southern China where people are used to taking much stronger precautions against viruses due to the prevalence of other tropical diseases — e.g., spraying disinfectant on pools of water and plants is common. This meant that it was largely wiped out in Southern China because people were used to strategies to contain nasties, and most of the other places it got too either had it in tiny numbers (e.g., Canada), or were also used to nasty viruses like Vietnam (as can be again seen with Covid). If it had hung around many months before being found like Covid and was exported everywhere, the situation would be have been very different.

            • Aaron says:

              SARS is far less contagious than SARS-COV-2, isn’t it? It also doesn’t transmit through people without symptoms, does it?

              • Conrad says:

                In terms of more infectious — I’m not sure — it was eliminated and presumably only exists in labs now (although probably less infectious given how infectious Covid is). It certainly could transmit without people being obviously sick — that was part of the problem. Many of the cases came from one area where it is thought poor plumbing was responsible (look up Amoy Gardens). Alternatively, there we 1755 cases of it in HK and 1100 in Covid cases. Hygene conditions were worse in SARS (this was when spitting was banned!), although people took it pretty seriously. It was the quietest I’ve ever seen HK.

                • Aaron says:

                  Well, the big thing about SARS-COV-2 is the large number of people who don’t realize they have the virus. People with no symptoms or mild symptoms. Why do you think there are all these serology studies finding massively larger number of infected people than we ever thought? None of this occurred with SARS.

                  • Conrad says:

                    If I remember, they did do such an investigation on a limited scale (it being 2 decades ago after all — I remember it taking weeks before they even knew it was a coronavirus) and found a similar pattern (The documents may still be hidden somewhere on the HK health department website..). It was thought bad enough that both the China and US developed vaccines, neither of which went into final stage testing because the virus was eliminated. This should give people a lot of hope in terms of a vaccine, and thus should be a factor in the balance of doing something and doing nothing. I believe you can also get some idea of how infectious this type of virus would have been based on the proteins it uses to bind to other things, so I suspect someone much more in the know than me on these types of things can probably give you an estimate of how virulent it may have been.

              • Conrad says:

                Here are some proper numbers for you: Estimated R for SARS before control measures was 2.9


            • Aaron says:

              The column width keeps shrinking, so I’ll reply here. From the document you linked to: “Transmission efficiency appears to be greatest from severely ill patients or those experiencing rapid clinical deterioration, usually during the second week of illness. Data
              from Singapore (Figure 1) show that few secondary cases occur when symptomatic cases are isolated within 5 days of illness onset”

              and that’s the key difference. SARS-COV-2 transmits well before symptoms making it much much harder to eliminate. Just not feasible once you get past a certain number of infections.

              • Conrad says:

                It still doesn’t mean SARs-I didn’t (we don’t really know) because it broke out in places where essentially 100% of the population wore a mask (HK + Singapore), which seems radically slow Covid spread even in high density areas (think Korea). I think this an unanswered questions and will presumably remain so. Incidentally, I assume that paragraph could equally as well describe Covid.

                • Aaron says:

                  See https://www.cdc.gov/about/history/sars/timeline.htm

                  SARS spread to 29 countries. Masks can’t explain this away.

                • Conrad says:

                  Aaron, as far as I can tell you are claiming that if SARs had occurred in the middle of China where spitting is a major sport (a disgusting habit noticed by Marco Polo!), had been hidden from the world for 3 more months, and where there were thousands more international Chinese tourists roaming the globe, it wouldn’t have made much difference.

                • Aaron says:

                  I am claiming this is a very different virus from SARS. Far more contagious. It spreads silently and there’s a gap between catching it and exhibiting symptoms. SARS had a 4 month lag between the first case in China and WHO notification. It had plenty of opportunity to spread, but it didn’t because it’s simply not as contagious. You’re stuck on this point for some reason and you’re missing my broader point: the ship has sailed for most countries on SARS-COV-2. Elimination is not realistic in most countries and the WHO has admitted as much.

  3. Nicholas Gruen says:

    Yes Conrad, it seems to me that this is a major problem with the way Paul is presenting his case. I’d like to be persuaded by Paul, but I don’t think he’s doing enough work at his end to satisfy me that he’s really trying to be as rigorous as possible (by which I mean candid with himself and others about the point you raise, not lots of numbers).

  4. paul frijters says:

    Conrad, Nick,

    do watch the whole video and the posts these slides are based on because you seem to miss how I Iook at various choices to various actors, not just one choice.

    What I do it look at the costs and benefits of one choice over another.

    One choice I examine is the choice of “humanity” in how to react to this crisis. I basically there juxtapose the likely outcome if humanity had treated this virus like just another disease (much like flu or AIDS), against the choice humanity as a whole has made. That calculation is extremely relevant for the question what we should do next time (or next wave), and also for what institutions we have or lack that made humanity take such a disastrously wrong option.

    The second choice I look at is much more local, much more “taken lots of other decisions constant”, ie what would it cost Australia or the UK to have another month of lock downs versus not doing so. Again, the outcome is overwhelmingly that its a major mistake to do repression strategies, independent of others. I also give numbers for anyone else to calculate costs and benefits of different here-and-now policy choices. It’s all rather simple because the damage of almost any repression is such a large multiple higher than the probable benefit, that the answer is the same.

    Nick, you seem to have missed in previous comments to previous posts that I have answered your question several times. I agree the fear and paranoia is a big problem, but the question then becomes how to counter it and to counter it as soon as possible. So a delay by a month in opening up whilst countering the fear is simply another month of misery with no benefit. Indeed, one merely deepens the paranoia. Even if this “a month delay in opening up” does not reduce the time spent in paranoia by a full month but only a fraction of it, the calculations still point to the huge benefit of pushing against the fear as soon as possible.

    Now, to be honest, you are being somewhat childish by playing the “but you need to tell me exactly what you think about 100 other things that would have happened if we didnt do X”. That’s whataboutery, self-exculpation in disguise. Any choice made is whilst neither knowing perfectly what is going to happen if one choice is made or the other. One builds a stylised view of the main elements, does one’s best to calculate the costs and benefits of that, and chooses. Or, in hindsight, one regrets and learns.

    In the background, there seems here to be the argument that government had no choice, it that “were all in panic and demanded our politicians did something really stupid so the politicians didn’t have a real choice”. Whilst there is some truth to that when it comes to the initial hysteria, that misses the point of looking at any past choice, and the role of responsible government and responsible science. Are you really going to give up the notion that the government should do the best for the population and simply follow the “will of the crowd”? Surely not. And are you really going to argue scientists should not aim for betterment but simply see the “will of the crowd” as a sacred absolute to be respected even if that crowd is hurtling towards disaster? I thought the big lesson of WWII is “Surely not”.

    So the questions are different and of the type “was this a smart thing to do for humanity” and if turns out the answer is “no”, then the question is why it seemed inevitable at the time and what we can do. Neither of you even seem to realise that that is then the big question: how to avoid such panics. They are not inevitable. We do not as societies simply want to accept and go along with waves of hysteria. Again, I wrote a whole post on this.

    • Nicholas Gruen says:

      Thanks Paul,

      I don’t think there’s much to be gained from calling us childish.

      You’re imputing all kinds of motives which are simply unhelpful. As you know, I’m pretty scrupulous in argument. I try to listen to what the other person says and respond to the strengths of what they say – or indicate where I think I can’t do that – i.e. where what they think are their strengths look like weaknesses to me.

      Throughout this I’ve challenged you to include in your scenarios government lockdown and some counterfactual which is self-lockdown. You simply refuse to do that and, in this case, say I’m ‘whatabouting’ and seeking precise numbers from you. I’m not. I’m after a realistic counterfactual. At least from what I remember of our exchanges you’ve simply not done that and told me that it’s really about the future (what to do next time).

      Now your scenarios may be of interest, but they’re not comparing two possible scenarios. They’re comparing what’s happened this time with some scenario in which in the future somehow we get people not to panic.

      But the humanity in one of those scenarios isn’t humanity – it’s tame putty in the social engineer’s hands – capable of being induced not to panic – and that by a social engineer who thinks the panic is irrational. As someone who’s been pretty focused on listening to my inner panic telling me getting COVID gave me a 1 in 200 chance of dying, I’m not very receptive to your desires to have me be more rational. I decide on what is rational for me and the terms on which I calculate it.

      I think what you’re arguing is an interesting exercise and I am quite well disposed to your argument about ‘humanity’ facing choices. But once the model is populated by human beings, they don’t behave like you want them to. They behave like they have in Sweden which – and you’ll be closer to the actual numbers as you’ve been making this case and I’ve just been looking on trying to do a bit of quality control on the arguments from the outside.

      But the stylised fact I have in my mind is that activity – for instance, travel in cities fell in Sweden by 70 percent and by 85 percent where there was lockdown. And further that GDP falls have been comparable in Sweden to its close Scandinavian neighbours. You may be able to update me on how that natural experiment is turning out. I haven’t kept up with the numbers. Meanwhile – and this isn’t based on any numbers but just my own commonsense, – I’m thinking people in Sweden are scared, and people where lockdown have succeeded are much less scared.

      I’ve just been out to dinner in Melbourne and the restaurant was full. There were two community transmission cases today, and none the previous two days. In New Zealand which went for elimination more seriously and faster, they’ve not had a case other than a returned traveller for several weeks.

      There was heavens be praised, lots of sensible things going on. Names and phone numbers taken, temperatures taken, spacing between tables, comprehensive cleaning of tables between sittings and so on. If we’d followed your advice we’d have lots more COVID about and I for one would not have been in the restaurant. So at least in my case the Australian government has got me spending more than I’d have been with a more liberalised approach.

      I’m not asking for any fancy or precise numbers from you. I’m after some sense that I’m dealing with someone who’s trying to think about two empirically likely courses of action. If you fancy you can get people to act differently in the future, that’s an interesting discussion, but it’s not very relevant to policy choices that have faced us through the crisis or which face us now.

      • paul frijters says:

        “it’s not very relevant to policy choices that have faced us through the crisis or which face us now.”

        I feel like I am talking to a stone here, I really am. You clearly feel irritated with me, but trust me, the feeling is mutual.

        Let me try one more time via a hypothetical similar discussion we could have had in 1915.

        The first world war is one year underway and its immense costs are becoming clear. Now, suppose we’re in 1915 discussing whether the decisions that lead to the war were the right decisions and what else should have been done. We agree on the immense cost of the war as it has turned out and how it looks like turning out the next few years. We are both appalled and agree the likely benefits of the war never stacked up against the likely costs, though some delusion in 1914 made people think differently. I make a calculation as to the total cost of the war and on that basis say we should set up institutions that prevent us from making those same mistakes next time, ie that we should think hard about the structures that failed us when it came to getting into that war. You say “good luck with that in the future, but how about now?”. I point to the running costs of the war and say we should make huge efforts to stop the war because it is in no-ones interest to continue it. What do you say? “Good luck with that” or “the only realistic options the UK politicians can choose in 1915 was whether we’d send the expeditionary force 10 kilometers higher up in Belgium or where we actually sent them, everything else was just not a possible choice at the time”? You might well have been right in such a statement (though there were attempts at stopping the war in 1915), but it really is not a useful way to look at it because it is defeatist and takes huge and ongoing losses as inevitable. There must be hope.

        Why else have we said for decades about the first world war “lest we forget”? And how else are we to argue for better choices now but to point out their running costs and compare the with a different alternative? When we see a clear improvement, we should argue for it. You do this too when it comes to the evaluator general, gov 2.0, the private provision of public goods, etc. You think you see an improvement and you argue for it, dismissing those who say “ah, but that’s not a realistic choice”. Your normal answer to that is “it should be and its really not such a big change if you think about it”. Quite.

        So for the choices in the here and now, I run these “running costs and benefits” scenarios. And of course we can snap out of this fear, as we have snapped out of other fears in the past (we are no longer afraid of witches or forest gnomes, are we?). I point out the immense running costs, even now, of the border control regimes in Australia, the social distancing, the disrupted schooling, etc., and I point out that the government should really push against the fear mongering. That to me describes the actual choices we now face in Australia, or do you really think its sustainable what is happening now in Australia? Do you really think it can quarantine itself till a vaccine?

        Let’s leave a discussion of Sweden be. I dont think it would help.

        On your current experience in that restaurant. You say that if there were no sensible precautions, then you wouldnt be in the restaurant. I want to ask wheher that is how you behaved in the past when actual risks were probably much higher? Have you ever been to a third-world country and eaten food on the street? Have you been to a malaria-infested country or a place with denge fever, or a dozen other much nastier diseases here your actual risks are much higher? And were you afraid then?

        If you on reflection find that in the past you were much less afraid of far higher risks, then surely you should conclude you have been fooled this time round, much more afraid of something than is rational for you to be. And that your current reaction to minor fears is basically irrational and costly to you and others. Or do you enjoy being afraid?

        Also, when you say you cant see a situation in which people would not be so fearful, you neglect the previous 50 years of our reactions of new viruses and bacteria. We have not shut down the country and the population has not been this afraid. The wave of hysteria is certainly an important phenomenon and one we should try to understand with the aim of preventing a recurrence, but we should not take them as inevitable acts of God. You take that fear reaction this time round as inevitable whilst it was not inevitable all the previous times, and we need to ensure it is not inevitable next time round.

        Let us just stop however. This is just not helping anyone. I have tried in various ways to convince you that we should not take the fear as a given but see it as the main enemy both in the here and now and in a next grand occurrence.

        I know you think you are trying, but it doesnt look like that to me as far. I would have expected you by now to write 5 posts on the difficulties of getting people to snap out of their fears. I would have expected you by now to write about how institutions get convinced of something stupid and then cover up. I would have expected you to recognise the theater around this fear, and how tradeoffs that would not have been made otherwise suddenly are made. I actually miss you being your normal self in this crisis.

        However, let’s just agree not to discuss it further until something external changes the way we view things and we suddenly see things in a different light. If even friends only irritate the hell out of each other in this debate, maybe reality needs to be allowed to intervene

        • Conrad says:

          Paul, your assumption is that people shouldn’t be afraid, but this is just an assumption — I’m interested in what people do, and there are endless examples of where it is almost impossible to get people to be have behave statistically logically and examples like loss aversion where the individual price is different to the overall one (i.e., insurance).

          I didn’t go through your examples I didn’t agree with. But here’s a few more: Your example of HIV highlights how well largely voluntary minor efforts solved the problem in many countries. Influenza is closer to Covid (somewhat less bad — I’d have influenza over Covid given the long term effects of Covid are potentially worse) but I can go and get a vaccination for that, so if I am fearful of influenza it is easy to solve the problem. Indeed I do this — I have had it every year for more than 20 because I don’t like being sick. If people are willing to put up with being sick for a month (which is costly for the economy and hence has a socialized price), they don’t have to, and many don’t.

          In terms of 3rd world countries, I’ve been to and worked in them, but got all the vaccinations despite most of the diseases being low probability. I also took further precautions (boiling water and so on). When I was in HK I did avoid contamination with SARs (as did most people) which would have been bad for the economy (which went from about 6% growth to about zero if I remember correctly). Because of this, the rest of the world didn’t have to put up with it (but it is too late now).

          I also think you are too flippant at dismissing pandemics simply because previous ones like SARs, H5N1 and MERs didn’t go anywhere. These didn’t go anywhere largely because of government intervention. Even if that wasn’t the case, the fact that a nasty pandemic that kills endless people will emerge some time has a probability of approximately 100% unless evolution stops. This means that any novel virus needs to be taken very seriously, especially at the beginning when it is unknown. This is because you cannot estimate the cost of the damage at that stage.

          So all I really want is an estimation of the cost including behavior change that you cannot stop by telling people they shouldn’t be afraid. This may still be cheaper than the cost of the things the government might change, but I am less convinced (although I think it was very useful that you put the cost figures of lock-down up). In Aus, for example, I am still surprised at the level of paranoia in Adelaide where I am despite almost zero probability of getting it. If this only affects 50% of the population, that is still a huge effect. So the problem of fear is largely unsolvable. In addition, if cheap measures (e.g., mask wearing) allow the 50% that are fearful to participate at 75% of what they would otherwise do without Covid, then that’s clearly a winner.

          • Hi Conrad,

            “the problem of fear is largely unsolvable”

            I think that is where we differ most. In a sense what I try to do is point out that the current fear is unusual and something we need to confront like many other fears have been confronted. Business as usual is not to do nothing but to treat a new disease as just another one, confronted like just another risk, of which there are many different ones. Mustering the hope and the courage to confront the fear requires a view of what we lose if we do not. In that sense, we are actors in this game. We dont need to take the hysteria as given and unsolvable. That’s not merely an assumption, but that’s what we have as a collective done with many fears now forgotten, such as vampires, werewolves, black cats, witches, other religions, etc. They paralyzed populations for centuries. An enlightened elite pushed the story that these fears were irrational and that we should aim for better. This was understood to be the job of science and the government. Perhaps we need a new civilization drive.

            • Conrad says:

              Good luck. You clearly have more faith in people to understand these things and change their behavior than me! It reminds of two graphs I used when I had to teach simple decision making. The most common form of accidental death is falling (not corrected for years lost — perhaps poisoning would be the biggest given kids do it to themselves), and the overall cause of death I think was heart disease or cancer (I presume these depend on how you want to count). Yet people still worry about sharks.

        • Nicholas Gruen says:

          Thanks Paul,

          As far as my tone is concerned, you misunderstand me. I’m not particularly irritated or any of that stuff. You were the one throwing round adjectives like “childish”. I don’t think I have have I? If you’re irritated with me, you’ll just have to own that.

          One of the conclusions I have come to – and notice again and again, not just on blogs but in life – is that a huge proportion of misunderstanding, which can then lead to animosity, is produced by nothing more than the impatience people have that others aren’t getting where they’re coming from. They can’t believe some of the misunderstanding of their position that they’re observing isn’t deliberate and off they go.

          Of course I’m guilty of that often – I can be impatient, but it’s mostly in life off-line. On the blog I try to push back against the tendency. I think I’m very rarely guilty of that here. (I offer this old blog thread as a [now] comic example, though of course, some take it to be a counterexample ;)

          I think I understand your position and in important respects I don’t think it makes sense because you’re not bothering yourself sufficiently with the counterfactual and you’re too confident you know both that people are being irrational and that you know what a rational response is. But I won’t try to prosecute that any further as per your request.

          My only accusation about your conduct of the argument is that you’re just not listening. Conrad feels the same way. There’s nothing in what you’ve said that is hard to understand. I’m pretty sure I follow it and support quite a bit of it. One of the questions I’ve tried to discuss is the trade-offs – in health and economic terms – between government-led ‘panics’ and the inevitable private action that will be taken if governments don’t.

          Hence my mention of Sweden. Apparently it’s all off-limits.

          As far as the future is concerned, I’m not particularly concerned. I think our handling of the pandemic has been so bad, and the measures we could have taken turn out to be so straightforward, that we’ll have this licked next time, even if the fatality rate of the pandemic is quite a bit higher.

          We’ll have lots of systems far more ready, we’ll use facemasks, we’ll act far earlier.

          In the meantime, we can sit back and watch the terrible carnage in poorer countries – which you highlighted from the get-go. It’s a terrible thing.

          • “you’re not bothering yourself sufficiently with the counterfactual”

            I think that accusation, which you have made repeatedly, is the bit that triggers me more than whether I think people get what I am saying. Like you pride yourself on being someone who really tries to listen, I pride myself on trying to play “what ifs” and “could we haves” from many different angles.

            But let’s rest it for a while and hope reality brings new perspective. Then we can have a go again :-)

      • Aaron says:

        Where do you get the “1 in 200 chance” of dying from? I hope you realize the risk from this virus is very heavily skewed towards the elderly and people with existing conditions. You can’t take an average fatality rate across the population and have it make any sense for you.

        You also talk about spacing between tables at this restaurant and then say it was full. Are you taking into account the fact that this table spacing will cause restaurants to lose capacity on an on-going basis? Perhaps many will close up for good.

        • Conrad says:

          This is my point to Paul, Aaron — that people will often change their behavior based on the risk they perceive, which. So, for example, Nick won’t go to a restaurant in Melbourne (where there is more risk) even if it is entirely legal — fair enough — why take a risk if you don’t have to (and if someone else goes broke because of it, this is just the way life is).

          So the effect of “do nothing” is not the same as “business as usual” because a large chunk of your population will take evasive action anyway — My favourite patisserie in Adelaide is now closed because of this despite the number of cases in Adelaide being tiny and almost no community transmission. Now imagine how much evasive action people will take if there are 500 new cases every day (which, in the scheme of things and ways you can become ill, is pretty small).

          • Aaron says:

            Conrad – the risk people perceive is largely driven by the media and government sharing actual risk factors. For young, healthy people the risk is practically zero. They’re taking far greater risks in their lives anyway. The media doesn’t like to share that information much, nor do public health officials.

            One must look at the tradeoff between shutting down everything and then continuing to pay a tax for keeping out the virus with a one time cost of a more targeted approach. You don’t seem to be doing that.

            • Conrad says:

              You’re just pointing out things occur, and what you want, not what I’m arguing. I never said we should shut down everything (I don’t think that). The most useful time to do that is when you don’t know what the properties of the virus are, lest we come across the big one. In this respect, if Covid had killed like SARs (or worse) the story people are using now would be entirely different.

              My point is that people arn’t logical, and are unlikely to become so anytime soon . That’s why they worry sharks. Paul has a solution above, although I’ll wait to see it working and implemented first.

              • Aaron says:

                You have to realize people are stricken by fear because the media is not sharing actual risk factors. The panic is driven by a skewed narrative. This can be changed and people will come around if they are bombarded with a different message.

                • Conrad says:

                  Good luck with that. Also, good luck trying to get people to understand base-rate information — Kahneman got the Nobel in part for showing people are terrible at this.

                  • Aaron says:

                    This concept is not particularly difficult to grasp. COVID mainly targets the elderly. If you’re young and healthy, you can get on with life. For some reason you’re stuck on the notion that people will be fearful no matter what the media and public health officials say while ignoring how the panic was created by those people in the first place.

                  • Aaron says:

                    Have you seen the recent images of people in Paris and Germany getting on with life? No masks, lots of people huddled together. That’s a pretty clear sign that the fear is not as inevitable as you’re making it out to be.

                • Conrad says:

                  1) You have the wrong control group. When you see people outside in France and Germany and what defensive behaviors they still exhibit, the correct comparison is a country that didn’t have a lockdown. Since Sweden is still going up in cases and France and German are basically near the end (hopefully), the total cost is still yet to be determined. The overall behavioral changes including more complex ones like what it did to trust in government is also still undetermined.
                  2) The concept of base-rate might be easy for you but isn’t for the majority of the population (and there are many other statistical concepts that fall into the same boat). Go and see if you can explain it to people (really — it’s almost impossible with many people). That’s why Kahneman got the Nobel Prize for it — because it is something many people just don’t get. There are other decision making biases which people never get over. That’s why we kill sharks and spend oodles on nets when driving to the beach is more dangerous.
                  3) We can argue forever about SARs. I personally think you are far too optimistic. If it wasn’t for a lucky set of circumstances, it would have been a bigger disaster. It may not have spread as much, but it would gone vastly further. Indeed, if people followed a “no need to do anything” scenario, it would have got everywhere. I suspect this is the reason the US and China developed vaccines for it — in case it came back. If it was easily containable as you seem to believe, it’s hard to see why they would have done that.

              • Aaron says:

                1. Swedes are just as fearless. I’m seeing this happen in many places, including Brazil where lots of people are gathering on the beach in Rio and where lockdown was a joke. Young people aren’t stupid as you seem to think.

                2. Whatever behavior change occurs will be temporary anyway until herd immunity is reached. When you go down the path of eradication, however, the behavior changes end up remaining for a much longer time. You’re basically not done until a vaccine arrives. It’s an on-going cost.

                3. I don’t know what base rate is or what sharks have to do with any of this. People carry on with their lives when the risks are put in perspective.

                4. I think you’re completely failing to acknowledge how much more contagious SARS-COV-2 is compared to SARS. First, you claimed masks, then you claimed it was caught quicker. When I pointed out you’re factually off on both you’re going off into some unfounded claim about doing nothing. The point is simply that it’s too late to eradicate SARS-COV-2. You keep going back to SARS without addressing this point.

                • Conrad says:

                  No they’re not — many have taken voluntary precautions, and they’re equally as screwed economically by the looks of it (not that I’d trust the Australian!): https://www.theaustralian.com.au/commentary/sweden-takes-the-big-coronavirus-risk-for-none-of-the-economic-gain/news-story/e8981b9c2f49086278e9ace1719ee67a . Presumably moreso if they get excluded from some trade things when other countries become more normalized.

                  Also, last time I checked, estimates of herd immunity were around 20%. So they’re nowhere near it. Is the population going to put up with seeing dead people for the next 2 years or however long it would actually take? I don’t think so. They already started taking action once too many people died and thus have the worst of both worlds (lots of Covid, high and extended economic damage). I’d rather bet on a vaccine. It’s amazing what billions of dollars will do to people’s creativity.

                  I agree it’s too late to eradicate Covid incidentally. I just don’t agree about what you think about of SARs.

                  • Aaron says:

                    Can’t access that link. There are lots of naysayers who claim Sweden is doing the wrong thing. These same headlines feature images of young people at bars and restaurants in Sweden, so they don’t look as afraid as you seem to think.

                    What’s going to happen to these other countries once they open up? They’re back to square one, while Sweden has herd immunity. They’re not that far from it in Stockholm. It only takes 2 years if you continue lockdowns and the other trace/track nonsense.

                  • Conrad says:

                    The estimates I’ve seen on herd immunity in Sweden in Stockholm range from less than ten percent (their own government’s testing — 7.9% a few weeks ago) to around 20%. Even if this was a good strategy, I find it hard to imagine the population is going to put up with seeing high levels of deaths for next few years to get high enough (unless they get massive winter season which would shorten the time — but that would put more pressure on the government to do something). They already closed upper school and the universities, and have had a large decline in consumer activity.

                    They will also have problems with tourism etc. from non-coronavirus countries which is presumably pretty costly too.

                    So I don’t see it as a politically realistic option.

                    • Aaron says:

                      The population is putting up with it and doing just fine. Once they get to herd immunity, there’s no risk any more to other travelers going there.

        • Nicholas Gruen says:

          I’m 63 with a history of bronchitis – though my lungs have been in good shape for a few decades now

          The case fatality rates for 60-70 were about 1% at least when I last looked. I’m also fairly healthy.

          Not sure why you expected me to provide economic analysis when I was talking about spacing in the restaurant.

          I’m not really arguing a line here. I was discussing my understanding of Paul’s case. Like I said, I wouldn’t have been in the restaurant if it were not for the lockdown – I’d still be at home, locked down. I’m not making grand claims based on my experience, I was challenging Paul to take that side of the ledger into account – and arguing that if we can eradicate the virus we get to the upside sooner – except for the costs of closing our borders which is probably 3 or so percent of GDP. But I don’t have a strong view one way or the other on Paul’s case, other than I’m sceptical that he’s considered this case well against plausible counterfactuals.

          • Aaron says:

            My point about restaurant capacity is we’ll be paying an additional tax on top of border closures. Reduced capacity at restaurants (and many of them shutting down because of this) and most businesses, elimination of certain businesses entirely like nightclubs, concerts, sporting events and so on. Many businesses rely on packing people into spaces and won’t be viable at all. This social distancing tax is also an on-going cost like the GDP hit from border closures that must be factored in for the duration that these measures remain in place.

            Regarding elimination, I’ve said so in another comment thread above, but we’ve got to be a bit realistic here. Do you really think elimination is achievable in countries that have crossed a certain threshold of infections? The window of opportunity to achieve elimination seems very small to me.

            For the risk: for the 60-65 age group, it’s 0.3%. So 0.5% is in the right ballpark. I’m unsure what bronchitis does, though, and whether it shoots up your risk massively.

    • Conrad says:

      Here’s some neat mutation data to add to that:


      • Fantastic graphs. I Love pictures:-)
        Any idea as to significance, if any ,of the changes to the viral genome over space and time ?

        • Conrad says:

          Some significance — in some cases you can post-hoc infer various mutations that occur that causes changes to the virus. In this respect, there was a recent article out that showed a few changes that occurred in the Euroland virus making made it more infectious, although whilst I know some things about modern genetics, I don’t have the scientific knowledge to consider the validity of it. In the least case, you get a good idea of how fast it mutates compared to other viruses.

      • Conrad
        The range of outcomes between nations for example Japan vs UK and within nations for example Rome vs Milan is huge. And the results for Indonesia even if they understate by a factor of 10 are surprisingly low.
        As well as ‘business as usual ‘ , ‘lockdown’ also seems a meaningless misleading nonexistent thing.

  5. A piece I wrote on the real similarities between wildfire and viral pandemics

  6. Nicholas
    While the economic effects of lockdowns have been much the same around the world the apparent health benefit side of the lockdown equation has varied an awful lot.

    It could even be that some aspects of lockdowns , in some places, may have made things far worse, health wise.
    For example if a group of eight workers who used a room for sleeping in, on a shift basis , were all forced to stay in that one room 24/7 for weeks at a time the possibilities for transmission would be horrendous . And in the UK areas were that kind of ‘shift sharing’ of living spaces map closely to the worst hot spots for Covid19.
    It could also be that links between lockdowns and better outcomes may be mere correlations. For example because of earthquakes and tides waves Japan has a very distributed community health nursing system that is I gather much more effective than the higher levels of Japan’s government – it may be that Japan’s fairly good outcomes had little to do either way, with the policies adopted at the top level of Japan’s governmental structure.

    And for Australia a bigger factor could have been, the large numbers of Australians with active direct family and biz links to SE Asia particularly Vietnam and or China , that meant that many of us were already changing our behaviour, well before the Australian government took any action.

  7. Conrad re Sweden
    None the less they have per capita done better than the UK.
    And even China has not been able to prevent new hot spots from emerging.
    It seems very unlikely that a vaccine could be available in the quantities and at the price needed to quickly treat most of the Sub Continent , China, africa and all of the Americas .

    Feel that we need to work out approaches to the ‘viral fire’ that are better targeted more nuanced than lock stock barrel treatments of entire nations. ( and inversely why in some-places the virus fire burnt like the area was choker with accelerants)

    • Conrad says:

      I agree — although Sweden’s problems are now ongoing in a qualitatively different way (international exclusion, cases still going up) to countries essentially in a low-volume targeting and maintenance phase, which apart from Iran, appears to work (although the effect of winter is yet to be seen). You also certainly don’t need the same rules in say, Adelaide, compared to Sydney, so it would be good if the government could be more flexible.

      I am also far more optimistic about vaccines — I don’t think they’ll be here any time soon, but given the race is on and there were already vaccines developed for SARs, I certainly don’t see it as impossible in a year (I wonder what the betting odds are from people more in the know..). If this happens, then sitting in a low-volume maintenance phase seems like a pretty reasonable compromise.

      • Aaron says:

        All those problems will stop once herd immunity is reached. It will take a couple of months more at most. The estimate of “years” you’re quoting is based on lockdowns whereby immunity develops very very slowly.

        It’s fine to be optimistic. You need to start looking at tradeoffs. A year of lockdowns (or some government pushed social distancing) vs. getting to herd immunity. I do not feel you or Nicholas are doing an honest assessment here. You are taking for granted the benefit of the former and asking Paul to jump through all kinds of hoops to make a case for the latter when he’s made a very compelling case already.

        You also have to be realistic about social distancing. People simply won’t comply. Even in places where mass gatherings are banned, they are happening underground. I’ve attended a few this week myself. It’s a fantasy land. Just like the war on drugs.

        • murph the surf says:

          Can you supply a study showing that immunity develops and how long that immunity lasts?
          I will add a link to a sobering read,
          It is incredible to me that people are suggesting we try and speed up the spread of infections when we still know so little about this virus and the consequences of its infection in humans.

          • Aaron says:

            There have been zero re-infections despite tens of millions of infections worldwide. Even the few suspected re-infections in Korea were ruled out. For other coronaviruses like SARS and MERS, we know immunity lasts for 2 years and 3 years respectively. We simply cannot keep saying “we still know so little” and say we need a study before making any reasonable assumptions.

            It’s incredible to me that people are suggesting we wait for a vaccine without considering the cost of doing so.

            • murph the surf says:

              That answer reveals no evidence of a study rather an anecdote.
              And reinfection may be rare in 2 years but the essential.point I was making was not only about biologicals but the
              total absence of knowledge of long term effects from Covid 19 infections.
              At the moment we only know vast numbers of people are chronically ill some very severely.
              The Imperial college has estimated that the lock downs reduced the death toll by 80%, self serving analysis aside, a cautious and economically damaging way forward is almost inevitable as people act in what they feel will be their best interests.

              • Aaron says:

                Tens of millions of cases worldwide is not anecdotal, nor is evidence from other Coronaviruses. There is a reasonable basis for this conclusion.

                As for long term effects, we won’t even know that for a vaccine.

                The Imperial College is the same entity that forecast Armageddon and was off by a huge factor. Now they use their own flawed model as a basis for saving lives. It’s laughable, frankly.

      • Haven’t kept up viz Iran. What’s the story there?
        Yesterday was talking with a old India journalist friend , from what he hears things there are not good and he doesn’t like the odds of them reaching a maintenance phase.

        I hope your optimism re vaccines proves true, however if in a years time it’s still ‘coming sometime’ ,what’s the exit strategy re maintenance mode?

        • Conrad says:

          John: Iran is going up again (back to high levels) — I’m not sure why. I also don’t think poor countries like India will get to maintenance mode (at least I can’t see how). It will be a long and miserable slog I suspect.

          If a vaccine doesn’t appear, it’s unclear what the realistic options are. It seems unlikely populations like Australia are going to accept dying in large numbers to get herd immunity, so I suspect we will just sit around with lighter regulations that hold things constant (which might be very light given the small numbers and good tracing we now have).

          Aaron — as someone that thinks that base-rates and statistical information is easy for other people to understand, you could think harder about how long herd immunity would take and what that would mean. At present, around 7% of the Swedish population has immunity to Covid. So 93% doesn’t. Of course, you don’t need everyone, so lets assume about 8 million more want to get Covid. If 10,000 people a day get it, this will take you two years.

          Now, if you want people to all get in 2 months (and if you actually had some way of happening), which is what you are saying is the maximum time, then you’re saying you’d like 4 million people per month to catch it. Unless you deliberately infect people, this is rather difficult to imagine, especially when a large chunk of the population is already taking evasive measures.

          • Aaron says:

            If you look at NYC, 20% infection was achieved rather fast. But, Stockholm isn’t as densely populated. So, let’s look at the numbers. Population under 1 million. 20% herd immunity threshold, so we need 200,000 infected people. 10% or are already infected, so you’re only talking 50,000 per month. That’s not crazy. The herd immunity threshold is likely lower anyway. Even lower in other parts of Sweden.

            Sweden’s hospitalization and death rates have both plateaued now.

            I’m simply going to ask that you evaluate the tradeoffs here. What is the cost for Australia to hold things constant vs. an intelligent herd immunity strategy? I haven’t seen you once look at that.

            • Conrad says:

              Aaron, no Australia government would ever survive 20,000 Covid cases a day for 2 years to get immunity, no matter what the trade off (let alone 5 million a month for 4 months, which would be impossible to achieve anyway unless you weaponise it into spores and dump on cities..).

              This is because no-one wants a 1% chance of death and 5% chance of serious and permanent damage (or whatever the actual risk turns out to be). As Murph notes above, it’s currently unknown what the risks are. I personally don’t want neurological damage or scarred lungs (the latter of which was common with SARs). Furthermore, those in the high risk group will be vehemently opposed, as will their relatives, and when they die their relatives will blame the government.

              If you need another insight into decision making from Kahneman and Tversky, they termed something called omission bias (not the best name). The basic idea of this is that people prefer to do nothing if there is a future possibility of something bad happening based on what they do, because if they do something and something bad happens, they feel guilty, but if they don’t do anything and the bad thing happens, they don’t feel guilty.

              So even if the trade-off is on your side (which I don’t believe for herd immunity), not only will the government not do anything, but people won’t want them to do anything either, because if their grandma dies, they’ll feel guilty.

              So your first problem is convincing people. Apart from the ridiculous claim (herd immunity in 2 months), I haven’t got into the argument over which trade-off is best, but there are clearly many people who will, so your position is neither so obvious or so ubiquitous that it is likely to happen soon.

              You can also calculate the probability of a vaccine too incidentally. There are at least 20 vaccines in testing, but only about 5% of drugs get through phase 3 trials (this varies based on the type of drugs — I suspect it would be better for this type of vaccine given it isn’t completely novel, but I could be wrong). This means the probability one will work is about 65%. So it is not too bad.

              We can take this further because if it takes, say, 3 years to realistically get herd immunity (if it’s actually possible — which it isn’t clear if it is), then that’s 3 cycles of people searching for vaccines if it takes a year per cycle. So the 65% chance gets even better for a comparable immunity outcome.

              • Aaron says:

                I really don’t know where you’re getting 3 years for herd immunity. It would take months, not years. Even the Ferguson models predicted months in the “do nothing” scenario. Like I said, Sweden’s hospitalizations and deaths are plateauing out now.

                Continuing lockdowns is not “doing nothing”. Quite the opposite.

                Of course, no one wants death or permanent damage (not sure where 5% is coming from). That’s hardly the tradeoff here. One must look at the lives lost by continuing lockdowns and social distancing vs. lives “saved”. You’re getting caught up in a lot of personal interest and not willing to do an honest evaluation of these.

                • Conrad says:

                  Aaron, just add up the numbers. To get herd immunity in Australia you need about 20 million people to catch it. How many people do you want each month to catch it? If it is 1 million, then you still can’t have it in a year. If you want “months” then somehow you need to infect 6 million people a month.

                  So if there is a break-out how are these people actually going to catch it when they hide, don’t turn up to work etc? For example, teachers will go on strike, as will other workers, parents will keep their children home etc. .

                  So it isn’t even physically possible without deliberate infection or germ warfare.

                  • Aaron says:

                    Not sure where you’re getting 20 million from. 20% of NYC got infected rather fast. You do realize the virus grows fast? It’s not linear growth. You’re not going to need a fixed number per month.

                • Conrad says:

                  Arron, 20 million is the number you would need to get herd immunity in Australia (i.e., about 80% of the population).

                  Until you work out how this could happen without massive behavioral change or germ warfare, you are not going to get it. People will run to the hills, not go out of their houses, leave work, not send their children to school etc. Western Australia, for example, still has its borders closed to South Australia (no cases vs. no cases), so that’s what the governments will think.

                  Even now in the US where people are pretty slack compared to many countries in terms of self preservation (i.e., not wearing masks, lots of people running around protesting close to each other etc.) and there are some super-high density areas where transmission is easy (unlike Australia), there are about 20K reported cases each day. If this means 10 times as many people have been infected (i.e., 200,000 people infected) a day it will take them 4 years to get to herd level immunity.

                  In addition, if they let the numbers go crazy, the greater the behavior change you would see in people, and hence the numbers will reduce.

                  • Aaron says:

                    I think 80% is not the herd immunity threshold. It depends highly on how densely populated the city or region is. If it’s done more intelligently by exposing the younger people to the virus, this can happen fairly rapidly. It won’t be linear – you can’t go off a certain number of people per day. It will be more rapid growth followed by a plateau, just like we’re seeing in Sweden. People aren’t running for the hills there. NYC is another example and probably the worst case scenario where the virus was just allowed to run around unabated.

          • Conrad
            The situation in India sounds worse than awful, the example given to me was , ten people who normally shift sleep in the same room are now all confined to that one room, except perhaps one or two of them that are still going out to clean floors in a hospital…

            From what Ive seen just how deadly the virus is and how widely it spreads, seems to vary a lot:
            In some places it seems like the ‘fuel loads ‘ were very dry, very flammable, dense and there were strong winds. In other places ,for example both in Japan and Australia it feels like the local ‘forests’ behaved like broad leaf deciduous forests whose leaves are (literally) flame retardants.
            Feel that we need to pay more attention to the lessons of the past.
            The great city pandemics of the 19C were not on the whole stopped by vaccines or by nation wide lockdowns rather it was by clean water, proper sewage systems and better housing.

            • Aaron says:

              John – any idea why the deaths per capita in India are so low? 100x lower than the UK, 50x lower than the US.

              Same thing for Pakistan, Bangladesh, Sri Lanka, Thailand, Indonesia, Malaysia. I wonder what’s going on?

              • Aron
                India is big and its governmental system(s) are ‘complex’. Anecdotal reports are there is a lot more deaths than the official figures. Time will tell.

                As for the SE Asian nations Id guess that their local cultures ( regardless of the merits of their higher level governments) are were more prepared for Covid. But nobody really knows.
                The question might really be ; Why was it so bad in the UK , Milan and New York ? rather than, why was it so relatively harmless in Japan?

            • Conrad says:

              It reminds me of those poor foreign workers in Singapore (and possibly elsewhere we are not hearing about them), except at least in Singapore they have exceptional health care and they did everything like bring them food when they were stuck in their hovels.

              As for not-so great pandemics — a few good habits certainly stopped influenza in Australia! I remember reading about London, in the no too distant past before they good decent sanitation — apparently more people died than were born in the city due to disease, and it only increased in population due to people coming from the country.

    • Conrad says:

      This is what Fauci says: https://www.businessinsider.com.au/fauci-offers-coronavirus-vaccine-timeline-phase-3-trial-2020-6?r=US&IR=T

      Chinese companies or/and their government are not sitting around doing nothing either (especially given the SARs-1 vaccine) — they already have vaccines in human phase II testing. Great communists beat Americans. They’d love that story at almost any cost. It would be better than putting someone on Mars right now so I don’t think the situation is comparable to many others in terms of time-lines (for which testing takes forever).

      I also doubt we know the natural course of the pandemic either, since there will be nothing natural about it given various interventions. To me it seems more likely the world will just fractionate into have-and-have-not countries for international travel etc. So this and other bits of the article are essentially irrelevant information to promote their position, as if herd immunity was inevitable. Another bit of promotion is that it can take 10 years to develop a normal vaccine. That might be true, and it might have been true, but clearly the situation is vastly more exceptional now, technology is vastly better, and countries like China simply have less regulations to do it quickly.

      • “The US plans to mass-produce vaccines before knowing whether they work. Fauci still worries about how much protection they will confer.” Hmm

        • Conrad says:

          Sure, you don’t know how effective it will be — this is a problem with all new vaccines, and vaccines for viruses like influenza that mutate quickly. But you don’t need 100% protection. If the vaccine reduced the symptoms by, say 50%, that would be enough such that it wasn’t a deadly virus anymore, and so you could start with it until better ones came along (which may simply involve fiddling the one that works — something vastly easier than inventing a new one). At least if you Covid, you wouldn’t die. Similarly, things like scarring on your lungs would be vastly reduced.

          It reminds me of the two times I’ve caught true influenza. One was just from being young and reckless :) and it knocked me out for ages (in France — a good trip wrecked!). I learnt from that getting the influenza vaccine each year was a great idea. The second was last year when I did have the vaccine but the vaccine only conferred partial protection. It was still bad, but not nearly as bad. Of course, that might be just be random variation, but even partial immunity is vastly better than none.

        • Conrad that quote has a whiff of desperation about it.
          Isn’t it possible that it might not work at all?

          Maintenance mode presupposes too much ‘luck ‘to my eye.
          What’s their plan B?

          • Conrad says:

            Anything can obviously not work, although the advances in biotech in the last decades have been astounding so one can reasonably contemplate solutions.

            An alternative way to look at it is that I think more than 20 companies are trying to find a vaccine and have candidates. This includes large companies that can evaluate risk vs. reward reasonably (e.g., Gilead, Pfizer..), and not just small biotechs that might see it as their last chance (there are some of those too). Obviously there will be a lot of free government funding on offer, but to me this suggests these companies, who have the best knowledge in the area and still have to put up a lot of their own money, don’t think it will take 10 years to develop and don’t think it will be redundant by the time it is developed due to herd immunity.

            I don’t see any other plan apart from building up groups of low-infection countries that can work together (countries like India that will probably get herd immunity in a few years can probably join too) — you are never going convince politicians or the general public in many countries that they’d all like to catch the virus to get herd immunity and that they will carry on as normal so they catch it. There was a massive change of behavior here, for example, despite the tiny numbers.

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.