How the UK's GDP figures are being gamed and manipulated

28th January 2014 by The Harried House Hunter

Today has seen the release of the economic output figures for the UK economy in terms of its Gross Domestic Product for the last quarter of 2013 and hence for the year as a whole. Whilst the numbers are good news as the UK is in a recovery phase I want to examine what they really tell us and how they do not tell us what is often assumed. Let me help out by showing a development that will take place later this year.

The current methodological framework for producing national accounts data (ESA 1995) will be replaced in September 2014 with a new European System of Accounts (ESA 2010).

This seems a harmless enough updating does it not? Well let me get to the meat of the changes.

Recognition that expenditure on research and development has the nature of investment….This will increase EU gross domestic product by around 1.9%.

That is the main change though there are some other more minor changes which mean that the total increase in GDP will be this.

the weighted average impact of 2.4% in Europe

Do you feel better off and wealthier already? If anybody should, it should be readers in the UK who are expected to do better than the average and see an increase of between 3% and 4% at the stroke of a statistician’s pen or perhaps I should say computer mouse.

There is of course plenty of scope for a wry smile at this change which has already taken place in the United States (last July) as we note that many metrics used such as national debt to GDP ratios will be improved. How convenient! There is of course the issue of telling everyone they are better off just at a time when the going is tough with falling real wages. How politically convenient!

Inflation has also been used

Back on the 25th of January last year I reported on a change in the UK’s national accounts from using the Retail Price Index (RPI) to the Consumer Price Index (CPI). It may seem innocent enough until we consider that if output is the same substituting a lower inflation measure will lead to real output being recorded as higher. Hey presto it is magic.

The economist Sam Williamson did some calculations about this. Below is before and after. Let me see if you can guess which is which?

For the last 62 years the United Kingdom’s real GDP has grown at a rate of 2.43%.

The new story is that for the last 62 years the United Kingdom real GDP has grown at a rate of 2.68%.

Good this isn’t it? Soon the credit crunch will be a figment of our imaginations! Can I spend it?

What about rising house prices?

The switch from the RPI to the CPI did have the effect of removing any influence from this area as whilst the RPI has an owner occupied housing component the CPI does not. Convenient eh? Even the most ardent admirers of the CPI system eventually caved in and admitted it required a new version which had a measure of housing costs. Regular readers will be aware that I argued against their choice of rents (Rental Equivalence) rather than house prices for this because it would, on past history, fail to give us useful data should house prices rise again. It has already turned out to be the case that I was right and the official bodies such as the Office for National Statistics (ONS),the National Statistician and the Consumer Prices Advisory Committee were wrong. I will leave you to make your own mind up about house price rises and give you the latest ONS rental series data.

In the 12 months to December 2013 private rental prices paid by tenants in Great Britain rose by 1.0%, down from a 1.1% increase in the 12 months to November 2013.

As you can see that is not really cutting it, even if we make an allowance for public-sector rents being likely to be growing faster. Although actually this seems to have little influence.

The OOH component annual rate is 1.0%, unchanged from last month

Our measure of housing cost inflation (CPIH) is a national embarassment at 1.9% – as in below the 2% of the ordinary CPI. Or as it would be summarised on twitter #fail.

Accordingly I have argued at the Royal Statistical Society for this.

Do we have a way of asking for the CPIH debate to be re-run?

Actually in an irony of modern language, this would be an actual improvement as opposed to the “improvements” regularly claimed by the statistical bodies. It also seemed to be the equivalent of kicking over an ants’ nest as the ONS rather unusually replied and here is Richard Campbell, the head of consumer price statistics.

None of the methods of measuring owner occupiers’ housing costs aim to measure house prices. It is generally accepted practice that consumer price indices measure consumables, not assets. The merits (or not) of a consumer price inflation index including house prices is a different discussion to the method used to measure owner occupiers’ housing costs.

I could write for days if not weeks on my disagreements with that paragraph but let me instead quote from the Eurostat technical manual. You see the first three words in the section defining owner occupier costs are House, Price and Index! Richard Campbell replied in what I would argue is a version of ‘they didn’t really mean it’ or ‘the dog ate their homework whilst chasing the cat’, or ‘it was raining that day’.

So, while the Eurostat index does include an element of house prices it does so as a compromise and does not aim to include them.

Well that’s all right then! But it would seem that they are not following this advice from Elvis Costello.

Oh, Alison, my aim is true.
My aim is true

This matters because in my opinion house prices are a big influence in the UK economy and therefore our inflation numbers should reflect them as best we can. It is not easy as there is the debate over what is growth and what is inflation but in my opinion a proper CPI should be recording it around 2.4% or 2.5% right now and it should go into the GDP numbers. The rub as Shakespeare would put it, is that recorded real output or GDP would fall for the same actual level of output and I suspect that is not the only reason behind the resistance.

It is one of the rules of life that there are always plenty of apparent reasons for doing the wrong thing.

Today’s numbers

These provided a bit of winter cheer although they were below the expectations provided by the business surveys and bring another in an apparently never-ending sequence of forecasting errors from the Bank of England. Some much for the reviews of this as a latter-day Sir Humphrey scuttles under the nearest stone.

Change in gross domestic product (GDP) is the main indicator of economic growth. GDP increased by 0.7% in Q4 2013 compared with Q3 2013.

GDP was 2.8% higher in Q4 2013 compared with the same quarter a year ago. GDP is estimated to have increased by 1.9% in 2013, compared with 2012.

If we break this down, the vast majority of the quarterly growth came from services (0.61%) but there was a solid contribution from production (0.1%). Agriculture rose but not by enough to be material and disturbingly, considering its recent nuclear winter, construction fell (-0.02%). The underlying index where 2010 is 100 rose to 104.4 and it is nice to see us moving away from 100 where we were stuck for a good while. Should you want a breakdown of the UK economy, the ONS estimate is shown below.

The current 2010 based weights are: services 77.8%; production 15.2%; construction 6.3% and agriculture 0.7%.

Although in not quite so cheery news, there is still one issue to be overcome.

In Q4 2013 GDP was estimated to be 1.3% below the peak in Q1 2008


We should welcome the apparent turn in the UK economy as we desperately need every scrap of economic growth that we can find. Post credit crunch it has been and I think will remain a much rarer commodity than it was before. However as I have shown in my analysis today the improvement is unlikely to be as much as we are being told as the manipulations and machinations persist. All of the “improvements” seem to be in the same direction and a lack of actual growth has seen them appear out of the woodwork. So as we also recall that the preliminary release of GDP data only has around 45% of the final information let me leave you today with the famous line from the US cop series Hill Street Blues.

Hey! Let’s be careful out there

28 thoughts on “How the UK's GDP figures are being gamed and manipulated”

  1. forbin says:

    Hello Shaun

    forgive my ignorance but shouldn’t there be a chart somewhere that posts both the forcast and actual GDP rates?

    We can have one for OBR, IMF and BOE . After all if we are to judge the effectivness…..

    Other than that I’d posit that GNP and the balance of trade are the real hallmarks to be watching


    1. Pavlaki says:

      Totally agree re GNP and balance of trade.

      1. DaveS says:

        GNP per capita.

        And yes balance of trade but as Shaun has pointed out nobody has been “interested” in that for decades. Why ? because hey didn’t want anybody to focus on it. Because globalisation meant huge trade imbalances – it was always a lie that we could pay for goods imports by exporting high value services.

        Point is, they set the agenda, they choose what metrics the headlines report and they choose how to manipulate them. It will only get worse until eventually the lies unravel like they did in Argentina.

    2. Anonymous says:

      Hi Forbin

      We do not get a GNP estimate at this stage (they come with more detail in the later releases). We just get the basic GDP ones.

      As to estimates of the GDP growth rate there was this from the Bank of England in the latest MPC minutes for the 8th/9th of January meeting.

      “there were upside risks to the Bank staff’s estimates of growth of a little under 1% per quarter in the fourth quarter of 2013 and the first quarter of 2014.”

      If they were a centre forward you would not want the seats in the rows above the goal would you?

  2. Hans van der Meyden says:

    Why not make up your index and publish that

    1. Drf says:

      Isn’t that what they are doing already?

      1. forbin says:

        LoL! no truer word said , soon we’ll have one for every day of th week 😉


        1. Jim M. says:

          one for every day of the week

          Or one for any given set of circumstances.It’s always nice to have options, after all!

    2. Anonymous says:

      Hi Hans and welcome to my part of the blogosphere

      I have been considering something along those lines. The problem at this preliminary estimate stage is that you do not get sufficient data to do so. Actually you could argue that the ONS does not have enough data so quickly after the end of the relevant quarter but that is another story.

  3. anteos says:

    Great article Shaun.

    i wondered how much gment spending and imputed rent increased GDP.

    1. Anonymous says:

      Hi Anteos and thank you.

      We do not get the imputed rent details until the final (3rd) estimate I am afraid. As to the government there is only the section “government and other services” (22.5% of our economy) which rose from 102.4 in the 3rd quarter to 103.0 in the 4th.

      Over the year that category has been a laggard as it was 101.7 at the end of 2012.

  4. GusBmth says:

    Hi Shaun

    Thanks for such a detailed analysis of the GDP data, so refreshingly informative in comparison to the banal guff that many supposed expert commentators produce.

    From what you know, do the people in the Treasury and Bank of England actually believe the official numbers to be accurate (and base their decisions on them) or are they aware of the distortions and manipulations that show a much rosier picture of the economy than, in fact, is the case?

    What do you think would be an accurate estimate for the real growth of the economy in 2013 and the Q4?

    Many thanks

    1. Drf says:

      Well clearly Carnage does not believe the Uneployment numbers either?

    2. forbin says:

      Yes, I think they actually believe RPI which is why their pensions are linked to it ……

      Dont forget bennies are linked to CPI……. sneaky cost cutting ? Well the public will not notice that until many years later when the pollies that brought it in are safely retired on their index linked pensions….

      Great isn’t it ?


      1. Drf says:

        It’s not that they believe RPI; they know this is rigged just like all the other government-controlled statistics. However it is just that they know that linking to RPI means they lose less purchasing power and lose it more slowly than they would with CPI linking!

    3. Anonymous says:

      Hi Gus

      I do not think that I am releasing any particular confidence if I reply that some at least would like better quality numbers than those produced by the ONS. However there are still those who seem to rely on them pretty much lierally.

      As to the numbers I think that around 0,1%/0.2% per quarter needs to be knocked off to allow for the inflation machinations that have taken place.

  5. Anonymous says:

    So increased rents will make reported GDP higher whilst impoverishing the poorer members of society, based on purchasing power parity ….

    Oh what a tangled web we weave when we practice to deceive

    1. Noo 2 Economics says:

      But that would increase housing costs simultaneously rendering that part of the GDP/inflation calculation a zero sum game.

      1. Anonymous says:

        GDP increase at 2% means the economy doubling every 35 years. Sustainable? Clearly not. What answer does the “science” of economics have for that little conundrum? For those who want a clue, the answer is no answer whatsoever.

  6. Paul C says:

    The numbers just can’t be trusted. I suggest that we abandon counting services an only pile-up the tangibles each week. How many new houses built, cars left the factory and loaves of bread baked. The hairdressing and legal advice kind of services are necessary but they are difficult to count. I guess it is a pretty extreme approach,it could be used as a common sense measure to compare with political rhetoric- when they say GDP is up we can count up the tangibles and compare to last year and say “nope its not in real things though”.

    1. Anonymous says:

      Hi Paul

      It remains a problem that the largest part of our economy is measured so badly. If you look at the IT developments in the Old Street area and the flow of skilled migrants to work there (If the Evening Standard and others are accurate). there are clearly growth areas. But how much? Remains a mystery..

      Let me illustrate from the latest balance of payments bulletin.

      “This bulletin also reports on trade in services. However, the information on trade in services is mainly obtained from quarterly surveys, in some cases underpinned by larger annual surveys. That
      means that the data for the latest months are inevitably uncertain.”

      Apparently not so uncertain that they cannot put them in a GDP release….

    2. Eric says:

      Remember when the stock of bricks at the London Brick Company was used as a measure of how well things we’re going in construction?

      1. Paul C says:

        Absolutely, a real measure of sorts, perhaps combine that stock holding measure with a basket of numbers of masonry products produced and sold then we would have a real measure to compare to debased £value measures that now require a confirmatory index. I guess the stock-holding measure was a 1950’s (olden days) thing so I don’t remember it specifically.

  7. Noo 2 Economics says:

    Hi Shaun,

    Looking at your replication of Richard Campbell’s reply to you, he has given a crafty response. He doesn’t say or imply that CPI is “better” or more accurate than RPI, simply that it measures different things. I guess he probably isn’t interested in getting a “true” picture because you can argue for the rest of your life about which things you should measure and what weightings to apply to them to arrive at a true picture.

    I find his argument about CPI not measuring asset price inflation intriguing – isn’t a house a consumable as well as an asset? Surely the act of living in it constitutes consumption of shelter?

    Does the CPI measure equities/bonds etc in it’s calculation?

    Another question – does the RPI measure include movements in mortgage rates as part of it’s housing cost calculation?

    Something to muse on – the CPIH should, eventually, reflect housing price increases in a lagged response as landlords with recently purchased properties increase rents to cover their greater purchase prices. Of course if you want current inflation information then the CPI isn’t going to help much and bizarrely the housing costs part of CPIH could be showing increases (reflecting purchase prices paid by landlords 12 – 24 months earlier) just as a house price crash arrives. This will make the BOE’s job ever more difficult in deciding the best policy (note I haven’t mentioned the Treasury purposefully!!)

    1. Anonymous says:

      Hi Noo2

      The issue of housing and how you account for it is quite a minefield. For example some argue that the land value element should be excluded because it is not consumed but then you find something else which is difficult to measure!

      “Does the CPI measure equities/bonds etc in it’s calculation?” Easy answer nope.

      “Another question – does the RPI measure include movements in mortgage rates as part of it’s housing cost calculation?” Yes it does and the weighting last time I looked was 2.9% of the RPI.

      Overall housing costs are 25.4% of the RPI.

  8. David Lilley says:

    You were the first commenter to point out that our economy had an £8.6b helicopter drop in the form of PFI compenstion in 2013. It will also get a helicopter drop of five times that amount in 2014.
    UK companies are set to unload £86b (possibly even £100b) of their cash stockpile on shareholders in the form of dividend this year. But since half of the FTSE 100 is owned by foreigners only about £43b will land here.
    Thanks to your post above I now understand why the sumation of the four 2013 quarters of growth, 0.5%, 0.8%, 0.8% and 0.7%, 2.8%, far exceeds the 1.9% annual growth rate. But the 2.8% (which I shall round to 3% to take account of future revisions) is the more interesting figure as it exceeds the 2.43% UK average trend rate mentioned above.
    The health of the baby (UK economy) is measured by how much weight it put on since its last weighing and not compared to ages ago. The 2.8% figure tells us more about the velocity of gain than the 1.9% figure. Remember Mark Carney was court out on the velocity of UK jobs growth/unemployment fall. What he expected to take two years took just two months as we hit 7.0% unemployment in Ocober 2013.

  9. therrawbuzzin says:

    Down in the pleasure centre,
    hell bent or heaven sent,
    listen to the propaganda,
    listen to the latest slander.
    There’s nothing underhand
    that she wouldn’t understand.
    Pump it up until you can feel it.
    Pump it up when you don’t really need it.

    1. Patrick, London says:

      Two Elvis Costello references in one article/talkback… today is a good day. :)

Leave a Reply

Your email address will not be published. Required fields are marked *