This morning brings us more on what has become the troubled construction sector in the UK. Or to be more specific what we have been told by our official statisticians is troubled. Regular readers will be aware that I found some of this bemusing partly due to geographical location as there is an enormous amount of building work going on at Nine Elms around the new US Embassy. The last time I counted there were 32 cranes in the stretch between Battersea Dogs and Cats Home and Vauxhall. Also there have been problems with the official construction data series of which more later going back some years which have led to me cautioning that the numbers may need to be taken with the whole salt-cellar rather than just a pinch of salt.
What happened last week?
I pointed out on Friday that there had been ch-ch-changes.
This has been driven by revised construction estimates, with its output growth revised up by 1.9 percentage points to negative 0.8%
This was the road on which total UK GDP growth was revised up from 0.1% to 0.2%. It takes quite a lot for something which is only 6.1% of total output to do that but as it was originally reported at -3.3% then -2.7% and now -0.8% you can see that the original number was way off. This is a familiar pattern albeit not usually on this scale and does pose a systemic question. After all if you are struggling to measure something which is mostly very tangible such as a building and the associated economic output how can you measure the more intangible outputs of the services sector?
Actually there was more as the reformist wave spread across the data for 2017 as well.
While the 0.8% fall in Quarter 1 marks the largest quarterly decline since Quarter 3 (July to Sept) 2012, it is now estimated that this is the first fall since Quarter 3 2015 – earlier estimates had recorded falling output through much of 2017.
This does alter the narrative as we had been given numbers indicating a recession and at the worst hinting at a possible start of a depression, so it is hard to overstate this. Let us drill down into the detail.
Today’s new construction estimates show a much stronger growth profile throughout 2017, with upward revisions recorded in each quarter except Quarter 3
The major shift numerically is in the first half of 2017 as the first quarter goes from growth of 2.4% to 3.2% and the second from -0.4% to 0.4% . However in terms of impression and mood the last quarter may have hit the hardest as after previous doom it had the gloom of -0.1% whereas in fact it grew by 0.3%, Adding it all up gives us this.
Construction output is now estimated to have increased by 7.1% in 2017, up from 5.7%
What has changed?
Reality is of course unchanged by the way that it has been officially measured has seen these changes.
As part of the wider improvement programme for construction statistics, ONS has introduced significant improvements to the method for imputing data for businesses that have not yet returned their ONS survey responses.
Oh! That rather sends a chill down the spine as in essence we are back to fantasy numbers yet again and yet again they are in the housing sector. I am willing to give them a chance but can we really not get a grip on the actual numbers? Also I note that things in terms of actual measurement seem to be getting worse rather than better and the emphasis is mine.
Quarter 1 2018 is affected to a greater extent than Quarter 1 2017 due to the higher number of imputations in more recent periods due to lower response rates, as well as the inclusion of the bias adjustment.
In addition there has been a change to the seasonal adjustment which I take as an admittal of the problem I have highlighted before with the first quarter of the year which has been a serial underperformer. The combination of the changes has seen the beginning of the last two years revised up by 0.8% in construction terms so maybe this is some help with this issue.
Where are we now?
Let us take Kylie’s advice and Step Back in Time to 2016 about which we were told this.
The value of construction new work in Great Britain continued to rise in 2016, reaching its highest level on record at £99,266 million; driven by continued growth in the private sector.
Just for clarity this is far from all being new work as shown below.
Aside from all new work, all repair and maintenance equated to £52,223 million in 2016. This is an increase of £1,679 million compared with 2015.
There was a common factor in both new and maintenance work in 2016 in that the growth was essentially in the private-sector.
That number represented quite a boom. The nadir for the construction sector had been unsurprisingly in 2009 at the height of the credit crunch impact when output was £65.9 billion. Things got better but then there was something of a double-dip in 2012 when it fell back to £69.7 billion. As you can see from the 2016 number it was then a case off pretty much up,up and away from then.
The numbers above are in current prices rather than the usual deflated version which reminds us again that the deflator has been singing along to Lyndsey Buckingham.
I think I’m in trouble
I think I’m in trouble
Today’s update and if you like revisionism represents quite a change. Previously 2017 had seen the UK construction industry behave like one of those cartoon characters who are going so fast they do not spot the edge of a cliff but even when they go over it carry on briefly before they drop like a stone. On the road we were in a recession with flashes of a depression. Now we see that it was a year which opened very strongly but then slowed which is very different. Annual growth of 7.1% does not to say the least fit well with a depression scenario.
Now we see that we are being told the same for 2018.
Construction output continued its recent decline in the three-month on three-month series, falling by 3.4% in April 2018; the biggest fall seen in this series since August 2012.
Sound familiar? Well Kelis would offer this view.
Mght trick me once
I won’t let you trick me twice
No I won’t let you trick me twice
This really is quite a mess and regular readers will be aware it has been going on for some years. There was an attempt at an ongoing fix by “improving” the inflation measure called the deflator. Then there was an attempted “quick-fix” by switching a services company into construction. Plainly they did not work and frankly the idea of having these construction numbers as part of the monthly GDP numbers we get next week is embarassing. They are simply not up to it.
As to where we are now the Agents of the Bank of England offer a view.
Construction output remained little changed on a year ago, and contacts were cautious about the short-term outlook .
So now some 3% lower then? Also the Markit survey has its doubts.
June data revealed a solid expansion of overall
construction activity, underpinned by greater
residential work and a faster upturn in commercial
Indeed it was quite upbeat.
There were also positive signs regarding
the near-term outlook for growth, as signalled by the
strongest rise in new orders since May 2017 and the
largest upturn in input buying for two-and-a-half
So apologies for reporting official data which has turned out not to be worth the paper it was printed on. However strategically I think it is correct to follow the official data whilst also expressing doubts about systemic issues. Next week when we get the monthly GDP number we will return to a media bubble analysing each 0.1% which needs to be looked through the lens of a sector which has just been revised by 2,5%.