Hello everyone and welcome, for those betting I would use the word penultimate in the first sentence – you win! 6 sleeps till xmas, 7 till the last supplement of the year and 13 until 2022!! Perhaps unsurprisingly, this week, the draft plans for the supplement got ripped up halfway through the week as the macro situation, and the covid situation, continue to evolve.
One more moan, first. There is just SO much noise. Once again the hogwash posted, on all sides of the debate, has gone into the stratosphere because there is uncertainty. Some is fear, some is poor mathematics, some is deliberately to mislead people and cause disruption, some is trolling, some is those who believe they should get their point across at all costs. I’m going to try and inject (pun intended) some reason into the whole thing, one more time.
We’ve also had a big week on the macro front, interest rates and inflation have taken the headlines but there are further macro measures to discuss which will influence 2022 strongly. What are we seeing, at a high level, at the moment – and what have we seen throughout 2021?
Things moving much more quickly than the modellers are used to
Upside and downside surprise
Some fundamentals ignored, not just for political reasons but for capacity reasons
I’m going to get stuck into the macro first. You will have seen the headlines this week. Inflation to 5.1%, surprising few people. The forecast was 4.7% but overshooting was no surprise given that comparator countries have also been overshooting. The psychology of the 5%, as I alluded to last week, was breached and the Bank of England’s Monetary Policy Committee decided that was enough to raise rates in a December meeting for the second time in 45 years.
We need to consider what the reasons might be. That’s 2 MPC meetings in a row where the base rate vote has gone in the diametrically opposite direction to the market’s predictions. Those are the moves that have the most of the desired effects, and also (on the back of it) serve to change expectations. Models are reworked. This is extremely effective, and this is quality central banking in my view, as a general comment.
The BoE also needs to ensure that the markets believe they are credible custodians of the interest rate situation and the keys to the printing machines. This will have served well to do that. So rates were hiked (I made that point in the week – “hiked” to 0.25%. 15 basis points. Wow). This strengthened sterling particularly against the dollar and set the policy more clearly for 2022. The current intent will be 2-3 rate rises, I’m sure (that’s what they would say anyway) – but, hold on a minute, don’t we need more scepticism?
Well – the emerging Omicron situation is very close to telling us an awful lot more within the next week. And what a week to be getting that data in with Christmas just 6 days away, after last year. One thing covid has proven – Murphy’s law is alive and well.
Hospitality is in tatters, with all indoor venues suffering to some extent or another. Measures are surely coming in support – despite the agenda of the “Covid Recovery Group” side of the ruling party who currently have the whip hand after a 100 MP rebellion this week on the vote on “plan B”. Now we’ve got SAGE hedging their bets saying we need more measures soon, and we are just where we were before every other lockdown so far – resisted to the point that that lockdown then needs to be far longer before being meaningful, a proper “cut nose off to spite face” play. But – is this time different? It could be, and more on that later on.
Sticking with the macro, however, inflation up (and now the BoE instead of forecasting 5% in spring, having already passed that, are saying 6%). I would bet on the upside of that 6% at this time – looking at the US, and looking at the guaranteed components coming into play of more energy price rises – the bank have factored this in, but are too invested in talking inflation down at this point. I’m thinking January 2022 will see some price rises in the region of around 9% on many goods – rises held off in 2021 but margins are under too much pressure. The contrarian I am, I am also seeing the next reasons to be optimistic right at this point, as we enter just one more covid nadir. So here’s what hasn’t made many, or any, headlines this week. Channeling the inner Ian Dury if you will. We may have to get a bit worse and the forecasts to get a bit more miserable before we see forecasts we can overshoot in a positive way, though.
Yes, there are those individual specific sectors in trouble once again. Travel. Hospitality. Some leisure. Retail I have seen some claims around a December only 60% as good as forecasts (we will see very soon). GDP will move downwards on the back of this as a slice of the poor old British people try to semi-lock themselves down in the absence of Government guidance. This has been an argument I’ve referred to all along – the Belarus argument, just look back at what the people actually did when their “great leader” denied the existence of covid and claimed it could be warded off with Vodka. Or Sweden and how the Swedes managed themselves outside of a relatively loose set of restrictions and guidelines. We’ve lost 0.5% of GDP on the BoE forecast since the November MPC meeting just 6 weeks before.
We also had some unemployment figures this week. There has been a big climb in employment on payrolls (be cautious – there are multiple impacts here, temporary xmas jobs, IR35 still having an effect I’m sure, etc. etc.) So – we’ve gained 0.5% in employment to go alongside that same drop in the GDP forecast, just in 6 weeks, as the data now looks overwhelming that the end of furlough made almost no difference at all. This is a good sign – put that alongside the 1.2 million job vacancies, and the labour market looks very tight in 2022.
This is a worldwide phenomenon – although we have a specific Brexit impact to contend with as well – and is one part of the argument against transitory inflation. The “great resignation” has continued – however, as inflation ravages savings, pensions, benefits and the like, there is likelihood that financial pressures will force a slice of the resignors back into work in 2022 at some point. Still, low unemployment is very rarely regarded as a bad thing, however you look at it.
Meanwhile, of course, the Treasury is delighted at what looks like a relatively guaranteed inflation rate at or above 3.5% next year, on average. Bad for pension triple locks of course, but great for the debt mountain. My feeling is that inflation won’t ebb away as quickly as the bank are expecting, forecasting, or hoping for (you decide which one of those applies!)
What is worth pointing out here, specifically to debt-based/sensible leverage advocates – is that even with the base rate up, the 10-year bond yields have been down. This is often more relevant to the setting of the mortgage rates than the base rate itself (not helpful to you if you are on base rate tracker mortgages of course). This does for the moment though imply that fixed rate mortgages for limited company investors, over 5 years, may stick around around 3% for a little while longer. The direction in the past 3 months has been downwards, and 5 years sit at 0.6% while 10 years sit at 0.75% as I write this. Cheap, cheap money. So when you hear “rates are going up” or even the “hike” word, the market isn’t buying into rates doing an awful lot.
I still remember the BoE under Carney and his different style including forward guidance mooting that 2.5% would be the “new normal” base rate, and the US Federal Reserve are actually quite clear on this – 2.5% – 3% is their desired “new normal”. They did get there for about 5 minutes under Trump but had to unwind that position very quickly, even before the pandemic. I still feel 2.5% is the number they are working towards in the long run – and remember, these are central bankers. If they can put off a decision, they will. They think in decades, and that’s a good thing for the running of the BoE.
The other news was a weak GDP figure of only 0.1% in October and the resultant fact that we were sitting just over 2% below the pre-pandemic economy.
So…..unemployment looks good and is likely to get back under that 4% level relatively quickly. Productivity – we still need more data and there’s such a lag – have to wait until mid-Jan 2022 for that one. The summary so far is more productivity per hour worked, but less productivity per worker (i.e. people are still working shorter hours than pre-pandemic) – but only in the small percentage levels, nothing major. There’s still caution over how you measure it, and what it means, especially in the backdrop of a pandemic/furlough/etc etc.
We will see a GDP lag as we get through what is certain to be a tough January, there are no two ways about it. The way things have been, the GDP forecasts will be revised downwards too far and we will then overshoot as we follow an even more cyclical, seasonal overshoot/undershoot pattern on predictions. The covid “noise” has just not been captured accurately by the Bank of England modelling.
Now to segue back towards covid – as it still grips so much of our decision making and day-to-day lives. I heard a great snippet from Chris Whitty on the radio this week that I thought was worth sharing. Professor W has been one of my favourite things about the pandemic since he’s been fair and balanced throughout, in my view. Despite slanderous claims about him by some extremists, he has little or nothing in the way of conflicts of interest either.
He’s been talking about the future and where we might be in 18 months plus, answering the question around “constant jabs”. Of course we know Pfizer would like to recommend a jab every morning, if they could, in the “Best interests of the Company”. There is understandable disquiet or, simply, further hesitancy from those who bought the “double-jabbed is doing your bit” – such as the ridiculous claims from Biden around it being a “patriotic duty” this week in the US. Let’s have a quick look at the numbers:
89.4% of the 12+ population have had one jab. 81.6% have had a second. On inspection of the last month, around 25k first jabs are still being issued each day. Around 30k of second jabs are being issued. So, that’s 20% more people (roughly) getting a second jab than a first. That – to me – is suggesting some second jab hesitancy – and that’s quite a gap between first and second. This is and will be a phenomenon – so there will be third jab dropoff too. At this time we are targeting 1m boosters per day, as you will know, and have broken the record on Friday for the most jabs issued in the UK so far, in the past week (not quite there, but 930k shots total). Between first and second jabs we would take 1000 days at the moment to catch up (very roughly) so that everyone had been double jabbed – some people have fallen off the process…..
So, Prof W is answering that in a roundabout way. Will some be hesitating on their booster thinking “are we just going to get constantly jabbed”. Chris doesn’t think so – he thinks we will be a long way further forward in 18 months with retrovirals being much more widely available from more companies, and further vaccine research and market entrants which will break some of the oligopolistic dominance (dare I use the word cartel) of Pfizer et. al.
He also referred to the ongoing emergence of our overall immune response. Ultimately, there are two ways in which we have been creating resistance. Vaccine, and recovery. Back in June the ONS had identified over 91% of the population with antibodies. This did not equal herd immunity (as evidenced now) but does give you an idea of headroom. The survivors get tougher as time goes on (basics of evolution, right?)
These are now where we get to the interesting numbers. There have been some shocking and frankly utterly irresponsible figures thrown around this week. A million cases a day! This shows no understanding of mathematics, populations, modelling or any other points that should be key to making a proper argument and forecast. If we are preparing for the “best worst case scenario” – which we should be – we should be clear about that.
I have identified at least 3 points in the past 6 months where there have been claims of over 100,000 cases a day “by the end of the month” or similar. There may be more. We never got there – in fact, we only just about reached halfway there. Why? One is variance. This week we saw a huge rise in the number of cases detected in one day – around 40% of a jump. We broke the record for case numbers 3 days in a row. Of course this is concerning. Does it justify a prediction of a million cases a day though?
Now – for one more of those maths lessons you don’t really want. I will try to keep it simple. But here’s the problem. It isn’t simple. It is multivariate. There are tireless efforts by many on both sides of the debate to keep it simple, but that destroys the nuance, and opens the door to either oversimplification which can be characterised as “lies with the right outcome”, OR bastardisation which can be characterised as “proof that the government is lying” or similar.
So – onto our multivariate problem and a revision of what the variables are. The government actually do help here because they track all of them and publish them on a transparent basis. As so often, this is far, far better than many countries in the world, but we get lost in our own little bubble argument and often forget this. Don’t worry – the government also does plenty wrong in how they contribute to the debate, and how they communicate, and how they make decisions – so they are going to get plenty of pasting too…..
The “variables of concern” (OK, couldn’t resist that one) are:
Daily covid cases
Number of hospitalisations (note – DUE to covid, not “with covid”)
Critical care bed capacity (number in ICU)
Infection fatality rate and Case fatality rate
% of adult population with antibodies
% of population having had first jab
% of population having had second jab
% of population having had booster (sticking with the official terminology)
As soon as you see that list, then you realise (I would think) that you don’t know what all of those things are. You may well know the cases – it is difficult to avoid the number. And then in many ways the more of this information you have, the more problems it introduces to solve. But we can characterise the bulk of the problems thus:
The entire situation is about the NHS, protecting it, and managing the pandemic on top of an already stressed public health system – the goal is to not have wasted lives that could have been saved by having better capacity (this is another way of saying – there is an acceptable level of death that we all went about our lives “knowing” before the pandemic and we need to get back there, ASAP)
The new variant is effectively known to be much more transmissible and less deadly than previous variants. The problem is how much more, and how much less.
Delta crowded out other forms of covid, particularly in the UK and a handful of other countries, and Omicron is likely to crowd out Delta (the estimated transmissibility makes this a near certainty)
Regardless of government intervention and support, people are now at a stage where they will take action as they see fit which will have economic, social and societal consequences. Lack of leadership is harmful here.
People have, as a whole, absorbed many messages of fear in the past 22 months, including continual bombardment via public signage and roadside campaigns, signs everywhere in shops and other retail premises, and this is not an “on-and-off like a tap” situation – there will be shadows
Covid does not give us a reason not to consider all other causes of death, health problems, unhappiness and general participation rates in society. There is a long shadow coming (which has no doubt already started) and over-prioritisation can and will lead to excess morbidity from other causes such as cancer, particularly if the first line of defence (GPs) are stopped from seeing the general public on other matters, and/or if the general public are too scared to use the first line of defence.
So onto the mathematical part. IF (and if there was a way I could use 55 font-sized capital letters on that first IF, I would do it) Omicron is proportionately less deadly vs. Delta and proportionately more transmissible, the fatality rate drops dramatically (there’s no guarantee of this of course). But wait. 95.3% (at last estimate, w/c 15th November 2021) of the adult population in England has antibodies at this time.
This doesn’t make them immune to Omicron. You can definitely catch this twice (because it isn’t the same thing you are catching, apart from anything else). However, your immune response is firstly likely far better if you’ve caught it before, or if you have the antibodies thanks to the vaccine. The more vaccines you’ve had, the better off you will also be. There is some evidence that antibodies are “better”/your immune defence is stronger from catching it than getting the vaccine – although using this as a reason not to get the vaccine is badly misguided (it’s a bit like playing russian roulette, with a low probability of blowing your head off, rather than playing with a much bigger number of empty chambers, with a tiny probability of blowing your head off).
This of course also affects the fatality rates. Release Delta on day 1 instead of Alpha, and it is likely many more people would have died, especially if we’d reacted in basically the same way as we did to Alpha in the UK. But at some point, when you water it down enough, the problem becomes far less in terms of our point 1) – the protection of the NHS and the critical care resources.
Then we move back to the cases. Here are the actual pertinent stats at this point:
Case numbers: 90k yesterday, 513.5k in 7 days
Adults with antibodies: (working on c. 56m adults in the UK, 95% throughout UK) 53.2m
Adults without antibodies 2.8m – potentially “low-hanging fruit” for the virus, (not the only ones at risk, but perhaps fair to say the ones at the biggest risk)
R rate – Currently published at 1.0 to 1.2, that’s from Friday, showing what a job they are doing of staying on top of things…..”real R” looks more like 2.0 to 2.2 right now, but that’s subject to the rest of the stats too…..
Number of hospitalisations – this started to dip under 1000 on about 8th November based on the 7-day average, this is a laggy piece of data so last efforts are from 13th December (Monday) – but the 7-day was under 900 on 13th December. It will need to shoot up in a similar way to the case rate to get to the 3000 now being publicised (and see today’s graphic, but also the critique)
Critical care bed – capacity – we got to 4000 in January 21, about a month after the last minute lockdown at/around Xmas 2020 (depending on where you lived in the country, apart from other things!). 3000 is being mooted as breaking the NHS in the media at this point – not sure why when the facts are we coped with 4000 without Nightingale capacity. We are currently at c. 900, and again the recent peak was around 8-November. The peak will come around a month after the peak case numbers – this is a critical point that much analysis is missing, for no good reason.
Fatality rates – lots and lots and lots published about these in the past 22 months! Hard to measure, for lots of reasons. The government don’t bother to try. You can argue an awful lot either way – but one thing you can say is that the current fatality rates are far lower than March/April 2020 and also January 2021, orders of magnitude lower. That’s a good thing. How low they are is traded off against how many people “have it” of course, or, I would argue, “have it” and are unvaccinated, especially if they have no previous antibodies. That starts to drill down to a much smaller subset, which you could call “live cases” with a much bigger percentage chance of a bad outcome.
The other percentages we have written – the booster % doesn’t tell us much at the moment.
One of the big omissions (it surprises me, because this is an actual number that means something) from the narrative of late has been the fact that double-jabbed Astrazeneca (vast majority of people over 40) has 0% effectiveness on transmission of Omicron. This means there’s a material probability based on confidence intervals that you are actually MORE likely to transmit Omicron as a double-jabbed AZer. Let’s be very careful here. What do we care about?
Probability of catching it
Probability of transmitting it
Probability of going to hospital from it
Probability of dying from it
Probability of having a long-term bad outcome from it.
I’d suggest the sensible order of play as an individual would be to consider these risks in this order: 4-3-5-1-2. We have to rely on individuals to be individuals and put themselves first. Death is the worst outcome. Hospital massively increases the chance of death (not because the hospital will kill you, on the basis that if you go to hospital you are much more likely to already be in very bad shape). No-one wants long term bad outcomes, but might be grateful for being alive. Not catching it in the first place is the best place to start. Not transmitting it would be helpful, a nice-to-have but not driving much behaviour, in my view.
What are we getting our attention focused on, gigantically, via the mainstream media machines? Catching it and transmitting it. There’s been so much written about transmission and vaccines – without much in terms of clear and robust evidence that I’ve seen, in meta-studies – and one variant comes along anyway and takes all that out.
What if you’ve had a double-jab of AZ already? The dominant strategy is to get your booster.
What if you’ve not been vaccinated, don’t know who to trust, and don’t know who is telling the truth? Well, I would normally say “go and look at the data yourself”. That’s the only way. What’s the point in me, or anyone, telling you what to do? I’ve written before about legitimate vaccine hesitancy and am seeing the same attitude at the moment that lost labour an election by a landslide, that meant the referendum went the way of brexit, and a whole number of other things. A place between intellectual superiority/snobbishness and arrogance. So frustrating!
The reason why I say “I would normally say” is because the ONS disappointed me deeply in a study they released on November 1st. They NEED to right this wrong. They don’t do conclusions, generally, but made an exception here – perhaps under political pressure.
They compare Age-standardized-mortality-rates (ASMRs) – and stay with me here. All it means is that they take out the age component of mortality, and just give you one number of the likelihood of dying. This isn’t that helpful if your vaccine hesitancy is based on your age and fitness level (it really shouldn’t be, but it is natural). But, mathematically, it is helpful to be able to standardize numbers, otherwise it is far more difficult.
They proclaimed from the piece of work that you were 32 times as likely to die from covid if unvaccinated versus vaccinated. Their numbers were badly polluted by starting from January 2021 (when basically no-one was vaccinated). The numbers needed to be “cleaned” by at least giving the majority of the population the chance to be vaccinated, and this conclusion was very simple to debunk.
I reworked their numbers, based on starting from August 1st 2021 (I would like to have started later, but all over-18s had been offered a first jab for at least a month and their data only goes to September, so its a trade-off – and like a lot of “covid math” – it is imperfect). The actual conclusions from there:
ASMR unvaccinated versus double-jabbed – 6.87 times higher
Chance of death this year from covid, if unvaccinated, no age specified – 0.007% or 1 in 14,545
Chance of death this year from covid, if double-jabbed, no age specified – 0.001% or 1 in 100,000
So – 7 times more likely to die. Very unlikely to die – but uncomfortable, especially if you start to introduce age and comorbidities into the equation. I’d encourage anyone who hasn’t used it – IF the data will help you make a decision – to use the excellent calculator at Q covid dot org. I ran it for myself, and ran it twice. Firstly, if I was unvaccinated – secondly, if I was double-jabbed (yes, this analysis is taking place while “vaccinated” still means double-jabbed, even though at the end of January or thereabouts to be “fully vaccinated” will mean you need to have your booster – but as you can see from the above, the numbers are clear that the dominant strategy is to have your booster when you’ve already had 2 x vaccines, particularly if they were astrazeneca – mine were, and I’ve boosted, for the avoidance of doubt!)
My QCovid stats:
Risk of catching covid and going to hospital: 1 in 455
Risk of hospital admission following a positive test result: 1 in 13
Risk of catching and dying with Covid-19: 1 in 14,493 (very similar indeed to my analysis above)
Risk of dying from Covid-19 following a positive test result: 1 in 149
Now – this is where some numbers aren’t that helpful. The two pertinent ones – today – if unvaccinated would be the 1 in 455, and the 1 in 14,493. But look at the numbers if unvaccinated me gets it – 1 in 13 chance of going to hospital? Not sure about the readers, but for me that’s uncomfortably high. And whilst 1 in 149 is still low risk in my eyes, for my life it isn’t a bet that I would like to lay. Bear in mind in that 1 in 13 situation it is too late for me to be vaccinated, and that’s why I say that number isn’t helpful – but it is extremely powerful.
So – what about if vaccinated:
Risk of catching covid and going to hospital: 1 in 8,065 (around 18 times less likely)
Risk of hospital admission following a positive test result: 1 in 244
Risk of catching and dying with Covid-19: 1 in 100,000 (very similar indeed to my analysis above) – so I am at an average risk of death compared to the population as a whole from covid.
Risk of dying from Covid-19 following a positive test result: 1 in 3,774
On a numbers basis, I’m pretty happy right now I’ve had the vaccine, having read all of that. That battle between “how likely is it” and “what’s the outcome if it happens” is similar to fire insurance on a property – or even life insurance. “But it’s so unlikely to happen” – but what if it does……
Am I telling you to get vaccinated, or to stop getting vaccinated, or what to do? Absolutely not. The idea is simple empowerment – here is the information. I am aware of someone who has a profile within property who has been in intensive care recently, unvaccinated, and rolled those dice above, unnecessarily. They’ve survived – they’ve also used hundreds of thousands (if not millions) of pounds worth of scarce healthcare resource. They might have needed to, anyway, based on the above (there’s a chance). But it is 7 times more likely that they wouldn’t have done if they’d been vaccinated (or thereabouts). I’m still seeing loads of posts from people who haven’t yet pitched a camp, and taken a “side”. This is just the way I go about solving these sorts of problems, just like any other problem where there are numbers available.
I ran a similar exercise when my wife was pregnant with our first child and we decided whether to have amniocentesis or not. That was 11 years ago and the answer was an easy, and overwhelming, no. At the time that decision was an absolute no-brainer on the numbers (and no-one helped with that, which really narked me at the time) – perhaps things have moved on in that department now, I’m sure they have, but it leads to my strongest conclusion so far from the pandemic.
Doctors and numbers don’t mix all that well. Politicians and numbers don’t mix very well at all. Numbers have a gigantic role to play in all of this – not just for those like me who are blessed by swallowing the calculator at birth, or whatever. To make sensible decisions. Why haven’t we wheeled out the National Statistician every night, or at least once a week – well, google Sir Ian Diamond and you might find out! But that whole story, in itself, shows the lack of respect that the one discipline that could have added value so far – mathematics – has been afforded throughout the process. Instead we’ve gone with Neil Ferguson. Not ideal.
This week has been huge – well done if you’ve got to the end. I hope it was worth it. For me – I needed to get it off my 2021 chest, so there it is. If I’ve upset you – good. Your thinking likely needs challenging. So does mine, so please comment, like and share. Thank you!