Construction output grew in cash terms in July, despite the expectations of many analysts that the figures would start to show a dip in the month.

And, by the way, the industry is actually 3.5% bigger than we thought it was a month ago, despite contractors making about £300 million less than we thought a month ago.

That is one interpretation of the main points to be extracted from output figures released today.

And if you think that is odd you should look a bit harder at the numbers.

It’s hard to know where to start with analysis of the construction output figures. Juggling jelly was never my forte.

But here goes.

Point 1.

Unless I am wrong the figures as published suggest that the crazy figure for construction growth in the second quarter of 8.6% (on a constant price seasonally adjusted basis) published last time has been revised even further upward to 9.6%.

To say I queried the previous figure is to be polite, having made an effort to defend – despite some qualms – an earlier estimate 6.6%.

But for ONS  to have revised the figure up yet again leaves me (obviously not speechless) but rather confused to put it mildly.

To put the 9.6% quarter-on-quarter growth in an historic context – assuming we can rely on historic figures that is – the only time we have seen growth of that magnitude is in the second quarter of 1963 after the yard or more thick snow that had been covering the country had finally melted.

Then on the same basis output grew 18.6%. The next largest growth rate, ignoring the latest quarter, was 6.3% in late 1982.

Now you’d expect a bit of a bounce back when the nation has come to a grinding halt as it did in 1963. And yes we had some nasty weather early this year, but nothing on the scale of the 1962/63 winter.

Yes, there was a fair amount of public sector money pumping up construction. Yes, housebuilders were a bit busier starting up rather than finishing off. Yes there was a twitch of activity in the private commercial sector and yes infrastructure was enjoying a bit of a boost.

But add this all together and you don’t easily get to the biggest gravity-defying surge in construction seen outside of a recovery from a paralysing winter. Well not to my mind anyway.

What will be worrying many economists – that is if those outside the industry ever delve into the construction figures – is that this suggests a further revision upward of gross domestic product (GDP) growth in the second quarter.

Now the original figure raised eyebrows at 1.1%. The upward revision to 1.2% raised those eyebrows a wrinkle higher, so a further lift will push them above most economists’ hairlines – not the bald ones obviously.

Point 2.

How do you measure the size of and industry and get it wrong by about 40%?

Well let’s look at the size of infrastructure in the second quarter of 2010 and compare the previous set of figure with the current set.

A month ago the volume of infrastructure output over the 12 months to June was £7.65 billion in 2005 prices. Today we discover it is worth £10.67 billion in 2005 prices.

Now the 2005 base is to provide a comparative volume measure. So what the statisticians appear to be saying is the amount of infrastructure work had been underrepresented by 39% in the figures.

But if we look at how much cash was spent (current price figures) we see that a month ago that was calculated to be £11.93 billion. Today’s calculation puts it at £11.86 billion.

Confused? I’m not surprised. A deflator problem apparently – the drop in prices was not as harsh as the earlier figures had suggested.

A technical problem this may be. But this kind of problem is causing all manner of problems for people who are trying to estimate where the industry is and where it is going.

More worrying is that if these figures are being used for policy making, the chances of making sensible policy – I am entitled to dream – is made all the more difficult. The wrong conclusions will be drawn.

I have had my problems with the output figures for some while, but they are the bedrock figures from which we can build an idea of how the industry is performing. They need to be right and they need to be trusted.

So come on Newport, we are behind you. We know you are under financial pressures, as is the rest of the economy. But even in these straightened times it must be well worth it to put a bit more resource into the construction output figures.

Ok you can argue that statistics provide the likes of me with little more than some room for joshing and that they provide politicians with the tools to be a second rate Machiavelli.

But they are very important to both the ordinary citizen and the ordinary business – they provide tools that are used to shape our futures. They are worth investing in.