It’s hard not to get caught up in the silly guessing game over whether the nation will tumble into a triple-dip recession or not.
Yes it is totemic. But actually measuring growth to an accuracy of 0.1% is pretty tricky and revisions over time can eliminate or even reverse growth rates.
The reality is that growth is very weak, if there is any, at the moment and that is horrible, especially for construction where its growth requires at least modest growth in GDP (see blog).
But what the heck. Let’s speculate. I’ll use as my excuse the rather disturbing latest construction data (see blog).
You already have the answer from the headline. But, importantly, how did we frame the question?
More importantly, is it meaningful?
Let’s narrow the game to guessing the March figure on the back of what we have, which is the January and February figures, and using the estimated quarterly figure as a marker for whether GDP is likely to go up or down in the coming quarter using a respected forecast for GDP as a benchmark.
Part of my prompt for this examination was a chat with Noble Francis at the Construction Products Association. We are pretty much agreed that there is a huge amount of uncertainty over where the March figure might lie. But we both seem to agree that as a whole the uncertainties are on the down side.
Bad weather events, revisions, changes to deflators, new data, workload-mix effects, the effect of the irritating lag between work happening and being recorded in the ONS data, bounce-back effects, underlying growth, variable seasonal factors such as the effect of a mobile Easter or leap years, one-off effects and many other variables can have a striking impact on assumptions made about the direction of construction activity. Gauging the effect of these is tricky. Gauging how the ONS will account for them, if and when they do, adds a further layer of trickiness.
To this list of uncertainties we need to add the behaviour of firms and local authorities in the run up to the end of the financial year. Will firms and local authorities act consistently with past behaviour or will the surge at the end of the financial year increase or decrease in our age of austerity?
The best thing to get a starting point is to see what happens if we put to one side all the above and see what would happen based on the latest figures and the historical data.
So if we start with the volume based data on table 3 of the ONS Construction Output spreadsheet we can see that on average over the past three years March workload was about 17.5% greater than the monthly average over January and February, although last year the rise was about 14.5%.
On the basis of the limited experiences over the past three years we might then expect March to come in at £8.3 billion at 2005 constant prices. Although you might fairly argue that as the industry is in decline rather than growth it would be lower. But let’s stick with the £8.3 billion.
This would put the first quarter constant price output at about £22.5 billion which is 6.5% lower than the first quarter of 2012.
Turning to table 2, the seasonally adjusted data, and reducing 2012 Q1 data by 6.5% leads to a figure of £23.5 billion for 2013 Q1. Here we are assuming no change in the effect of seasonal adjustments.
If that is where the figure comes in, we would have seen a fall from the final quarter of 2012 to the first quarter of this year of 3.4%.
Now there will be revisions to the earlier data, seasonal adjustments are mobile things, we have to assume behaviour is pretty similar to the past etc, etc, etc.
Indeed from my chat with Noble Francis we think you could make a reasonable case for a growth rate in 2013 Q1 of anything between -1% and -6%, given the huge level of uncertainty. Both of us would tend towards the lower end if pressed.
But let’s make the assumption that all other things remain equal and let’s stick to the -3.4% fall in first quarter construction output and see how this compares with the predictions made earlier last week from the well-regarded independent research organisation NIESR for first quarter growth.
NIESR estimated that GDP grew by 0.1% in the three months to March. Its latest forecast saw construction falling, on my calculations, by about 1.6% in the first quarter. This was admittedly produced before the latest and rather dismal construction output figures were released, so is not surprisingly very much at the top end of our range of expectations.
If we crudely account for the extra fall we see in construction above that currently forecast by NIESR the net effect it to lower GDP first quarter growth by about 0.12%, based on adjusting the 2009 industry weighting of 68 out of 1,000 parts of GDP.
So given that NIESR predicts 0.1% growth (and we assume that from the index it is below rather than above 0.12%) the data used in this crude fashion suggests we will be very very narrowly below zero growth. That is making the massive assumption that all the assumptions made by NIESR and all the assumptions made here are reasonable and (importantly) that reality doesn’t interfere and lead to a more positive outcome.
I would add that if anyone bothered enough to check my maths finds any glitches would they please let me know and I can adjust the headline accordingly.
Anyway, our little guessing game comes up tantalisingly close to zero, but shades on the side of a triple dip. Which we knew from the headline.
There is clearly political mileage that can be made from whether the nation is in a triple dip or not and there’s some fun to be had coming up with predictions that put GDP growth either side of the highly politically sensitive zero GDP growth mark.
But let’s be honest the real value of determining which side of zero growth falls is limited.
The truth is that growth is pitifully poor whatever assumptions we make. Tha is damaging the nation and, more parochially, pushing away hopes of a much needed revival in construction.