It has long been said that construction is an important bellwether in determining the shape of the nation’s economic progress.

Today the performance of construction, or rather revisions to its measured performance, seemingly determined how close the UK is to recovery.

The upwardly revised construction data put the nation yet closer to its official exit from recession.

Whether we are in or out of recession has huge political significance.

But for me it is interesting to note just how much store is put by the construction figures, partly because I am hugely sceptical of their accuracy.

Here is why.

The construction figures plugged into the latest calculation of GDP suggest that the industry saw growth in both the second and third quarter of this year.

Fascinating that construction should have been enjoying growth of 0.6% and 1.9% and be cutting jobs at an average rate of more than 1,000 for each working day of that period. It is possible, but...

Furthermore, we have seen savage cuts to the prices on offer to construction’s clients from contractors.

Why would these things be happening if the industry had enjoyed two quarters of consecutive growth?

What do I think? I am not sure construction did enjoy growth and I wouldn’t be the only construction number watcher to share that view, I just enjoy a bit more freedom to say it.

It is intensely tricky to measure construction and even more so when prices and volumes are fluctuating wildly, as they are at the moment. So I put little blame on the overworked statisticians who have to make sense of this complex industry.

For the record I have suspicions over the deflators used to convert cash to volume measures and again I am not the only one with such concerns. These are extremely tricky to get right, yet their accuracy and timing has a huge bearing on the level of the calculated volume of construction output.

But for all that, how amusing to watch non-construction sector economists and policy makers putting such meaning to construction numbers that may well be very wide of the mark.

Interestingly, the production method of the construction output figures is in the near future to be radically revised. This is a tricky time to do such a thing.

Let us hope that the statisticians are gifted the resources they need to do this job properly.

And hopefully we will be spared the uncomfortable spectacle of very smart economists and policy makers trying to put meaning to figures that may be, to put it bluntly, wrong.