Lies, damn lies and the UK’s statistics
BRITAIN’S economy shrank even more than previously thought during the recession – that, at least, is what the Office for National Statistics informed us yesterday and what we are therefore expected to believe. The economy is thought to have collapsed 6.4 per cent from peak to trough, a figure that will probably go on being revised for years (previously, the estimate was a contraction of 6.2 per cent).
What I find deeply irritating is the way the City, the media and politicians alike all take every pronouncement from our official statisticians as akin to the gospel – even though the figures keep being changed, often dramatically so. The 0.3 per cent growth in the first quarter went unrevised but crucial details were unrecognisable. Until yesterday, we were told that investment spending had gone up by 1.5 per cent – yet yesterday this was changed radically to 4.6 per cent. The new figures show that government consumption went up 1.5 per cent in the first quarter, three times faster than previously thought, prompting many to argue that this meant that the economy would have shrunk without the public sector. There are two flaws with this reasoning – the first is that the figures could easily be revised again; the second is that the whole point of cutting spending is to ensure that it no longer crowds out the private sector, allowing it to grow faster.
Exports of goods and services fell back by 1.7 per cent on the quarter; again, this is a big change. One driver was new estimates for inflation of exported and imported goods – of course, these new numbers may be right but I do wonder. I once spent ages poring through ONS data on economic growth and comparing changes – the findings were astonishing. Economic history was completely revised over and over again at the stroke of a statistician’s pen; the timing, duration and depth of recessions in particular tends to be modified for years afterwards (and even more depressingly, the definition of GDP has changed several times over the past 40 years, injecting even greater confusion into the picture).
The figures leave GDP 5.7 per cent below its peak, Investec calculates. By contrast, the US is just 1.3 per cent smaller than it was at its peak. This is an implausibly large gap given that the UK labour market appears to have been more resilient than the US. Another oddity is that UK public finance numbers have been outperforming for months (though they remain grim). All very strange.
So what next? Ruth Lea, a director of Arbuthnot Banking Group, has crunched the numbers. If the Budget forecast is to be realised the private sector will have to grow at around five per cent a year on average. There are precedents for this. After the early 1990s recession, government spending was tight but the economy grew strongly, as a weaker sterling helped drive exports. Few jobs were created by the public sector, but private employment grew strongly. There is another precedent: over 1m jobs were lost in 1929/30 (a large figure in those days). But by 1934 employment had surpassed its 1929 peak, even if unemployment also remained high.
None of this changes my forecast that we are in for years of weak growth (but no double-dip). But one other thing is clear: beware the statistical spin-merchants who claim to know more about the minutiae of the economy than they really do. Excessive precision is impossible in economics, as yesterday’s data shows all too well.
allister.heath@cityam.com