FT Alphaville has an article this morning that has the headline:
The UK's inflation figures were wrong and now we just have to live with that
I won't quote it at length, although it is good investigative financial journalism of the sort that this team is known for. That is because the essence of the story is easy to summarise.
What the Office for National Statistics, which is responsible for preparing UK inflation data, has been doing is taking UK list prices for consumer items as the basis for their estimates. They have not been using the prices we actually pay, which are invariably lower. That is because of, for example, the existence of things like the Tesco Clubcard and equivalent schemes, which discount prices for those who join such schemes, which anyone can do without charge. They have pretended that everyone pays the higher price, when the majority of people do not.
This has a big impact. As they note, it was suggested in March 2024 that consumer prices rose at an annual rate of 3.2 per cent, down from 3.4 per cent in February. But if prices actually paid had been used to prepare the data, the ONS thinks that the CPI figure for that March should have been 2.8 per cent — 0.4 per cent below the originally-reported figure.
This is a massive difference, given the obsession with this inflation data, and its importance in policy making and in establishing things like price increases within the economy, as well as influencing pay rises. And all the while, the Office for National Statistics acted like typical economists and pretended the world was not what it really is, but was instead what they would have liked it to be. The consequence was a serious misstatement of inflation data.
I have long criticised the Office for National Statistics for its misstatement of the national debt. Now we know it also misstates inflation. It needs a radical overhaul. And one day it might even get a usable website, but that might be too much to hope for because then we might all realise just how bad it is.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
In a recent article, you wrote about the irrelevance to most people of the relentless economic news, and suggested a more relevant ‘basket’ of purchases to represent what most of us face. My council tax has gone up by over 10%, energy 6%, water has gone up (I haven’t tried to work out as moved house & now have water meter). I am curious about my personal inflation (not talking carbohydrates here!) as those bills are quite a large chunk of my monthly outgoings. Does anyone know of an online calculator for this? Am I right in thinking the lower our spending (most on necessities) the higher the inflation % on that person’s ‘basket’?
The ONS did start to talk about this a while ago but appears to have lost the plot on it
Jack Monroe drove the demand. She has given up because of social media abuse.
Inflation statistics usually leave out items such as housing, food and energy. I believe the ~3% figure that Richard is talking about would exclude such items. The ONS website reports an inflation figure above 5% when energy, housing and cost of living are included. They also report a figure called “owner occupiers’ housing”. This figure is even higher still (7%) and does not correlate with other causes of inflation. This figure is what people experience when their household bills increase. I agree with Richard that the problem that people face is not ‘inflation’ as economists define it but as you said it is rising bills which inflation figures completely ignore.
Beware all those numbers
They a re deeply judgemental
FT Alphaville has been like a dog with a bone on this topic…. indeed, their pieces on “small caged mammal prices” have achieved cult status and “merch” is available. The ONS must hate them more than you!!
The simple solution (as FTAV says) is just use the data that the retailers already have when they scan your items at checkout. This is give you the true prices paid as well as the correct “weightings” for each product in the CPI basket. Job done.
The broader question about does it reflect the “true” cost of living for ordinary folk is still an important one… but that, I think, revolves around people freighting a single number with too much importance.
So, get better data first. Then think carefully what the data do and don’t do before using them to determine policy. Jack Monroe was good on this – I wondered where she had gone. Such a shame.
The pressure became too much.
I met her several times, including one extended trip together, and liked her a lot.
Jack did some really good practical work. Her books are so helpful, her Twitter advice line during Covid was invaluable and her work with ONS could have made a big difference.
Such a shame she got hounded by social media.
I hope she is safe and living a quiet, happy life now.
Me too. We are no longer in touch.
“…just use the data that the retailers already have when they scan your items at checkout.”
Looks like they have taken that on board?
https://www.youtube.com/watch?v=BJJ72RicBNI
It looks like it….
Economists? Find better data? They wouldn’t know where to look, or what to do with it. Or even recognise it. It isn’t in their curriculum. Trite armchair maths is all they know.
True
And data quality appraisal appears to be quite beyond them. I used to teach it.
It’s a fair point, but what interests most people, or should do, is not the absolute level of inflation, but the trend. Is it higher or lower? I’m not convinced that using market prices rather than list prices would have a great impact on the trend of inflation.
I think it might, because this data is used to set prices
Agreed. If you’re comparing the non clubcard price now to the non clubcard price of a year ago, the trend will be reasonably close to the comparison of the discount price now to the discount price then.
But the data may actually run the other way.
e.g. headline price of an Easyjet or Megabus could be say £30
Actual price paid could be higher as you’ve brought some hold luggage or even god forbid chosen a seat.
Anecdotally I’ve heard a private primary school trying to bill add-ons in order to keep the headline price of a term’s tuition down. £1 for a slice of toast with toppings for example, when previously snacks were included in the fees.
What if the ONS are just measuring the list price and not accounting for the extras. That taxi journey that includes a drop-off fee likely enters the ONS data as just the base taxi fare.
I don’t see how it can have a long term effect, if prices now are over-reported due to discounts, then surely last year’s prices were also over-reported for the same reason.
The ONS think it does, now
To my mind the change will give them a one off cut in inflation this year only
Maybe…but the ONS seem to think not
In otehr words, they seem to think the moves are not synchronous, as you suggest
The question that I ask is what purpose does the CPI rate of inflation play these days, other than to give the politicians something to quote occasionally?
Like when Starmer and Reeves come out with their mantra that average wage increases are currently higher than inflation?
It’s a bogus claim, because a) CPI gives us a figure that is so far removed from every day experience that it has become meaningless for many of us, and b) not everyone is enjoying the benefit of that “average” wage increase. As always, millions are getting left behind as real inflation eats away at their lives.
I’ve long argued that there are only two forms of inflation for most people, “need” and “non-need” inflation.
Need inflation are the bills we have no choice to pay — they are often covered by law; if you don’t pay, you could end up in court, or the bailiffs will come knocking. They take up a large percentage of income (or increasingly savings) for most people. Mortgages and rents, Council tax, water rates, energy costs, travel costs, telecoms, insurance (for those that can afford it), and food all spring to mind. All of these are going up by a lot more than CPI. I suspect partly because the greed merchants that provide them know that they have a trapped market, where genuine “free market” choice doesn’t really exist. If anything, the greed merchants behave like cartels — there is no real competition between them, other than how much they can get from us.
This year my rent was up 12%, water rates 26%, energy costs, difficult to work out because the energy companies tend to lower in summer (this year they just raised them) and then bang the rate up just in time for winter. I do know from my own personal record keeping that my energy bill has doubled in the last 3–4 years. My food bill has doubled as well. Now, I’m frugal, a penny pincher, because I always have been, but what I see as real inflation is horrendous.
When a politician tells me that inflation is just 3%, I wonder what planet they are on. I would love to pay just 3% a year for all the need items just mentioned. Think how wonderful life would be if that were the case! The fact that I might only be paying 3% more this year for a trinket I don’t need is meaningless.
CPI inflation mostly reflects the non-need goods and services of life, so I wonder what purpose does it serve? And I’m afraid the RPI and CPIH index are not much better at showing real inflation either. I see the official number as a fraud, but I’m sure the politicians might be attracted to another way of lowering it.
The CPI number is purely for the politicians who consistently fail to address the real issues we face. Ultimately, there is a massive shift of wealth going on here, from the have-nots to the haves. I see Labour doing very little to address it, and I think they will pay a price for that.
They will pay a price for it….we might know the scale of it on Friday
I wonder if the ONS takes into account regional variation in pricing, which I believe some large supermarket chains to employ?
I very much doubt it
When looking at inflation figures I often think about the price of a pint of milk. Should be a simple figure, you would think, but:
At Tesco:
1 pint £0.85 (£1.50 per litre)
2 pints £1.20 (£1.06 per litre) to £1.85 (1.63 per litre)
4 pints £1.55 (0.68 per litre) to £2.25 (0.99 per litre)
That is only fresh cow’s milk sold in multiples of pints, the higher figure being organic. There are ordinary and premium cow’s milks sold by the litre at different prices.
So what does the ONS use as the price of milk?
As I have said before the inflation figure has no basis in reality, and should not be used to set pension or state benefit levels unless it can be shown to be relevant to the costs incurred by those in receipt of ‘money from the state’
Good question
It seems that they have taken the highest priced option…
If ONS just took the total money paid for milk (by Tesco) across all types/sizes and divided by total litres sold then you would have a “true” number for the “average” price paid for milk.
Indeed, you could broaden this out across all groceries…. Tesco certainly has those data.
It has
Will they share it?
Why don’t they publish it?
Inflation is a much overated measure of financial life and the basket of goods used to calculate inflation is unrepresentative of daily life.
What does it mean that inflation is only 3.2%. Do i only need 3.2% more money than last year? A much better measure would be how much money a person/couple needs to maintain their life style.