If we cannot measure construction accurately what about the rest of GDP?

The modern era has seen economic growth as measured by Gross Domestic Product or GDP become something of a Holy Grail. It is also rarely out of the news for long. For example early this morning we saw the latest number for China. From Bloomberg.

Gross domestic product rose 6.7 percent in the first quarter from a year earlier, meeting the median projection of economists surveyed by Bloomberg and in line with the government’s growth target of 6.5 percent to 7 percent for the full year.

Quite a heroic effort in a country as far-flung as China to know the number with such accuracy only 15 days after the end of the quarter! The more homogenous UK which takes around 10 days longer thinks by then it only has around 40% of the final data. But this obvious problem is not my subject as i wish to focus in on one subject which is construction and the enormous problems the UK has had in measuring what you think would be a relatively simple thing to do.

On Tuesday I did a little of it myself as I cycled around the developments at Battersea Power Station and Nine Elms. For those unaware of the geography the site extends from Battersea Park to Vauxhall and must be on the scale of what happened at Canary Wharf. Residential tower block after tower block is being built there showing what I know as I thought they had gone out of fashion after the concrete jungles created in the 1960s and 70s! Also the American and Dutch embassies are relocating there. As a principle you would think that this would be easy to measure as the buildings are obvious and many of the machines used to build them are on a similar scale. The local public houses must also be getting an enormous windfall from thirsty construction workers.

Problems, Problems

We only need to look back to the latest UK GDP update to get a hint of the issue here.

Construction output decreased by 0.4% in Quarter 4 2015, revised down 0.3 percentage points from the previously published estimate.

As you can see there was quite a change and this is for the new supposedly improved system. Indeed even the new version for 2015 poses questions.

Construction output in 2015 as a whole was 3.4% higher than 2014, much lower than the rate of growth for 2014 (7.5%). This was largely due to the 2 consecutive quarters of negative growth in the second half of 2015, with construction output falling by 1.7% and 0.4% in Quarter 3 2015 and Quarter 4 2015 respectively.

If we compare this to the surveys conducted by Market for its Purchasing Managers Indices we see an answer so different someone has to end up being embarrassed. You see over the period in 2015 where the official numbers showed a contraction the PMI varied from 55.3 to 59.9. So we had solid growth or surging growth on that measure in every month. As Kipling so aptly put it.

OH, East is East, and West is West, and never the twain shall meet,

Analysing the difference

The UK Office for National Statistics has had a go and they find themselves having to admit the central issue .

(they) have painted a different picture of the underlying performance of the construction industry; raising questions about the quality and accuracy of both sources.

This seems to be the limit of current expectations.

post 2013, both show an underlying upwards trend for output in the construction industry.

Frankly I can see that by looking out of my window! Also whilst the first bit below is true the latter is not as revision follows revision.

the Markit PMI is the more timely estimate, but our slower release of data enables us to provide a more comprehensive coverage of the industry.

I can agree with this bit.

It is not surprising, given the conceptual differences explained in this article, that the 2 measures sometimes give different signals.

You see the PMI series does have issues of its own and it is not a direct comparison with the output series on a monthly basis. Okay. But you see this does not address the “up is the new down” issue over the latter 6 months of 2015! The PMI series records very strong growth and the official data shows a contraction.

February 2016

This morning’s data just continues what is becoming an even longer sequence. Here are the official numbers.

In February 2016, output in the construction industry was estimated to have decreased by 0.3% compared with January 2016. Both all new work and repair and maintenance reported decreases, falling by 0.2% and 0.5% respectively.

This compares to a Markit PMI reading showing steady growth as both February and as it happens March had a reading of 54.2. Now the ONS can make a critique of monthly differences but we are getting quite along period now of completely different patterns for the two series.

Oh and the Markit series is slowing down overall whilst if we switch to the quarterly numbers we may be seeing a pick-up in the official data overall.

Comparing the 3 months, December 2015 to February 2016, with the previous 3 months, September 2015 to November 2015, construction output increased by 1.5%.

I like to Move It,Move It

There have been some major changes along the way and people ask me how things are categorised. Well the ONS is sometimes not so sure either.  From Brickonomics last October.

Having explored the issue more, but far from exhaustively, it appears in March there was a reallocation of a major business from the services sector to construction. When I say major I mean a firm turning over significantly more than £1 billion a year, probably more like £3 billion.

Oh and it gets better as the numbers were imputed and turned out to be rather different from the actual ones and this was discovered because Brickonomics had wondered what was going on here.

in April the ONS decided that on an annualised constant price basis the industry was £3.7 billion bigger than it was thought a month earlier. Blimey.

Or to put it more officially.

Construction output increased by 1.4% in Quarter 2 2015, revised up 1.2 percentage points from the previously published estimate.

Blimey indeed.

There have been various attempts to improve this such as a formal meeting which appears to have disappeared from the new ONS website. But for now let me just point out that it has lost the National Statistics label and solutions are described as “interim”


If we look at the UK construction series we see that this all began with a problem that is my central theme. It started with the inflation measure or deflators which misfired which posed its own problems for the output series. Next the output series hit more trouble and the seasonal adjustment in another familiar theme is not going so well either. There isn’t much left really is there?

This poses quite a problem for the overall GDP numbers. Whilst construction is dwarfed by the services sector it is still a bit over 6% of GDP or at least we think it is. But our inability to measure it poses a serious question for those attempting to measure more inanimate output such as the “sharing” economy and indeed much of the services sector. After all if we are struggling with cranes,diggers and buildings…

I’m fixing a hole where the rain gets in
And stops my mind from wandering
Where it will go ( The Beatles)

My personal view is that the situation is better than the official numbers show. I was only half-joking when in the past I said that I would be counting cranes! But whilst there is an obvious bias in what is happening near me other anecdotal measures including industry contacts are positive. Also the Federation of Master Builders was so too at the end of 2015.

activity in construction rose for the eleventh consecutive quarter in Q4. This growth was reported by firms across all areas of the industry, and was led by new building activity in the private housing, commercial and infrastructure sectors.

But how much is less sure.


Here are my views on how the IMF was changed to suit political rather than economic ends as expressed to TipTV on Wednesday.










11 thoughts on “If we cannot measure construction accurately what about the rest of GDP?

  1. “If we cannot measure construction accurately what about the rest of GDP?”

    given that so much is imputed , or frankly guessed at , does it matter?

    so long as HMG remembers its all hand waving and guesswork …… ah I see the point ..

    oh I do love how the world is gonna end because China has 6.7% growth

    my gods , our pollies would be gagging for that level , or even half !


    • Hi Forbin

      The output numbers for GDP are the main ones and should be in terms of manufacturing and construction the easiest to count and measure. After all they are physical things rather than inanimate but even so we have more than a few problems.

      The imputed rent numbers are there to square the circle for the income series of GDP which is then supposed to equal in the output ones. But they have problems of their own and invariably are not the same in total. Indeed the imputed rent numbers seem to be behaving differently to the CPIH rental series. Oh dear.

    • Hi Foxy

      After that very quick time period to calculate results no we cannot. I note this from Gordon Chang in Forbes magazine earlier this year.

      “So what does the “Li Keqiang Index,” as these factors are now known, tell us? The usage of electricity remains the most reliable single indicator of Chinese economic activity. In the first 11 months of 2015, electricity consumption increased 0.7%.

      Rail freight volume fell 10.5% in 2015, according to Caixin, which cited the National Railway Administration.

      New renminbi loans last year, according to the People’s Bank of China , amounted to 11.72 trillion yuan, an increase of 1.81 trillion yuan from 2014. Foreign-currency loans in 2015 fell $50.2 billion. In 2014, such loans increased $58.2 billion.”

      He thinks the economy may have contracted in 2015. Now I am not so sure about that but it appears to be growing much more slowly than we are told.

    • Hi Eric

      I like it! I am pleased that my influence reaches down to NZ 🙂 More seriously it does give a guide to construction output at a basic level and should be used up here too because we are in a mess.

      You may also be amused to note that an FX website I follow has the Kiwi Dollar as one of its major currencies.

  2. Mark Carney, Governor of the Bank of England, said: “On behalf of the Bank of England, I am delighted to welcome Michael Saunders to the Monetary Policy Committee

    frigg N odin , Shaun , there aint no justice !!

    that should have been your job !


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.