We have the opportunity to look at what is good news and to see that many of our themes have some true again.
UK gross domestic product (GDP) is estimated to have fallen by 2.9% in January 2021, as government restrictions reduced economic activity…………January’s GDP was 9.0% below the levels seen in February 2020, compared with 4.0% below October 2020 (the initial recovery peak).
Care is needed as of course this is still an economic depression with an 9% decline since pre pandemic. However the lockdown this time around has had a much smaller impact than last time meaning there is more hope for the recovery.In terms of our themes then we see that the “expert” forecasters were wrong again as they were expecting something like -5% on a monthly basis and yes the first rule of OBR Club ( that it is always wrong) has had yet another good day.
Should February and March follow a similar pattern then there is genuine hope that the Bank of England is wrong too.
GDP is expected to fall by around 4% in 2021 Q1, in contrast to expectations of a rise in the November Report.
This leads to this according to Alpesh Paleja of the CBI.
Worth noting that even if GDP is flat in Feb and March, the decline over the quarter will “only” be 1.3% – much smaller than analysts were expecting (initial estimates were for a decline of 3-4% in Q1 2021).
If so the forecasts will have been out by quite a bit and we will be on a much better trajectory. The caveat as ever is that the monthly numbers can be erratic and we do not know February and March yet.
The Breakdown
We see two by now familiar patterns in the numbers.
Falls in consumer-facing services industries and education drove a contraction of 3.5% in the services sector in January 2021.
We see that declines through this phase are invariably led by the services sector and we see that the UK quirk in terms of measuring education is back. This pattern repeats if we look at the annual comparison.
The services sector was 10.2% below the level of February 2020 compared with 4.9% below the level seen in October 2020.
We can break this down into the various areas.
The service sector saw falls in 9 of the 14 sub-sectors between December 2020 and January 2021. The largest contributor to the fall was wholesale and retail trade, followed by education services, accommodation and food service activities, and other service activities such as hairdressing, because of the reintroduction of restrictions across the UK. There was growth in five sub-sectors, most notably in health activities, and information and communication services.
Of the 3.5% fall in the service sector some 1% was the retail and wholesale trade sector which is hardly a surprise in a lockdown. Also an issue I was the first to raise back on August 12th last year has returned. By this I mean education which reduced services output by 0.94% so let me explain the issue.
Health and Education
This is an area where the UK has chosen to measure GDP differently from elsewhere as we measure what is output whereas pretty much everywhere else matters inputs. So when the pandemic hit other countries recorded higher GDP from higher wages, overtime etc in the health sector whereas we noted some areas closing and thereby reduced output. We did so by an arcane method which was deciding inflation could take the slack and set the deflator for this area at 32% for one quarter.
Education was recorded as having lower output due to it switching from physical teaching to online. Whereas abroad it was regarded as pretty much unchanged and the same has just happened.
The education sector contracted by 16.3% in January 2021; the second largest contributor to the monthly decline. While remote learning activities are included for the education sector, the estimated output from these is lower than pre-pandemic classroom-based learning, as described in our School’s Out: measuring education output in the summer of the pandemic blog post.
I have left in the hyperlink as our statisticians have been poor in explaining this so I welcome any effort to do so. But if we do a back of the envelope calculation we see that there was a 0.7% fall in UK GDP in January due to this.
Just when you thought that might be enough there is something else to factor in and we are back to health.
Health had a large contribution to growth in January 2021, increasing by 8.7%, mainly through coronavirus testing and tracing and vaccine schemes across the UK. Adjustments for these programmes increased overall GDP by 0.9 percentage points in January 2021; some of these adjustments are in other industries than health.
You see after proclaiming we are using outputs we have switched to inputs/expenditure.
These very approximate initial estimates are informed by available in-year spending data for testing and tracing; the available estimated cost to secure and manufacture vaccines for the UK and deploy vaccines in England; available testing and vaccination data and estimated imports.
If there is any output measure here that is in another new section which is vaccines.
The January adjustment across all industries for vaccines, in volume terms, is £400 million, and for testing and tracing is £3,200 million. The equivalent total December 2020 combined adjustment for vaccines, and testing and tracing was £2,200 million.
I can vouch to some extent for the vaccine side of the argument as I had my first dose ( AstraZeneca) yesterday so I guess I will be a small part of the March figures. But if we return to the Test and Trace situation we see that we are counting an expenditure or income number in something of an inconsistency or if you prefer type of U-Turn.
Trade
This is something which has caused a minor furore on Twitter so let me start with why. This is what is being reported.
The ONS #Brexit trade stats are ugly, even allowing for December stockpiling & “teething problems” – exports to EU down 40.7% – exports to Ireland off 47% – imports from EU off 28.8% No similar falls with rest of world via @ChrisGiles_
This is from Peter Foster of the Financial Times and one might on GDP day take that to mean the UK has seen a substantial GDP fall. But the opening swerve is the use of percentages because actual numbers give a different picture.
Exports of goods, excluding non-monetary gold and other precious metals, fell by £5.3 billion (19.3%) in January 2021, because of a £5.6 billion (40.7%) fall in exports to the EU……..Imports of goods, excluding non-monetary gold and other precious metals, fell by £8.9 billion (21.6%) in January 2021, driven by a £6.6 billion (28.8%) fall in imports from the EU.
So as net trade goes into the expenditure version of GDP we see that EU trade boosted GDP by £1 billion and overall trade boosted it by £3.6 billion.
Care is needed as we also need to factor in the important services sector.
The trade in services surplus widened by £0.1 billion to £8.2 billion in the month of January. Imports fell £0.3 billion (2.4%) to £10.7 billion and exports fell £0.2 billion (0.9%) to £18.9 billion.
As you can see that changes the picture little.
Comment
The conclusion is that these are good numbers and put the UK on a better trajectory than we were told which has become a consistent theme. Care is needed with that as it means we have less of a depression than before but unnecessary gloom ( with the OBR to the fore) does not help.
Under the surface there are some big issues going on and two major ones relate to health and education. When back on August 12th I pointed out the issue it was health that seemed bizarre.
I helped Pete Comley with his book on inflation a few years ago with some technical advice and proof reading. I recall him telling me that he had looked into the deflator for the government sector and had discovered they pretty much make it up.
Bringing that up to date I have spoken to some physio’s who have been deployed to Covid wards and another who when returning to her normal role found all the PPE fogged up her glasses so it was very difficult to treat a patient. So it is complex and therefore a time for producing numbers which are internationally comparable.
Now education GDP has been hit again and the methodology has been dropped and in an irony for something about which many would argue has had issues with its output. Step forwards the “Test and Trace” system. On the other hand putting in the vaccines seems much more sensible.
“Contrariwise,” continued Tweedledee, “if it was so, it might be; and if it were so, it would be; but as it isn’t, it ain’t. That’s logic.” ( Alice In Wonderland)
Notayesmanseconomics's Blog 
Hi Shaun,
I used to believe that I understood GDP as some sort of yard stick, but I am not sure now. In particular, I simply do not believe the measurement is accurate to within several (many?) percent, let alone fractions of a percentage point. For example:
1. I followed that link in your article and, in a nutshell, they make up the figures. What exactly is the out put from a lesson in GDP anyway? It is apparently more when face to face than remotely – why? If 20 people have a Latin lesson, does that really increase GDP? The comical way in which they smoothed up education GDP in the holidays while locked down is just priceless.
2. How do they work out healthcare GDP? It looks like the more vaccines that are given, the bigger our GDP is, although all a vaccine does is try to return us to the status ante quo…
3. Imputed rents. I know that you have covered this before, but the concept that an imputed rent on the house that you own is added to GDP seems beyond parody. As people have suggested, why not add the cost of a taxi fare for every journey you take in your own car.
4. Seasonal adjustments appear to be just a smoothing technique. For whose benefit exactly are the figures smoothed? If sales go up before Xmas, well then, they go up. Why should a statistician arbitrarily declare that so much of that shouldn’t be included in order to smooth it out?
So, I am afraid I doubt whether they really know if the right figure is 2.9 or 5.9% It is a guess.
ah don’t forget the BoE ‘s sustainable blockchain seaganistic GDP-fluencer will be working in the ghost kitchen for those green swans from the loot box……
failing that , lick a finger and stick it in the air…… and order a double on the rocks
C tripadvisor
Forbin
James: if you remember that the primary (secondary & tertiary) purposes of state education is to provide us with the tools to better serve the elite, then every lesson learned makes us more useful to that elite. (I almost wrote, “more valuable!!!!” lol)
Since a rising GDP is of absolutely no consequence to the pleb, but is to the elite, it is quite correct to measure it thus.
As for health, £37bn for track & trace, (for no discernible benefit, according to the audit office) surely must be counted, as it is the elite who will eventually trouser it.
To put this into perspective, the original cost of HS2 going to be £36bn (although that was always a lie too)
Does anyone ever compare GDP movement to tax returns and other ways to measure economic activity? It would be interesting to see the correlation (or lack of)
Like wise – will anyone ever attempt to relate the Census results with say mobile phone usage? I would have thought that mobiles would give a good estimate of overall population as even illegals use one. Historically the census appears to under report the actual population according to later revisions.
Good God that sounds far too sensible
and anyways, it might let the cat out of the bag* ………
Forbin
* and frighten the horses …..
I was a census enumerator in 2001 and my better half was a census team lead. We collected information in South London directly along a route that is now sadly in the news at the moment. Going up to strangers’ front doors and knocking on them in some interesting estates is quite an experience.
It was very evident that a surprisingly large number of residences did not have the residents they were meant to and when the results where filled in it was clear they were reporting less people than those who were actually living there. We could tell when knocking on doors that there was more than one person fairly obviously. Sometimes we saw up to 6 in a 1 or 2 bedroom flat.
London has always been rife with sub-letting both legal and illegal. It is just so expensive to live there that having extra people living and paying the rent is an entirely normal thing to do. I knew of people who had moved out of housing agency property who let it to someone else and did not return the flat. The justification was that if they needed to be back in that part of London then they would move back into it. I suspect if Housing Agencies or cancel did an audit they might get a nasty shock.
It is really is no surprise to me that the census under-counts the population. For 2011 I was away on a work trip. The census states you record who was in the property on the day of the survey. I was therefore not included.
Tax records, benefits claims and school places combined would give a far better indications of our true population.
how about this method
” One best estimate of population comes from sewage: Thames Water says the outflow of rich or poor alike reveals true numbers – and Slough has 30,000 more people than officially registered. Bagley fears census-takers will have little time to check the estimated 3,500 houses in multiple occupation, or the estimated 2,500 living in unregistered sheds in gardens illegally let out, unlikely to return forms in the 56 languages they speak.
Besides, there are only slots for six people on each form.”
( the article is over 10 years old btw )
Forbin
I have found the most accurate method of measuring GDP, it is called Period Indexed Numbering
Basically, take a blank sheet of paper, write numbers 0 to 10 in 0.1 units and insert your P.I.N anywhere you like – infallible!
Hi Foxy
Perhaps the Bank of England was ahead of you although it would make an odd PIN.
Hi Shaun
If one of the g20 nations decided to make statistics
accurate re- GDP,Inflation or unemployment can you
imagine how awful they would look in comparison to
the “Rigged” figures, therein lies the problem!
A litany of evermore left field ideas by TPTB is certain
The film starts, the film ends.
Nothing is said in between. Leftield.
JRH
Hi JRH and thanks for the link
It would require the undoing of many years work. There is an episode of Yes Minister when Jim Hacker says something like “nobody believes the unemployment numbers, surely?” It is from 1983! It also gets worse because he is later informed that the school leaving age was not raised for education reasons as he had previously believed but to reduce the unemployment figures.
I have a wry smile at that whenever someone takes the claimant count numbers seriously.
Great blog as usual, Shaun.
Regarding measuring education output for the UK, I agree that the piece by Katherine Mills was difficult to understand. Just the same, it is a strength, not a weakness of the UK GDP estimates that as much as possible they try to estimate outputs and not inputs. If your measure of government output is based solely on inputs then productivity growth is always nil by assumption, so one can’t get any feel for public sector productivity. It was the 2004 Interim and 2005 Final Atkinson reports that really drove the ONS effort towards direct output measures, although the ONS actually brought in direct measurement of school output much earlier, in 1998. The measure used then was number of pupils with a 0.25% quality adjustment. If most countries still use input-based measures, one can find substandard methodologies that are the most commonly used everywhere in official statistics. The treatment of seasonal goods is probably the most glaring example.
My former colleague, Kam Yu, pointed out that education output can be based on” the total number of students, total number of student hours, or total number of student days.” In the Canadian SNA, total students, or more precisely, student enrollments, was the measure, until Statistics Canada lost its nerve and bended to pressure from teachers’ unions to revert to input-based measures. I had never thought that it made too much difference which of the three measures one used, but in the current pandemic conditions it obviously does, and student hours would seem to be the best measure.
Anecdotal evidence should generally be heavily discounted, but my own stepson, Miroslav, has gone through every situation from March 2020 on, from the schools being closed, to online instruction only, to the current situation, where he has two days of online instruction and three days of in-class instruction one week, three days of online instruction and two days of in-class instruction the next week. He is better set up for online learning than a lot of students, but he still much prefers in-class learning, because there are fewer distractions. If you think about it, it is really hard to decide how you measure quality of output for education with the switch to online learning in the pandemic, since, in a sense, home infrastructure takes on a big role in education that it never had before, so the education output is really produced jointly by the government and the household sectors.
Hi Andrew and thank you
Your middle paragraph raises a smile as it links to the reply I have just given JRH. In the Yes Minister episode I mention Sir Humphrey goes on yo explain that state education is run for the benefit of the teaching unions who run the show and then in a different country 20+ years later I read
“until Statistics Canada lost its nerve and bended to pressure from teachers’ unions to revert to input-based measures”
Same as it ever was?
As to the output point it is circumstances which have exposed it with 2 of the major areas affected being health and education. There are real issues with measuring it.I had the first AZ vaccine jab yesterday and that is pretty clear cut. But when I had the ACL in my right knee repaired Dr.Thompson at Chelsea and Westminster hospital did a good job for me but that has never been measured. So whilst in theory you are right in practice I think there are loads of issues.
In a sense though it could easily be fixed by producing both output and input numbers so things could be compared internationally
I hope your step sons education goes well. At least you were able to provide home infrastructure so he could do it as not everyone has been able to.
This is what I read from a number of sources today:
“UK economy shrinks less than feared in January”
https://uk.news.yahoo.com/uk-gdp-january-2021-covid-19-coronavirus-brexit-070452101.html
Am I missing something as the data released was much worse than forecasts?
GDP yoy -9.2% forecast -7.8%
Industrial production jan -1.5% forecast -0.6%
Industrial production YOY -4.9% forecast -4%
Manufacturing production mom jan -2.3% forecast -0.8%
Manufacturing production yoy jan -5.2% forecast -3.2%
???????????????????
newspeak
today’s chocolate ration has been increased from 4 oz to 2 oz
forbin
Well, if you multiply by fractions!
Come on Forbin, you’re more switched on re economics than I, got to be a joke in between fraction(al) (p)reserve and ba(n)king regarding your chocolate!
Past my bedtime, have a great weekend everyone; maybe we can inflate our GDP with our humour quotient?
Iain
People in the Isles of Scilly eke out a living by taking in each other’s washing – Mark Twain
Hi Eric
If they also cut each other’s hair and grass they could generate quite a GDP boom.
Watch out for the blood clots, Shaun. Didn’t think you looked that old!
Hi JimW and thank you for the good wishes.
So far so good. As to the clots I had seen the news from Italy on social media before I went so I asked about it? I was told that the UK had used lots of AZ vaccine so it should be news if true.
It is a complicated issue as there is a lot that is uncertain about all this but I decided that I would feel dreadful if I did not then found I had passed Covid on.
Oh and I have always looked younger.It was less fun when I was 27 and had to get my id to be served in a bar in Colorado.
No good statistical evidence yetr that it cuts transmission rates ( certainly not the recent point to point NHS Scotland data), but here’s hoping.
My 30 yr old son had exactly the same problems both in MA as well as CO.
And my wife many years ago in Glasgow, even in the ‘lounges’.