Weekend reading: Winging it when it comes to the economy

Comments on this entry are closed.

• 1 Snowman May 23, 2020, 10:22 am

  @TI
  I think the John Kay link needs fixing to

  https://www.johnkay.com/2020/05/18/the-meaning-of-r/
• 2 The Investor May 23, 2020, 10:26 am

  Thanks @Snowman! I thought I’d checked them all, but there’s so many nowadays.
• 3 Gentleman's Family Finances May 23, 2020, 10:37 am

  If you really want to make some money investing in start-ups then have a look at VCTs and find that beyond the tax breaks, mega successful funds are as rare as unicorn feathers!
• 4 SemiPassive May 23, 2020, 10:55 am

  UK gilts going negative yield is the story of the week for me, and they closed the week with even the 5 year going negative. How long before the 10 year is also?
  The USA seems to be heading in the same direction.

  I realise Germany, Switzerland and Japan have been this way for a while, but it still feels like a watershed moment with a lot of implications, especially with the latest comments from the BoE governor.
• 5 Al Cam May 23, 2020, 10:57 am

  Get Rich Slowly article is definitely worth a read!
• 6 Pre ka May 23, 2020, 11:27 am

  I am interested to find out the views on the economy if a treatment or a vaccine is available by the end of the year…. especially as a lot of the companies like moderna seem cautiously optimistic it can be done.
  It is notoriously difficult to predict the economic progress from here but most projections do not seem to take the human technological advances in finding a tx or vaccine into account.
• 7 The Investor May 23, 2020, 11:52 am

  @Pre ka: Morning! You write:

  It is notoriously difficult to predict the economic progress from here but most projections do not seem to take the human technological advances in finding a tx or vaccine into account.

  On the contrary, I think pretty much all of the projections assume we’ll either get a good treatment or a vaccine in the next 6-12 months. (Which I expect, too).

  If we don’t get either a treatment and/or vaccine and we have a free-floating disease around indefinitely *and* the IFR is around 1% (still TBC but not impossible) then I’d say the outlook is probably worse than predicted, unless it fizzles out like Spanish flu of course.

  Modelling the economy is as you say very difficult. But we know for a fact we’ve switched much of it off, everywhere, for two months. This will have consequences.

  For instance, more than half of all Americans could soon not have have a job (for the first time in history IIRC).

  As Napoleon once quipped: “Quantity has a quality all of its own.”
• 8 ermine May 23, 2020, 11:59 am

  > most projections do not seem to take the human technological advances in finding a tx or vaccine into account.

  I dunno. Take a look at the S&P500 on a five year basis and it seems to be saying pretty much just that. Ticketty-boo, nothing much to see here, move along now.
• 9 Sara May 23, 2020, 12:01 pm

  You’ve found some interesting Covid links this week – the Atlantic one on “stay at home” especially. It reflects my experience where those who have completely stayed at home are now overly scared to leave their homes. The initial point was to scare us all into staying put but one big problem now will be reducing that fear (as the England government are now discovering with schools, teachers & parents). I have been out during this time – both for food shopping and returning to work after the first 3 weeks. Initially I was terrified. Now several weeks later it still gives me low level anxiety but I can live with it and in fact find my mental health is better for being in work and seeing a few other people (at a distance). But my elderly mother who has not been outside with others for over 12 weeks and is only experiencing Facebook and the TV news is increasingly reluctant to go outside even for a walk and is obsessed with the infection numbers and deaths on TV – even as they are dropping to very small numbers.
  Everyone went “Huh” at the Stay alert message but one of the government’s problems is going to be reducing the fear level to a point where people will go out but where they willl continue hand washing & social distancing. Not easy.
  And how many public area toilets have we all visited which were gross and useless – no soap, touch flushes, trickle of freezing cold water or unusable boiling water, no proper dryer, overflowing bins?

  I have also had the fun (as a department manager) this week of reading my employer’s H&S guidance and the government’s guidance on making a workplace “Covid Secure”.
  I have A LOT to do & really should have invested in makers of plexiglass screens.
• 10 xxd09 May 23, 2020, 12:14 pm

  Many thought the amount of debt that economies were carrying was unsustainable and would lead to a crash and reset sometime soon
  As always the Fates deemed otherwise and sent us a pandemic (man made out of a Chinese lab?) instead
  So the crash that was coming -because that one of the default human conditions- ie life is up and down-is here
  We will recover because that’s what we do but lots of debt and lots of illness will test us all along the way
  Tin hats everybody
  xxd09
• 11 ZXSpectrum48k May 23, 2020, 12:31 pm

  People need to differentiate between what the economy is doing right now and what asset prices are doing. While consensus forecasts for 1-year aggregate earnings have collapsed, 1y,1y earnings (1y earnings, 1y forward) is only down modestly and 2y,1y (1y earnings, 2y forward) is virtually unchanged.

  So the market assumption is that earnings are going to mean-revert relatively quickly. Now why that is being assumed is open to debate. It could be a vaccine, it could be therapeutics. It could be that the market thinks that, even without a vaccine, aggregate company earnings will improve as people adapt to COVID being a persistent issue.

  The key point is that there is an assumption of only a short term collapse in earnings being reinforced by an assumption of easy monetary policy for the next decade. Basically no rate hikes till 2030, and in the near-term, the risk of further cuts or more unconventional policies. So asset prices can happily diverge from where the ‘economy’ and ’employment’ etc are.

  It also makes this very different from something like early 2009. In early 2009, consensus forecasts for company earnings were down for many years. Not only was the 1-year earnings forecast on the floor but so was the 2y,1y. The market did not expect the global economy to rebound at all. It assumed permanent damage had been done. By comparison, the in the US bond market (or to be precise swap market) the 2y,1y rate in the was pricing 200-250bp of Fed rate hikes. It was assumed rates would mean revert reasonably quickly.

  So in this crisis, the market has looked at what is got wrong in the last crisis of 2008/09, that earnings would not rebound and but that rates would, and totally reversed it’s view based on that error. It’s not unusual for the market to use the previous crisis as a template for how the current one will play out. It does create clear risks if those assumptions turn out to be invalid.
• 12 Snowman May 23, 2020, 1:32 pm

  Interesting Sara, to read about how and you are mother are dealing with the situation. Glad you now only have low level anxiety on going out. It’s completely understandable how you and your mother have reacted and it reflects true to me when you say ‘those who have completely stayed at home are now overly scared to leave their homes’.

  I’ve said in earlier comments about how appalled I am about the fearmongering that has been going on about the risks, both by government and media, and equally appalled by those saying that the fearmongering was justified as it got people to socially distance.

  The right thing to do in my view is always to give people clear information on what a risk actually is, so that they can take informed decisions, but it is not right to scare them.

  When you go into hospital for an operation they sometimes say things like there is a one in ten thousand risk of this or that complication etc. That’s the right way to approach things surely. It’s not right to pretend there is no risk at all, but instead the risk and its quantum should be clearly explained.

  The actual risk if you get covid-19 has been described as follows. And this annual risk assumes you actually do get the virus which many, perhaps even most won’t (and the extra risk is now being generally revised to be less than a year of extra risk)

  “To put things in perspective, the virus is now known to have an infection fatality rate for most people under 65 that is no more dangerous than driving 13 to 101 miles per day. Even by conservative estimates, the odds of COVID-19 death are roughly in line with existing baseline odds of dying in any given year.”

  Source
  https://www.realclearpolitics.com/articles/2020/05/21/how_fear_groupthink_drove_unnecessary_global_lockdowns_143253.html
• 13 Naeclue May 23, 2020, 1:56 pm

  I see that MSW is spouting dubious numbers to give her fund management pals a boost

  – IA UK All Companies return since December 2009 84%, Vanguard FTSE UK All Share 77%.

  How much of this supposed outperformance is explainable I wonder by:
  – Survivorship bias in the IA UK All Companies return calculation?
  – UK fund managers investing in non-UK stocks?

  Vanguard FTSE Developed World ex-U.K. Equity Index Fund has trounced the UK fund, so statistically the inclusion of some non-UK stocks in active managers funds should have given the active funds a boost

  Whenever I read that active managers have, as a group, outperformed, I always wonder who actually underperformed. Simple arithmetic tells us that there must be actively managed portfolios somewhere that did.
• 14 MrOptimistic May 23, 2020, 2:06 pm

  Interesting article in the FT from a reliable statistician.
  One point is the age dependency of death correlates with the natural risk so the virus more or less doubles your chance of dying in the next year, which is unusual for a pandemic he says ( does beg the question as to how many pandemics he knows of).
  Might be behind a paywall.
  https://ftalphaville.ft.com/2020/05/22/1590156197000/Spiegelhalter-says-vast-majority-of-Covid-deaths-would-not-have-occurred-in-coming-year-/
• 15 Hari May 23, 2020, 2:07 pm

  @Sara @snowman

  “obsessed with the infection numbers and deaths on TV – even as they are dropping to very small numbers.”

  Still around 350 people a day…. a plane crash every day….carry on for a month and thats 10,000 plus, not exactly trivial if it’s your mum/dad…..

  Life can start to get back to some degree of normality, but some caution and behavioural changes are still necessary.
• 16 Tony May 23, 2020, 2:25 pm

  That Valley of Coronavirus pic is brilliant!
  For those with a greater grasp than my limited one, is it reasonable or not to assume that UK gilt bond fund prices will rise in step with (if) future yields going further negative (as some predict)? Thinking of the Vanguard UK Government Bond Index Acc. Disclaimer: I recognise this is not the fundamental question when investing for the long term, gilts are water in the equities whisky and if I’ve got this right, some are avoiding buying gilts, preferring cash or more equities and future inflation could cause much damage to non-index linked bonds. But I struggle to grasp gilts, despite reading up.
• 17 Naeclue May 23, 2020, 2:26 pm

  @Sara @snowman

  “obsessed with the infection numbers and deaths on TV – even as they are dropping to very small numbers.”

  I think we need to be careful about extrapolating our own experiences to the whole population. I am a member of a sailing club with quite a few over 70s actively racing. Many if not most are chamfing at the bit for racing to start again and cannot understand why we cannot restart racing straight away. Relatives I have in their 90s are only staying in because we are urging them to do so and follow government restrictions.

  Everyone’s experience is different. I have very little experience of people obsessing over death numbers or fear of going out. This may be a problem some people have, but how widespread?
• 18 Naeclue May 23, 2020, 2:57 pm

  “At least 5% of UK has had coronavirus, with 17% in London”

  The very simple observation here is that the rest of the country has a lot of catching up to do. That catching up, were it to happen, would lead to roughly a trebbling of the numbers who have died from C-19 outside London. Possibly worse than that due to the different age profiles between London and the rest of the UK.
• 19 JohnG May 23, 2020, 3:21 pm

  @TI – As an MBA I feel entitled (generally) to enlighten you that the clear message is that you should no longer look for Unicorns and instead search for horned pegasi; and for just a few thousand pounds a day I can provide you with slick briefings and vague guidance on how to find them 😉
• 20 Snowman May 23, 2020, 3:30 pm

  @Hari

  Everyone is free to obsess or not, but would it be a good life strategy to obsess about the 10,000 or so who die of all causes every week in England even when there is no pandemic occurring? If not why do the 2,500 per week or so (350 x 4) who are currently dying with or from covid-19 deserve special merit and sympathy? Our personal death rates are all 100%, and we of course mourn and celebrate the lives of all our friends and family who die.

  Around 15,000 died in England in week 2 of 2018, that compares with an average of about 10,000 dying over an average week of an average year. So about 5,000 more in a week than the average. Should we obsess about them too? That’s what happens with seasonal flu it kills the most vulnerable. It’s sad but it’s life.

  How would you feel Hari, if a family member of yours died because they missed their screening for cancer or didn’t get a lump examined, or they died of a heart attack or stroke through not seeking or being able to access treatment at the hospital or their doctor’s surgery, or they committed suicide because of losing their job or they died indirectly through the financial and economic affects? I don’t like this sort of emotive argument, I’d rather assume we all have everyone’s wellbeing at heart, regardless of how we think we should progress. But if you are going to play that game then I will do the same. Whatever action we take we are trading lives against lives.

  Do you think the loss of future years for someone dying with or from covid-19 is the same as that for someone dying in a plane crash? Perhaps at a complete guess it’s 40 years of mainly healthy life lost for someone in a plane crash vs perhaps it’s 4 years of less healthy life lost for someone from covid-19. So do you think your plane crash analogy is a reasonable one?

  Some caution and behavioural change will clearly continue even if most lockdown measures are removed. Are you saying we should have exercised caution and behavioural change in mid January 2018 when people were dying of flu at a higher rate than they are dying of covid-19 now? Or should the healthy get on with life and we should concentrate our efforts on protecting the vulnerable?
• 21 Bob May 23, 2020, 5:34 pm

  @Sara. Good point about overcoming resistance to going out again. Back in the day I had a job to deal with to identify the small number of underperformers in a large organisation.

  A common theme was for staff to go long term sick when challenged. Without the day to day benchmark of a workplace, there was an increasing reluctance to the return. Such a fear of resuming work that some hoped for an ill health retirement rather than come back.
• 22 Charles May 23, 2020, 5:49 pm

  I didn’t actually believe that unicorn graphic was genuinely in the SoftBank slide deck. I thought you just made it up to make them look silly. Then I went and checked.
• 23 Learner May 23, 2020, 6:43 pm

  @Snowman, wrt fear of leaving the house, fatality rate is a pretty high bar. How about using hospitalization instead?
• 24 ermine May 23, 2020, 9:18 pm

  > Without the day to day benchmark of a workplace, there was an increasing reluctance to the return

  Ever occur to you that your sort hectoring people for underperformance was part of the problem?
• 25 Ruby May 23, 2020, 9:26 pm

  @ Hari – for you I prescribe a further homework with Lord Sumption; available freely online. I suspect @ Snowman may second.
• 26 Ste May 23, 2020, 11:31 pm

  @Hari @Snowman:
  I find neither plane crashes nor the flu particularly helpful when weighing covid-19 deaths against economic and societal costs of the lockdowns. The exposure period for plane crashes seems too different to reasonably compare it to an infectious diseases (we aren’t really all on a plane every day). As for the flu, there is certainly a wider philosophical and ethical argument to be had about what causes of death we accept as background mortality and which ones we don’t. But I would argue that simply accepting deaths from a new condition that roughly doubles mortality across all ages once infected as part of said background mortality would be extraordinary and likely not acceptable to most. Source has probably been posted here before: https://medium.com/wintoncentre/how-much-normal-risk-does-covid-represent-4539118e1196. Seasonal flu is recognised as a major contributor to mortality which is why there is a vaccine, we are encouraged to wash our hands and stay home when sick (probably just hasn’t been overly engrained in our collective behaviour before this year).

  A few (pre-)publications from a quick search suggest somewhere upward from 10 years of life lost per covid-19 death (e.g. here https://www.gla.ac.uk/news/headline_720672_en.html).
• 27 arty May 24, 2020, 1:02 am

  @Ermine

  When I worked in the civil service in Whitehall, one of my bugbears was all the appraisals and associated bullshit. However, it was a fact that there were a considerable number of staff whose output was effectively zero (leaving aside those that were achieving what they were asked but whose work was entirely pointless). In the main, they were left to continue being useless, because many managers took the approach of preferring to be nice and avoiding any difficult conversations. Occasionally, through normal staff movements, they would come up against a manager who couldn’t tolerate it, and went down the route of poor performance procedures, followed in pretty much every case by long term sickness.
  Apart from trying to get rid of them, therby hopefully precipitating a move into another job that they might be able to/actually have to do, what would be your solution? Leave them sitting in the office looking at facebook, reading the guardian or doing their online shopping, at our expense, until retirement?
• 28 Snowman May 24, 2020, 7:18 am

  Good Telegraph article ‘Lockdown saved no lives and may have cost them’ about Nobel Prize winner Michael Levitt’s analyses of the data.

  https://www.telegraph.co.uk/news/2020/05/23/lockdown-saved-no-lives-may-have-cost-nobel-prize-winner-believes/

  ‘Michael Levitt sent messages to Professor Neil Ferguson in March telling the influential government advisor he had over-estimated the potential death toll by ’10 or 12 times’………..The problem with epidemiologists is that they feel their job is to frighten people into lockdown. So you say there’s going to be a million deaths and when there are only 25,000 you say ‘it’s good you listened to my advice’.’

  The Ferguson model applied to Sweden (predicted 65,000 deaths by the end of April) is a good example of why the epidemiologists can’t spout this sort of nonsense.

  Naeclue, I’m certainly not suggesting everyone is frightened, so useful for you to point out that we shouldn’t extrapolate to the whole population. I’m just saying many many more are unnecessarily frightened than should be if they were given clear information and were able to make informed decisions. And as there is no realistic possibility that there is going to be a second wave that overruns the NHS, for the multiple reasons I gave earlier, I’m saying the government and media shouldn’t tell people that, as it serves to stop us getting back to something resembling normal life.
• 29 Snowman May 24, 2020, 7:34 am

  The Lord Sumption interview on the BBC is here for anyone who is interested. Not posting it to be antagonist, but because I think it gives a clear reasoned argument to challenge a view that suppressing people’s freedoms is justified for societal reasons.

  https://www.youtube.com/watch?v=eB1eqvAxQAQ
• 30 Bob May 24, 2020, 7:53 am

  @Ermine. Understand your point. Shortened the story for the post. A few really difficult cases, managers amongst them, who made their colleagues lives miserable and where everything else had been tried before. I was their last chance. As a union rep I wouldn’t have tolerated institutional bullying.
• 31 Snowman May 24, 2020, 9:07 am

  @Ste
  ‘A few (pre-)publications from a quick search suggest somewhere upward from 10 years of life lost per covid-19 death’

  I looked at a study such as this that came up with a 10 year figure in detail a while back and worked out the obvious and major flaw in about 5 minutes. I then went to see what others were saying on twitter and others had independently found the same flaw.

  What the studies have generally done is to split people into the same age, sex and co-morbidity mix. They then assume that people of the same age, sex and co-morbidity mix are a homogeneous group and work out the years of life lost from that related to number of covid-19 deaths in each group. In practice it seems likely that the 70 year old male with very poorly controlled diabetes is much more likely to die in the absence of covid-19 than the 70 year old male with very well controlled diabetes. And that same person with poorly controlled diabetes is also more likely to die from covid-19.

  Covid-19 is selecting as its sad victims those out of a heterogeneous group of people who have a lower life expectancy not an average life expectancy for the group. For example, having diabetes or not associates with covid-19 outcomes. So my starting point is that the severity of diabetes might be expected to associate also with covid-19 outcomes. The report I looked at in detail had a throwaway line about they didn’t think it made much difference, but gave no reason for them saying that. I hope that will get picked up at peer review as a major flaw.

  I actually did the sums myself accurately (not based on an average age at death which has a skewed distribution) and if covid-19 picks as it’s victim someone of average life expectancy for each age and sex grouping and using the UK population distribution, that already gets you down to about 14 years of life lost.

  I’m not pretending that I know how to adjust for this, although my feel is it’s significant, and 4 years is reasonable as a guess, but to say it’s more than 10 years based on a quick google is not.

  When you’ve had the chance to properly look at your study in detail perhaps you can come back and rejoin the discussion.

  There are a few people out there who seem to have an agenda that the years of life lost is a significant period perhaps 10 years or more, and there are those who seem to have an agenda to say people who are dying would have died anyway at exactly the same point. Looking objectively I don’t think either position is justified by the data. I think you can say it is somewhere between 0 and 10 years perhaps, but I don’t think it is 0 and I don’t think it is 10.
• 32 Snowman May 24, 2020, 9:21 am

  @Ste
  ‘But I would argue that simply accepting deaths from a new condition that roughly doubles mortality across all ages once infected as part of said background mortality would be extraordinary and likely not acceptable to most.’

  The article you linked to actually contradicts that the risk is completely additive and so it is double the risk

  ‘Also, as Triggle points out, there will be substantial overlap in these two groups — many people who die of COVID would have died anyway within a short period — and so these risks cannot be simply added, and it does not simply double the risk of people who get infected’
• 33 The Investor May 24, 2020, 10:02 am

  Morning all!

  We’ve been fortunate to enjoy a great discussion about Covid-19 here over the past couple of months, which has allowed a bunch of different viewpoints to be put forward and certainly helped with my understanding of the extent/limit of the pandemic and so on. 🙂

  Unfortunately across the Internet the debate is now devolving into the ‘with us or against us’ mentality that infects every disagreement these days.

  It’d be a real shame if that happened here.

  By all means challenge the facts / present counterpoints, but where possible can we please try to avoid lumping individuals into either faction that’s is/for or against either extreme of the spectrum when it comes to the appropriate response.

  There’s nothing too egregious here so far, just the inkling of a change in tone. So please let’s continue to reach for understanding and assume others are only doing the same. Thanks!
• 34 Dan H May 24, 2020, 10:03 am

  I think the penny will finally drop if there is a vaccine. I don’t see the NHS providing free yearly (assuming an endemic, mutating virus) vaccines to the entire population. I expect it will be provided to the elderly / high-risk only. This will leave the rest of the population finally realising they were always gonna get it anyway.

  Hopefully government policy will lead us to that before then.
• 35 Jim May 24, 2020, 11:39 am

  It’s depressing that other western nations have all followed the same plan, barring Sweden. What happened to liberty? One things for sure I wont be taking the mickey out of those gun toting MAGA yanks again, they might have some crackpots there but at least they value their freedoms and constitutional rights.
• 36 ermine May 24, 2020, 12:08 pm

  @Bob Fair enough – my bad! I saw a little bit too much of that methodology to thin out the ranks in the last GFC. But agreed, companies do have to get rid of some people, preferably in a civilised manner, which sounds like it was the case.
• 37 Factor May 24, 2020, 1:06 pm

  Personally I don’t mind in the least being labelled a reductionist, if that is what I am, but I await the final % excess deaths data across all the significant economies before taking any kind of view on the “rights and wrongs” of C19 policies. Until then, “what is” is what is, and I deal with it as it occurs.
• 38 Bob May 24, 2020, 2:35 pm

  @Ermine. My fault entirely. In an attempt to keep my post short, I accidentally offended my favourite blogger. (Well equal with MV). I should have provided more exposition.
• 39 Snowman May 24, 2020, 2:55 pm

  On the subject of vaccines, has anyone seen any good information or articles on how a vaccination programme would work, assuming of course an effective vaccine is developed and is scalable?

  Presumably if there is a vaccine that appears to stop you getting covid-19 or partially works then it may be the most vulnerable who will first get the vaccine, a bit like the flu jab? Or perhaps they will be the only ones to get the vaccine? But I’m guessing that the most vulnerable may not generate a sufficient immune response, or will there be an issue with the immuno-compromised also stopping them from being vaccinated? And you’d have to test it in them as a later test stage also rather than just healthy volunteers so would that delay things?

  If it’s a wider vaccine programme to achieve herd immunity because you can’t vaccinate the vulnerable and immuno-compromised, or because the tests have been on healthy volunteers, how will that work? You may end up having to vaccinate children who have a very low risk of dying from covid-19, or do you just assume they are low transmitters and give it to adults only? Any vaccine that was being made available say this year, can’t possibly have been properly tested for it to be safe to give to children, can it, you haven’t got long enough to identify any medium or long term side effects? And so it would be seriously risky to vaccinate children? And many healthy adults won’t want to get vaccinated because of safety concerns about a vaccine hastily produced.

  The chief executive of Astra Zeneca has been quoted as saying ‘We have to run as fast as possible before the disease disappears so we can demonstrate that the vaccine is effective’. I find that a bit puzzling, surely we want the virus to disappear and then we won’t need a vaccine? Or is he assuming it goes and comes back in the Winter or something?

  I’m not expressing any views at all here. Just trying to locate some good information and think aloud. I’ve seen some stuff on the stages of developing, testing and manufacturing a vaccine, but I’ve not seen anything properly discussing how a vaccination programme would work, other than waffly discussions about which country will get any vaccine first.

  (for the avoidance of doubt I am not an anti vaxer. I’m talking specifically about a covid-19 vaccine)
• 40 Ruby May 24, 2020, 4:10 pm

  @ Snowman – I believe what the CEO of Astra (along with an Oxford scientist I read about this am – maybe in the Telegraph) is saying that if they trial the vaccine on people they want the virus to be sufficiently common in the community to see whether those being trialled actually get it or not. If there is no virus around I guess its efficacy doesn’t tested in the real world.
  In my ignorance I thought they would vaccinate, sometime later give the volunteers a covid dose, retreat 10 paces and see what happened.
  I don’t think I’ll be rushing to get a rushed job vaccine. Rather ungentlemanly I know, but if the take up amongst the Nervous Nerys’ is sufficient won’t that do the job for the rest of us?
• 41 Snowman May 24, 2020, 4:50 pm

  @ruby
  With prevalence at 0.5% perhaps you’d need 400 volunteers and if half of those are in a control group then you might expect one person in the control group to get covid-19 each week and none in the vaccinated group if it worked. It would take a while to get significant results I guess.

  I read somewhere a while back they might try testing it on medical staff, on the basis that the may be more likely to get the virus (naturally) than others.
• 42 Jonathan May 24, 2020, 4:53 pm

  @Snowman, a comment on a different forum pointed me to a paper by Chinese scientists analysing immune responses in patients who had Covid-19 in Wuhan. (Sorry, don’t have the link any more, but I found the PDF still in my downloads folder). They actually tested for the neutralising properties of any antibodies, so asking directly whether they were likely to fight another infection. Elderly patients were actually found to have stronger antibody responses than their younger counterparts which is encouraging (but there is a possible selection artefact, these samples came from patients who had recovered without ongoing problems). The antibodies were shown to react with the spike protein portion of the virus, which is encouraging for the vaccine developers.

  However there were a few patients who had had a poor antibody response, but given they had recovered maybe they had a T cell response. There are so many strategies being tried for vaccine development, it is quite likely at least one will activate a T cell response too.

  September still sounds early to me for a validated vaccine. Basically, in a trial you ideally need 100 or so of the control group to get infected so that you can meaningfully measure protection in the vaccine group. Current transmission rates in the UK mean that needs either a very large number of participants, or a long time collecting data, or both.

  Assuming success, the chosen vaccination strategy will depend to an extent on the efficiency of the vaccine and the number of doses available. In a briefing the other day Chris Whitty said they would need to use that information to choose between the strategies you suggest (go for protection of most vulnerable like flu, or herd immunity like measles). They could always progress from one to the other – if a miracle happens and the vaccine is available in September it could be delivered alongside the annual flu vaccine to the older and clinically vulnerable population very efficiently, and only afterwards rolled out to everyone else.
• 43 Snowman May 24, 2020, 5:14 pm

  @Jonathan

  Thanks that’s useful and interesting information.

  With prevalence actually at about 0.25% at the moment (in the latest infection survey pilot) to get 100 of the control group to test positive you would need a group of 10,000 if my back of an envelope maths is correct, to get 100 in the control group to test positive over a 8 week period say (assuming generation period is a week to keep it simple and prevalence staying at 0.25%)

  2 x 100/(0.0025 x 8) = 10,000
• 44 Ruby May 24, 2020, 5:56 pm

  @ Snowman – 10,000 it is. Below is the article I mentioned earlier. If you can’t get to it the lead scientist is one Professor Adrian Hill.

  https://www.telegraph.co.uk/news/2020/05/23/oxford-university-covid-19-vaccine-trial-has-50-per-cent-chance/
• 45 Seeking Fire May 24, 2020, 7:21 pm

  Is there any chance we can get the lockdown madness extended for a few months? It’s been great really, financial services actually doing quite well, working from home so seeing the family lots, government policy has boosted asset prices significantly – thanks very much, even the weather’s nice and we’ve just about convinced the loony liberals that schools can open for our age group. In fact in a week’s time, when the children are back – I might be humming the greatest day of our lives by Take That as long as BoJo doesn’t p*ss on my chips and bring in wealth taxes. Obviously the small business owners who run the hair dressers, pubs and local restaurants are totally f*cked and poorer people are going to be disproportionately hit but hey you really can’t pin this one on the bankers can you eh no matter how much the government tries. So it really could be a case of win win for some people.

  I do get that the economy would have been very substantially hit anyway as per Sweden etc but I’ve always had the view if something is on fire don’t throw more fuel on it. And anyway not me guvnor this time.

  I haven’t quite figured out this clapping for the NHS. Are people clapping the decision by the NHS to push out old people into care homes and kill thousands of elderly people? Or are the masses banging the saucepans for people to avoid discussing difficult questions about properly funding it? Sort of like the labour government giving out medals to the armed forces like confetti in the Afghanistan when someone had their leg blown off to avoid difficult questions like why aren’t we funding the armed forces properly and sending people out in Snatch Land-rovers. Anyway, I’m afraid I’m a heretic and have refused to clap one week and join in the hypocrisy. Meanwhile if anyone still doesn’t realise the number one objective of the government was to stop the NHS being overrun not to save lives but to save political capital then worth having a re-think of your political compass.

  I’m not sure much behavioural change will come out of this. Depressing to see the queues for McDonalds. I suspect one change might be more people realise that the lack of trust in politicians is….entirely justified. The fable of the boy who cried wolf too often might have some resonance here. I admit to having been taken in by it for a few weeks before realising the bullsh*t of ‘we’re always guided by the science’. Political risk feels like it has been gently rising in the UK over the past decade and a global index tracker nicely hedges you.

  I don’t know if taxes will rise or not but I’m going to go out on a limb and say not. I don’t see any need – the UK can borrow at negative interest rates so why is there any need to raise taxes. And the conservatives are less keen than the other lot. Obviously I can think of better ways to utilise this strange turn of monetary events such as angling ourselves away from some of the inherent weaknesses in globalisation, rebuilding our infrastructure and decarbonising the economy but if the govt wants to p*ss this unique opportunity up the wall on monthly salaries then up to them.

  Back to personal finance, everything is highly highly levered towards low interest rates. If rates were to say go back up to five per cent, then prices across all asset classes would fall very substantially. Now I don’t expect interest rates to rise at all (at all!) but I don’t know they won’t rise. So it seems prudent to not be fully levered into the base case scenario – always invert your thinking and consider the alternative and be humble as to how little you know. Historically NS&I index linked certificates would have been quite useful here for some liquidity but sadly withdrawn…hmm I wonder why :). The next best alternative would seem to be holding some cash. Not in sterling though beyond say 12m of expenses – the old fruit bat dicky davies previously let the cat out the bag that he was in favour of a depreciating sterling so dollars seems a better bet. The downside risk to holding cash is clear – more of it is being printed and typically the more availability of something the less value it has.

  Some talk about negative interest rates…I expect that interest rates will go negative from here over time. How far who knows, probably not much further – maybe minus 1 or 2% over some time period. But mortgage rates could fall – I have a friend in the EU who has a 25 year fixed rate mortgage at 1%. Put that in your pipe and smoke it for real estate asset prices- we can manufacture that here if we really wanted to. What happens when interest rates can’t go any lower and we finally realise that party is over – well let’s keep printing that money. Am preferring $ TIPS and Gold here to Sterling (after 12m if expenses) for liquidity noting that the volatility, interest rate and currency risk is inherent.
  Enjoy the lovely weather next week.
• 46 Jonathan May 24, 2020, 7:49 pm

  Thanks @Snowman and @Ruby for the calculations which I admit I didn’t get round to myself earlier. If they are well funded 10,000 might actually be achievable, there may well be the willing volunteers nationally but the hard bit is recruiting the multiple local investigators and their teams on the requisite timescale (in a trial of this sort, it needs a proper medical screening of each person’s suitability for the trial and massive frequent monitoring; usually involving paying for local consultants with research expertise to do hours on top of their NHS work plus for each of them recruiting support from dedicated research nurses and scientists). Possible, but unprecedented.
• 47 Griff May 24, 2020, 8:48 pm

  So seeking fire. Your happy for the lockdown to continue to suit your own personal gratification and yet you want the Liberal teachers to now wing it with Germ infested kids just to suit your selfish needs and government policy. We’ll I’m not. This is not going to end well for teachers. But, you enjoy the sun you have earned it.
• 48 Algernond May 24, 2020, 9:43 pm

  @Seeking Fire
  $ TIPS – hedged or unhedged?
• 49 Snowman May 24, 2020, 9:54 pm

  Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals

  https://www.cell.com/cell/fulltext/S0092-8674(20)30610-3

  Don’t understand the detail at all, but it does give some fascinating insights into the CD4 and CD8 T cell immune response to covid-19
• 50 Seeking Fire May 24, 2020, 10:18 pm

  Hi Algernond – Unhedged. Ishares ITPS. Nothing magical about it. I wanted the exposure to $ away from £ – nothing on this earth could make me buy long dated gilts at the moment. Note this is just for a portion of assets and a slight hedge against the £ going down the gurgler (bought pre brexit fiasco) – worked well so far tbh but I know enough to know it’s basically ‘luck’ – I’d much rather they’d not worked tbh and the £ was stronger. They are also cheaper than UK ILG, which have a deeply negative YTM whilst $ TIPS have just started nudging below zero (I believe) and so the inflation protection is less expensive (someone correct me if I am wrong). Still I bought some more recently not withstanding the £ at near historic cheap levels – hopefully I will be wrong. The duration is around 8 last time I looked so yes if I/R went up they’d fall but tbh I felt I/R would fall in the US medium term so that has helped too – monevator wrote a very illuminating article on this a while back. Griff – fair enough, I was being a bit provocative.
• 51 Jonathan May 24, 2020, 10:24 pm

  @Snowman, that paper is heavy duty immunology but the real McCoy. CD4 T cells are involved in recognising foreign substances and initiating an immune response (often ending up with antibody production) while CD8 cells are associated with a “cell-mediated” response which allows the body to destroy its own virus-infected cells in a separate immune mechanism.

  All the patients investigated produced some sort of immune response. What was especially interesting to me was the fact that 40-60% of patients who had never been exposed to SARS-Cov-2 (they used historic samples from before the pandemic for this, so there should be no false positives) showed signs of some response: their speculation is that prior exposure to other coronaviruses (certain cold infections) might have conferred a level of protection. Which might explain why some people had such a mild infection.
• 52 Sparschwein May 24, 2020, 10:44 pm

  @Tony – that’s a good question that deserves more discussion. Have a look at @ZXSpectrum’s comments from last weekend.
  My (more simplistic) view is that gilts are poor value at <0.2% yield for the 10yr. Expecting any higher return is a speculation against the bond market. Compare gilts with ~1% on cash at NS&I. It's a rare advantage for retail investors, worth taking.
  And I'd second @Seeking Fire's comment – consider diversifying your risk-off position (USD, perhaps other currencies; TIPS, physical gold).
• 53 MrOptimistic May 25, 2020, 10:01 am

  @Jonathon. On an opposite tack I read an article which looked back at the1918 epidemic. This was especially bad for young people but on top of that there was a marked spike for 28 year olds. Not 29, not 27. The reason for that isn’t known but the authors wondered if prior exposure had sensitised their immune systems and increased the probability of a cytokine storm. Looking back 28 years they found a prior epidemic (Russian flu?).
• 54 Stoic Cyclist May 25, 2020, 11:00 am

  I’m thinking that the next few years might be a good time to finally buy a house, so I am reassessing the asset allocation in our ISAs (potential deposit money) to reduce volatility. The usual recommendation would be to put the lot in cash, but I can’t bring myself to put 1/3 of our life savings in cash at current rates, for possibly many years.

  I’ve been exploring Portfolio Charts and come to the surprising conclusion that gold-heavy portfolios like permanent portfolio or golden butterfly have had downside in the short term (a few years) barely different to cash, yet upside, over any time frame, that is dramatically better (median real return 4.8%).

  The worst case drawdown for permanent portfolio since 1970 is about -12%. I could live with that. And according to Portfolio Charts, cash is not really any safer – it experienced real-terms drawdowns of similar magnitude in the 1970s, when inflation spiked.

  This seems like a compelling case for permanent portfolio rather than straight cash for holding of a few years, but I’ve never seen anyone else propose this and I wonder if I am overlooking something important. What do people think?

  And is Portfolio Charts trustworthy? I’ve checked its predictions for permanent portfolio against data for PRPFX (a fund similar to ‘pure’ permanent portfolio) on Morningstar and they’re pretty similar, so it looks good.
• 55 Jonathan May 25, 2020, 11:08 am

  @MrOptimistic that is a fascinating story. At this distance, any explanation can only be speculation.

  But I take the point, the part of the population with pre-existing reactive T cells might be at risk of over-response to a Covid-19 infection rather than being partially protected.
• 56 Snowman May 25, 2020, 1:56 pm

  @Jonathon

  On a related note is there a danger of ADE (antibody dependant enhancement), for any vaccine that has been rushed a bit, especially given the numbers potentially exposed to the virus may be quite small at the moment, because of the low prevalence?

  Or have they got ways of ruling out ADE?
• 57 Snowman May 25, 2020, 2:27 pm

  The UK death toll has today risen by 59 deaths in England alone. The increase is one of the smallest yet since the pandemic began and the UK went into lockdown in March with the death toll now at 36,852.

  https://www.thesun.co.uk/news/11701665/coronavirus-death-toll-uk-small-rise/

  Good example of the lack of quality of reporting. The reasons the deaths are so low is that the England reporting system was down for most of yesterday!!
• 58 Snowman May 25, 2020, 2:37 pm

  They’ve now updated the article to add in the reason for the low number
• 59 Jonathan May 25, 2020, 3:17 pm

  @Snowman, yes there will be additional risks through not following trial patients long enough to pick up possible side effects. Though I assume one of the reasons for teaming up with a big established vaccine producer – separate from their scale-up capability – is that the novel immunogen will be presented in well established and fully characterised excipients, so perhaps they have reasons to think that risk is manageable.
• 60 ermine May 25, 2020, 5:07 pm

  @Stoic #54

  > This seems like a compelling case for permanent portfolio rather than straight cash for holding of a few years,

  I hold about three years running costs in cash. I have lifted this more recently, and I hold the excess in gold. The typical rule of thumb is that money you expect to call on in less than five years time (or is there to insure against emergencies eg redundancy) should not be in the stock market, bonds or cash are OK. I don’t use bonds (I have a DB pension which is bond-like) but I have held cash. I hold cash in NS&I ILSCS and small amounts in Premium Bonds and NS&I direct saver, but gold is the bulk of non-equity holdings

  Regarding the permanent portfolio, that is more suited to those in decumulation. Someone young would pay a greater amount in lost performance due to the poorer returns on cash and gold because they have a longer accumulation period, particularly if they are at the start of their working lives.

  For someone like me that is different, because I have no human capital left, and I have half the time horizon that someone who is in their 20s

  How compelling you find the case for the PP therefore depends on where you are on your life journey, though it also depends on how you feel about the future trajectory of gold relative to the pound. People say gold is a hedge against inflation and while that may be true over decade-long periods, gold can spend long periods being over or undervalued. If you bought into it in January 1980 you’d be a sick puppy for 20 years and if you bought into in 2013 then you would have had to wait for the Brexit vote folly to save you.

  If you were steadily buying gold over years, fair enough, but you tend to hear a lot of buzz about the Permanent Portfolio in times of turmoil. Now is one of those times.
• 61 MrOptimistic May 25, 2020, 5:48 pm

  I hold a fair amount of gold and one issue I have with it is I never want to sell it to rebalance because when it’s price is high, like now, there are issues stalking the streets and the mood, and a lot of advice, is buy gold! Hard to swim against the tide.
• 62 NewInvestor May 25, 2020, 6:57 pm

  I never want to sell it to rebalance because when it’s price is high, like now…Hard to swim against the tide.

  Interesting. How does it differ for you when rebalancing requires you sell high-performing shares?
• 63 Matthew May 26, 2020, 8:20 am

  I’ve heard that if you have a lockdown fine you can now appeal it under the law of “Cummings and goings”

  Twitter link for Barbosa (potc) meme for lockdown “guidelines”:

  https://mobile.twitter.com/PJBenson82/status/1264314960970203136?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed&ref_url=https%3A%2F%2Fd-32636644911818951642.ampproject.net%2F2005151844001%2Fframe.html
• 64 VS May 26, 2020, 8:56 am

  It wasn’t the pandemic that made you a cynic 😉
• 65 Tony May 26, 2020, 10:21 am

  @ Sparschwein thanks.
• 66 Snowman May 26, 2020, 2:08 pm

  Weekly ONS death data out for England and Wales for week 20 (w/e 15th May); they point out weekly data skewed this week and last by delayed registrations after bank holidays

  https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending15may2020

  4,385 more deaths registered in week 20 than the 5 year average, with 3,810 having covid-19 on the death certificate. So fewer covid-19 deaths than excess deaths.

  From the previous publication 3,083 more deaths were registered in week 19 than the 5 year average, with 3,930 having covid-19 on the death certificate. So more covid-19 deaths than excess deaths

  However if you dig deeper into the data, and look at the dates deaths occurred rather than the dates deaths were registered, the real position becomes clearer and the reappearance of non covid-19 excess deaths seems to be illusory. My figures are

  w/e 15th May: 1,941 excess deaths occurring over 5 yr av, vs 3,007 covid-19 deaths
  w/e 8th May: 3,282 excess deaths occurring over 5 yr av, vs 4,078 covid-19 deaths
  w/e 1st May: 6,161 excess deaths occurring over 5 yr av, vs 5,359 covid-19 deaths
  w/e 24th April: 8,779 excess deaths occurring over 5 yr av, vs 7,150 covid-19 deaths
  w/e 17th April: 10,561 excess deaths occurring over 5 yr av, vs 8,569 covid-19 deaths
  w/e 10th April: 11,709 excess deaths occurring over 5 yr av, vs 8,498 covid-19 deaths
  w/e 3rd April: 7,473 excess deaths over 5 yr av, vs 5,302 covid-19 deaths
  w/e 27th March: 3,586 excess deaths over 5 yr av, vs 1,920 covid-19 deaths

  If you look at these figures you see a trend that during the period up to the peak of the epidemic at the beginning of April, the difference between these figures increased. So for example the difference reached it’s highest point for the w/e 10th April where there were 3,211 (= 11,709-8,498) non covid-19 excess deaths occurring.

  The difference then starts to fall away and in the past 2 weeks we have moved to a position where the covid-19 reported deaths now exceed the excess death figure rather than being less than it. So for example in week 19, covid-19 reported deaths exceeded excess deaths by 796 (4,078 – 3,282)

  My very tentative broad explanation is that as the pandemic worsened, those with other severe conditions, typically those normally living in care homes, weren’t able to get treatment and died prematurely to some extent. In addition some people who had covid-19 but didn’t have covid-19 on their death certificate died prematurely. And this caused excess deaths to exceed covid-19 recorded deaths.

  As the pandemic passed it’s peak, some of those now dying with covid-19 are people who would have died anyway regardless of covid-19, and that some who died in earlier weeks of covid-19 would have died around now because of their other conditions. Hence why covid-19 recorded deaths now exceed excess deaths. The indirect deaths through missed cancer screening and economic affects etc haven’t yet materially occurred.

  Anybody got an alternative/better explanation?

  (entirely my own figures, and analysis so could have made an error. 5 year average deaths figure assumes a 1 week reporting delay, so week 19 occurring deaths compared with 5 year average of week 20 reported deaths, and adjustment made for deaths occurring in week 20 for deaths not reported by 23rd May)
• 67 Snowman May 26, 2020, 3:17 pm

  And just a mention that anyone who wants to see how far we have come from the peak of deaths on 8th April, nearly 7 weeks ago, and how things have improved in different settings hospitals, care homes and at home, just have a look at the graph in figure 9 of the ONS link in the previous post.

  Then factor in that deaths should be going down further from 15th May to today (26th May) and then factor in that deaths post-date infections by about a further 3 weeks.
• 68 MrOptimistic May 26, 2020, 9:22 pm

  @Newinvestor. True but gold has this last resort cache which is, to those susceptible to it, almost like another dimension of value.
  @Snowman. The FT analysis I posted was first off an attempt to quantify ‘how many would have died in the next year anyway. His answer was not more than 15%.
• 69 Vanguardfan May 26, 2020, 10:45 pm

  @mrO – thanks for the speigelhalter FT article, he’s always worth listening to.
• 70 Snowman May 27, 2020, 6:52 am

  @MrOptimistic

  Yes, thanks for the FT alphaville link interesting.

  His conclusion is that the % of those dying with covid-19 who would have died anyway this year is well below 25% and probably somewhere between 5-15%.

  Neil Ferguson claimed it was plausible that of the order of two-thirds of the people that have died so far might have died in this year anyhow. So I suspect two thirds is completely wrong as Ferguson has a particular knack for getting things wrong, and doesn’t even seem to ever get things right by chance!

  I sense that David Spiegelhalter’s figure is based on his feel for the numbers and he has a pretty good feel for the numbers of course. But I don’t think the source he quotes directly shows the estimate to be right, it just helps with the feel.

  If we do see a continuing trend for the covid-19 deaths to exceed the excess deaths, or even for the excess deaths themselves to turn negative later this year (before the indirect deaths such as cancers that have progressed from stage 1 due to delayed screening) then we do need to explain that.

  It can be explained without assuming that some of the covid-19 deaths are people who would have died anyway this year. For example if you imagine someone in the last few months of their life living in a care home, they may have been unable to access the hospital treatment they needed to prolong their life for that last few months. So they may have died in week 15 say rather than week 20. But that person may never have had the covid-19 virus. So it may reflect a limited partial unwinding of excess (genuinely) non coivid-19 deaths from the height of the pandemic.

  It is worth mentioning the David Spiegelhalter’s rule of thumb that covid-19 adds an extra year of risk on was calculated when the case fatality rate looked to be about 0.9%. So it’s going to be less than that now I think. And it is worth saying that extra risk is on top of normal risk but it isn’t completely additive.

  The ONS figures in coming months, especially when cause of death is analysed also, may make analysis easier.
• 71 Grumpy Old Paul May 27, 2020, 8:58 am

  @Vanguardfan,
  Spiegelhalter is giving a talk at the virtual Hay Festival today 6:30pm available from their website.
  I always find him worth listening to. I have a very low threshold for being irritated by obvious flaws in arguments or facts and he always manages to stay well below it!
• 72 The Investor May 27, 2020, 12:21 pm

  Preliminary research from France suggesting that Covid-19 infection almost always produces protective antibodies. The study (not yet peer reviewed) looked at 160 medical staff who had all tested positive for Coronavirus.

  It found:

  …all but one developed antibodies within 15 days after the start of infection, Institut Pasteur and university hospitals in Strasbourg said in an early version of their findings released before peer review.

  Almost all of the staff tested had antibodies that were capable of neutralizing the virus within 41 days of developing symptoms.

  https://www.bloombergquint.com/onweb/hospital-staff-with-covid-19-had-protective-antibodies-in-study

  Interestingly only 79% of the volunteers had neutralizing antibodies at 20 days, compared to 98% at 41 days. So there *is* a lag effect, albeit not of a game-changing magnitude.

  Great to see someone finally re-testing a population known to have previously had the virus. Very encouraging if this holds up in future studies.

  Edit: Missed the crucial word *neutralizing* in my first post of this comment! As I read it you see antibodies show up fast, but they take a little more time to reach a virus-defeating status.
• 73 Vanguardfan May 27, 2020, 12:29 pm

  It’s encouraging in the sense that it suggests mild infection (maybe also asymptomatic?? Dunno) does provoke an immune response.
  However that also suggests that the antibody prevalence studies aren’t missing some significant segment of the population that have somehow acquired invisible cell mediated immunity. (Notwithstanding the time lags of course). On that topic, I have been unable to find a publicly available report for the UK antibody study that Hancock gave some partial/garbled results for at the press conference about a week ago. Why not I wonder?
• 74 The Investor May 27, 2020, 1:06 pm

  @Vanguardfan — Agreed on the latter point, but *if* this study holds I’d say it’s more encouraging than that. People still routinely state “we don’t know if infection confers protection going forward“. According to this study, it does, at least for a period.
• 75 Jonathan May 27, 2020, 1:49 pm

  Yes, I saw that paper too. Very thorough. Only one of the 160 participants had no detected antibodies and s/he was sampled early (18 days after infection) when textbook responses take 3-4 weeks. And by 4-5 weeks almost all had some virus neutralising activity.

  They all showed symptoms of some sort to prompt testing, though for 5 these were described as “minor” which could be what someone would have thought asymptomatic were they not working in a medical environment and worried about infecting patients.

  Very encouraging for the vaccine researchers. I hope these individuals are followed over time so that information about how long antibody levels last, and whether they protect against further infection, eventually becomes available.
• 76 MrOptimistic May 27, 2020, 2:12 pm

  I’d have more confidence in my understanding of this is if all the immunology explained the asymptotic carriers. Can you carry the virus, amplify it through breeding, transmit it without your own immune system not provoking a symptom of licking the thing down?
  Last time I looked at the human immune system, and I didn’t get very far, I wondered why rocket science was considered hard. We’ll get to a better understanding of the origin of the universe first. The immune system is a true marvel and if I get religious and wonder about intelligent design I might just start with that.
• 77 Jonathan May 27, 2020, 2:38 pm

  @MrOptimistic, arguably immunology and theology are equally dense subjects!

  Basically, for an infection to “catch” you need a minimal number of infecting viruses (which as far as I know has not really been estimated yet for Covid). They infect a few initial cells in the nose/throat and replicate to generate a small population explosion of viruses to in turn infect nearby cells, passing steadily down the respiratory tract via a succession of affected cells. To pass between cells the viruses are exposed to the immune system, and will start the process of stimulating a response. For Covid, the way I rationalise it is that for patients with a quick response the infection stays in the nose and is shut down before it is large enough to cause significant symptoms. With poor responses the infection makes its way all the way to the lungs to cause the most serious disease. For the majority there is enough infection for fever or cough symptoms but not much more. In all cases, if the body has brought the infection to a finish the immune system will have been stimulated; you would expect that to mean detectable antibodies but could theoretically in a few cases only mean sensitised T cells (they cause a different type of immune defence, attacking the cells which have been invaded by virus rather than the virus itself when it moves between cells).

  That is a bit of an over-potted summary of immune responses to respiratory viruses, but it means that until full evidence emerges of Covid disease progression at the cellular level there is a strong expectation that any infection which the patient fights off – even if asymptomatic – will raise an immune response.
• 78 The Investor May 27, 2020, 3:41 pm

  @Jonathan — Interesting, thanks for those insights.

  On this:

  Basically, for an infection to “catch” you need a minimal number of infecting viruses (which as far as I know has not really been estimated yet for Covid).

  The piece that went viral from Dartmouth professor Erin Bromage the other week stated:

  In order to get infected you need to get exposed to an infectious dose of the virus; based on infectious dose studies with other coronaviruses, it appears that only small doses may be needed for infection to take hold. Some experts estimate that as few as 1000 SARS-CoV2 infectious viral particles are all that will be needed (ref 1, ref 2). Please note, this still needs to be determined experimentally, but we can use that number to demonstrate how infection can occur. Infection could occur, through 1000 infectious viral particles you receive in one breath or from one eye-rub, or 100 viral particles inhaled with each breath over 10 breaths, or 10 viral particles with 100 breaths. Each of these situations can lead to an infection.
• 79 Jonathan May 27, 2020, 5:39 pm

  I think Bromage’s explanations excellent. I don’t think he attaches any significance to the number 1000 which he has rather plucked from the air, it is just easier with an example number when explaining the balance between potential virus amount and exposure. So sitting across a table talking for an hour might give you the same total infective dose as someone cooughing in your face at close range in a crowded place.
• 80 Vanguardfan May 27, 2020, 10:36 pm

  @TI. The paper isn’t saying that there is definitely protective immunity. It’s just saying that all the people they tested had antibodies. Direct quote from the paper ‘although NOT YET DEMONSTRATED, several lines of evidence SUGGEST that the presence of neutralising antibodies MAY be associated with protective immunity…’

  So although people believe it’s likely that there is protection arising from these antibodies, we don’t know for definite. This was an in vitro study of the antibody activity, not a clinical study of reinfection.

  (For what it’s worth, I assume that the experts are probably right and there is some degree of immunity. But I also assume that the large majority of the population remains susceptible at this time).
• 81 Sparschwein May 27, 2020, 11:09 pm

  @Vanguardfan makes a very good point.
  Generally it’s important to read the original papers (or preprint manuscripts, as in this case) instead of relying on what a journalist thought they had hoped to be able to conclude…
  And read the papers in full, to get the context and the all-important caveats and limitations that any good scientific paper should mention.

  From the same manuscript:
  “For patients with SARS-CoV-1, antibodies persist for at least 2 years after symptomatic
  infection [16]. In the case of Middle East Respiratory Syndrome (MERS)-CoV, the antibody response is variable, not robust, and often undetectable when disease is mild [17–20]. Future studies will help evaluating the persistence of antibodies upon SARS-CoV-2 infection.”

  So, IF there is a protective effect from an infection (still TBC), we have *no idea* how long it might last. It could be very short-lived.
• 82 The Investor May 28, 2020, 12:37 am

  @Vanguardfan — Ah, important correction, thanks! (My fault for not reading the associated paper, just the article summary.)
• 83 Snowman May 28, 2020, 9:13 am

  @TI
  Re the study: Serologic responses to SARS-Co-2 infection among hospital staff with mild disease in eastern France

  You had me a bit worried that the study would give some evidence that everyone who had fought off the virus would have detectable antibodies. But having read the study it provides no evidence for that at all.

  I think all the study shows is that all (or almost all) of those who test positive for the virus through the PCR test and have classic major symptoms that don’t require hospitalisation, develop serologically detectable antibodies.

  Everyone in the study, apart from 5, had major symptoms. They give no detail of the just 5 people who didn’t have major symptoms. Perhaps they had minor symptoms such as a dry cough but didn’t think it was that bad, so decided to have a PCR test at the time because they were worried about infecting patients, and were then classified as having minor symptoms and not major symptoms for the purposes of the study.

  They should have called the study ‘Serologic responses to SARS-Co-2 infection among hospital staff with major symptoms of the disease not requiring hospitalisation in eastern France’. How can you have mild in the title when only 5 had mild symptoms and 155 had major symptoms? Perhaps some sort of bias showing up there?

  MrOptimistic is spot on that we need to study the asymptomatic.

  What would be interesting would be to test those who are working in high risk (for covid-19) healthcare settings where they had no detectable antibodies regardless of whether they had a positive PCR test but never seem to go down with symptoms. What is protecting this group? And also they should study those living in households where someone has had the virus, they share bathroom facilities, but have never had symptoms. Why did they not encounter the virus in a detectable way? What is protecting them? Similar conundrums such as this exist for influenza.

  So whether there are a number of people who have fought off the virus (through their innate immune system for example) but without detectable antibodies, and what immunity that group if it exists would have, remain completely open questions.
• 84 FI Warrior May 28, 2020, 9:16 am

  A minority report from an epidemiologist:

  https://www.belfasttelegraph.co.uk/news/uk/leading-scientist-urges-faster-exit-from-uks-lockdown-39225900.html

  Perhaps, as in finances, sometimes doing nothing is the best solution, maybe that is why our rulers aren’t personally worried for their health and their rules are to calm subjects.
• 85 Snowman May 28, 2020, 9:41 am

  Norway ‘could have controlled infection without lockdown’: health chief

  https://www.thelocal.no/20200522/norway-could-have-controlled-infection-without-lockdown-health-chief

  Don’t know the context, but on the surface it is refreshing that Norway’s Director of the Institute of Public Health is saying this.

  At the moment the UK is following the anti-science. Just say ‘lockdown worked’ enough times, rather than engage in genuine scientific enquiry. That’s group think not science. And let’s conveniently not discuss the evidence that infections were already reducing before lockdown in the UK and other countries. And so smart distancing may well have been all that was needed.
• 86 Jonathan May 28, 2020, 12:01 pm

  @Snowman, reading between the lines I suspect that quite a few of the patients in the French study had symptoms mild enough that they might have continued working if it had been a common cold. The study must have been initiated 2-3 months ago, when I suspect that France like the UK was finding testing capacity under heavy strain. Here most tests were directed at patients with symptoms severe enough to require hospital attention. But there were exceptions for testing health workers who they wanted at work as long as there was no risk of them passing on infection – that seems to have been their source of those with mild symptoms, as well as their choice of the word “mild”. The five with “no major symptoms” might have been tested because of family members having clear symptoms.

  The manuscript is quite clear about the scope and limitations of the study. Obviously we look forward to more comprehensive studies in future, but to be honest no one seems yet to have published a good survey showing the proportions of those infected with different levels of symptoms (including death).
• 87 Snowman May 28, 2020, 2:46 pm

  @Vanguard Fan
  ‘On that topic, I have been unable to find a publicly available report for the UK antibody study that Hancock gave some partial/garbled results for at the press conference about a week ago. Why not I wonder?’

  Released today, section 4 gives the percentage testing positive for antibodies in England between 26th April and 24th May as 6.78% overall (with a 95% confidence level of 5.21% to 8.64%).

  https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/28may2020

  Can’t find a regional breakdown.
• 88 Snowman May 28, 2020, 3:04 pm

  Here’s a regional breakdown of sorts of sero-prevalence in the Public Health England Weekly Coronavirus Disease 2019 Surveillance Report released today. Based on samples from blood donors

  Relevant bit is page 22 – 23.

  https://www.gov.uk/government/publications/national-covid-19-surveillance-reports
• 89 Vanguardfan May 28, 2020, 3:46 pm

  Thanks for those links. So the seroprevalence study is a subset of the active infection study. On less than 900 samples, less than 60 positives, regional breakdown would be pretty imprecise.
  Back of the envelope calculations (excluding care home deaths, since care homes residents are excluded from the seroprevalence study) suggest infection fatality rate in England of about 0.7%, which seems pretty much as expected. If care homes were included it would likely be higher.
  None of this information so far is challenging the consensus about where we are in our epidemic. Still nearer the beginning than the end.
• 90 ZXSpectrum48k May 28, 2020, 7:05 pm

  @Vanguardfan. Have to agree. As you say this does not challenge the consensus and, if anything, might be at the higher end of consensus. Once care homes are accounted for, IFR could easily be getting toward 1%. That would be broadly compatible with the Spanish seroprevalence study that came back with an IFR of 1.1-1.2%. If herd immunity is at the 60-70% level then total deaths (without medical intervention) might well have been be in the 350-400k region. Unfortunately, Ferguson’s “shock” fatality numbers may well have been about right. Not good at all.
• 91 The Investor May 28, 2020, 8:55 pm

  @ZX48k @VF @Snowman — Another disappointing suggestion (to my eyes anyway) in that surveillance report is that it says antibody prevalence is in general higher in younger populations, 17-29 year olds.

  This would imply that older populations have in general been less exposed to the disease than younger populations, and thus we’ve seen fewer deaths than we might have, which could I presume have implications for the IFR. I suppose it also reinforces the case for shielding the elderly et cetera, too, but yes, as I’ve said for weeks now this antibody data has been a big disappointment to me.

  I didn’t see much in the French study to indicate known defeated infections not producing antibodies, either, although I suppose it hasn’t categorically disproved it, given the selection bias likely in that population group. But I wouldn’t be thinking in that direction unless I wanted it to be true, anyway, and don’t hold out much hope.
• 92 Jonathan May 28, 2020, 9:58 pm

  @TI, I share your disappointment with today’s surveillance data. But actual data is more important than my instinct* – that IFR will turn out to be significantly less than 1% once the dust settles, due to all test data being biased in favour of those with suggestive symptoms.

  We will see eventually I suppose.

  [*unlike Dominic Cummings’s instincts being more important than national social distancing rules].
• 93 Seeking Fire May 28, 2020, 10:31 pm

  I must admit that when I read the ONS study, one of my first ports of call was this site for more info. A credit to the people tapping away. Am I right in saying this is first antibody test and its of just 900 people? This doesn’t feel like that govt has done a very good job here. Anyway, I agree the results are disappointing. The results doesn’t effect behavioural planning for me or family as I continue to think we’re low risk based on ONS data but I continue to encourage the elderly in-laws to stay at home. I agree it would seem as if we’re closer to the beginning than the end based on this study. Think govt messaging continues to be inconsistent. Oh and don’t sell.
• 94 ZXSpectrum48k May 28, 2020, 10:46 pm

  I’m also not particularly liking Figure 22 on page 17 of: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/888254/COVID19_Epidemiological_Summary_w22_Final.pdf

  They say “For the most recent week, more samples are expected therefore the increase seen in the above graphs should be in-terpreted with caution.” Yet it only seems to be London that is rising sharply?
• 95 Vanguardfan May 28, 2020, 10:59 pm

  @jonathan, the seroconversion study is a population based study, it should be reasonably representative, although there is of course always some response bias in every survey. Remembering also that it excludes residents of care homes. I’m not seeing how the IFR could come in lower later? Unless of course treatment gets better. I’m not sure the pattern of community infection suggests higher rates in vulnerable populations to date.

  I actually think the national rolling survey of infection seems well designed. It’s 10,000ish households per weekly wave, testing for active infection, with a subset doing the antibody test in addition. This will build up numbers quite quickly. I thought the most interesting result was the higher rate of infection in those working outside the home. This does suggest that there is a risk of rising infection rates as lockdown measures are eased, with increasing numbers back at work and in shops (not monevator readers surely!)
• 96 FI Warrior May 29, 2020, 6:59 am

  Straight outta Italy. Interesting article on why Lombardy is apparently worst affected by the Corona situation, diminishing the contribution of pollution, instead suggesting the particular nature of localised healthcare privatisation was the main culprit, with population density and an incompetent local administration contributary factors.

  https://www.theguardian.com/world/2020/may/29/why-was-lombardy-hit-harder-covid-19-than-italys-other-regions

  If that is confirmed to be true, it’s no wonder the UK’s record is turning out at least as bad, given those afflictions are institutionally embedded, systemic, signature traits here.
• 97 Vanguardfan May 29, 2020, 7:39 am

  @fiwarrier, I was hoping for more from that article. Very light on facts/research.
  I remember early on in Italy’s outbreak, that a lot of transmission seemed to have occurred in hospitals (I think one was a cancer hospital), due to delays in recognising and testing cases. I read a report that there was an investigation into these failings but I haven’t read anything about what it found. That said, I agree that dealing with covid is more about having good joined up public health infrastructure and less about state of the art healthcare for the sick.

  Once the virus starts circulating widely in settings like hospitals and care homes it is extremely difficult to contain, and these places act as reservoirs. Similarly some workplaces can also become hotspots. Until we actually start taking outbreak control seriously, with short local communication chains feeding test results directly to contact tracers/public health outbreak teams, we are never going to get on top of this. It’s negligent that even in hospitals they aren’t testing everyone weekly – again, we are never going to get safe non covid healthcare back up and running without this.
• 98 FI Warrior May 29, 2020, 8:36 am

  @Vanguardfan, I remember working at a London hospital a long time ago when MRSA was terrifying, with people justifiably scared to go in for ops because they thought they risked dying of infection instead. It seemed lunacy that something as basic as the neglect of systems like cleaning procedures were killing people in a nominally advanced country, but creeping privatisation and the curse of over-management culture was already entrenched in the NHS. ‘Low value’ staff like cleaners, porters and technicians checking slides all day while on minimum wages and more worried about making the rent are not the best motivated to concentrate and when these errors cost lives immediately, tragedy is guaranteed. Not enough people cared back then and so the mission creep continued, inevitably then, today even nurses, the backbone of the service end up on the list of demoralised staff, topdown decisions are always political and micromanagement in a blame-culture the norm. We’re collectively just reaping the fruits of our past complacency.
• 99 Snowman May 29, 2020, 7:55 pm

  @TI

  I’m pretty relaxed about the UK serology sero-prevalence %s. I think they show there is a good level of immunity built up in the English population especially in London.

  It still seems very likely that there is a lot of immunity other than that picked up through detectable antibodies. Here’s some more stuff pointing that way

  https://twitter.com/BallouxFrancois/status/1266265084143087628

  That we need 70-80% of the population to have detectable antibodies to have herd immunity is looking quite wrong now and that needs re-evaluating, the SIR and SEIR models that put the herd immunity threshold or herd immunity level at those levels don’t really seem to be good models. I suspect London has something approaching enough immunity to stop any further spread at the current time allowing for the undetected immunity.

  Karol Sikora is increasingly looking to be right that the virus is petering out in the UK, and I think it is petering out because of the seasonal affect not just because populations have been socially distancing (lockdowns add nothing or next to nothing).

  There is a possibility there will be some small outbreaks locally and it is possible that it will come back in Winter to some extent as per the OEBM wave link, and perhaps it will be worst in the areas of the UK currently with the lowest antibody %s currently, but impossible to know.

  In truth nobody knows what is going to happen, and we have everything to learn still, but my hunch, unlike some others, and ignoring what may or may not happen next Winter is we are at the beginning of the end.
• 100 Snowman May 29, 2020, 8:12 pm

  There is another graph of regional sero-prevalence (based on blood donors) here on this gov.uk page

  https://www.gov.uk/government/publications/national-covid-19-surveillance-reports/sero-surveillance-of-covid-19

  However the graph seems to show different %s than on page 22 of the National Covid-19 surveillance report for week 22 linked to earlier.

  Can’t immediately work out why the %s are so different
• 101 Sparschwein May 29, 2020, 9:48 pm

  @Snowman – Balloux never said what you imply. You are referencing him incorrectly.
  In fact, in another thread on 27 May he said: “Cross-immunisation is a ‘wilder’ hypothesis…”

  That’s what assuming some other form of immunity is, at this point. A *wilder hypothesis*. Everyone should know this, be responsible and act accordingly.
  Protect others from an infection that could kill them.
  Already now the UK has a terrible Covid-19 death toll, one of the worst in the world.

  Sikora is actually a cancer doctor. He is *not* an expert in any of the relevant disciplines: virology, immunology, epidemiology, biostatistics…
  Sikora is not an expert in this pandemic at all.

  Sikora does however have a huge conflict of interest, that should always be mentioned when quoting his personal opinion on Covid-19.
  Sikora is co-founder and owns a multi-million stake in Rutherford Health (name changed recently from Proton Partners), a chain of private cancer centres that got embroiled in the Woodford mess. They made a huge operating loss already before Covid-19, and surely they are struggling even more now.
• 102 Snowman May 30, 2020, 8:18 am

  @Sparschwein

  I merely linked to the Balloux thread as it contained a paper that further discussed some of the possible mechanisms for non-antibody immunity. The point is the experience and epidemic curves in different countries suggest something like this is going on, and the twitter thread just indicates that there could be mechanisms that do explain it, wild as any individual theory may or may not be at the moment. Better to remain open minded while the evidence is unclear rather than assume all who encounter the virus produce detectable antibodies. Form a view, but be prepared to change that view. That’s where I’m at, and I respect the opinion of those who have coherently posted an alternate view above.

  If you want to be constructive then please link to a single paper that has studied specifically the asymptomatic or those in households where somebody else has had covid-19 but had no symptoms themselves and shows they still have detectable antibody protection? I am genuinely interested if you can provide study a study.

  I agree conflicts of interests are always important, but for balance Professor Sikora has stated his motives as

  ‘The whole reason I joined this debate was to stand up for the forgotten cancer patients. We’ve made some progress, but it isn’t moving quickly enough. Despite what we read, Coronavirus is not the only thing killing people.’

  Sometimes it is experts outside the direct field (but still medically eminent in this case) who can look at things with an open mind and avoid the group think that partly derives from conflicts of interest amongst other things. Yes maybe he has a bias towards the cancer patients, but it’s good that someone is able to make a minor dent in the bias in the other direction.

  The immunologists and virologists don’t generally have any expertise in cancer. So presumably they should give no opinions on how to control the epidemic because they can’t judge the indirect deaths side of things?

  Your use of emotive language isn’t helpful, and when I see that it rings an alarm bell that someone isn’t thinking clearly, it’s understandable, but dangerous. What about all those people who will die of cancer because they aren’t diagnosed, what about all those who die through the indirect results of all this? The longer this goes on the more who will be affected.

  You say ‘already now the UK has a terrible Covid-19 death toll, one of the worst in the world’. Do you judge success (for want of a better word) as how many have died so far? Do you think we should take into account those who die in the future from covid-19 and the indirect deaths in judging success, and take into account the average years lost for each of those deaths? And should we look at excess death figures or covid-19 recorded deaths only? Do you think a country that didn’t take action before covid-19 had widely spread through the population, could stop the virus entering the population indefinitely or until a safe scalable vaccine is found, because that is the only scenario I see for judging things based on deaths to date. All deaths are sad of course past, future, covid-19 or non covid-19.
• 103 Minikins May 31, 2020, 11:06 pm

  The big data on testing, symptoms and outcomes will come with the test and trace programme which they will hold for 20 years