Stephen King: Disputed GDP figures probably are true picture of economy

Unemployment has remained low because wage growth has come down dramatically

A statistical storm is brewing. In one camp, we have the doughty employees of the Office for National Statistics (ONS), who seem to think that the UK economy stagnated in the first quarter. In the other camp, we have Anatole Kaletsky of The Times and many private-sector economists who cannot believe that the UK's performance in the first quarter could have been so pitiful, particularly relative to our European neighbours. As Mr Kaletsky put it a week ago "either the housing, retail and employment figures are false or the GDP reports are misleading".

Who is right? Given the number of times that official estimates of GDP are revised, we won't really know for some time. Nevertheless, we can at least make a rough stab to see whether the criticisms of the official data stack up.

For starters, the official data may not reflect changes in the quality of goods and services. This is a point well made by Mr Kaletsky, quoting work by Tim Congdon of Lombard Street Research. For the record, it's also a point made by John Butler, my colleague at HSBC, who argued at the beginning of the year that sterling's strength may simply be a reflection of our growing comparative advantage in services. Put simply, more foreign demand for UK services will push the exchange rate up. In turn, this will choke off the profitability of manufacturing – where we no longer have an overall comparative advantage – and divert resources to areas where we offer a more competitive package.

Statisticians cannot easily deal with this kind of change. The supposed advantage of manufactured goods, from a statistical point of view, is that you can count them. You can see the number of widgets being produced. With services, you cannot be sure. You don't know whether the increase in the value of a particular service is simply an increase in price or an increase in quality. Get the calculation wrong and you may end up overstating inflation and understating growth.

Certainly, this view is consistent with the relative stability of the current account deficit – as opposed to the trade in goods deficit – in recent years. Yet it is difficult to see the particular relevance of this argument to the first quarter of 2002. The shift from manufacturing towards services has been going on for decades. It is odd to think that all the measurement problems associated with this shift should have specifically fallen into one quarter and one quarter alone.

In any case, the argument is not specific to services. Faced with rapid technology changes, it is a difficulty that also now applies to manufacturing. It may well be that GDP is mis-measured but the distinction between manufacturing and services may be less helpful than might appear to be the case at first sight.

Mr Kaletsky's other arguments – which, to his credit, are more focused on the first quarter alone – are, at best, misleading. He suggests, for example, that the UK's problems effectively lie within the measurement of export and import volumes, pointing out that domestic demand rose 0.9 per cent in the first quarter relative to the final quarter of last year. Yet this is a highly selective use of data. Half of this increase represented a turnaround in the inventory cycle, as companies moved from destocking towards restocking. As every economist knows, this sort of thing is, in effect, a "one-off" and cannot be regarded as an ongoing source of economic growth.

Final domestic demand – consumer spending, government consumption, capital spending – rose a little over 0.4 per cent in the first quarter, hardly the stuff of economic success. And, within this overall increase, the two-tier domestic economy is alive and well. There's no doubt that consumer spending is very strong. It rose 0.7 per cent in the first quarter. The same is true for government consumption, up 0.9 per cent. Capital spending, however, was down 1.3 per cent. Of course, it may be that capital spending is also mis-measured but, given weak corporate profitability, a lack of decent productivity growth and heavy, but ultimately unrewarding, capital spending in the late 1990s, I somehow doubt it.

Other claims do not stack up terribly well. Mr Kaletsky suggests that the gap between the published nominal and real GDP growth rates over the last year or so implies an inflation rate of 3.6 per cent, totally inconsistent with measured inflation, whichever definition of the retail price index you care to use.

In one sense, he's right but there is a good explanation for the discrepancy. Given that GDP is the sum of domestic demand and exports less imports, it follows that falling import prices count as a "double negative", effectively adding to the GDP deflator. This sounds counter-intuitive but it simply reflects the way in which the numbers are compiled. It may mean that the low rate of RPI inflation is in part the result of declines in import prices, not a clear structural drop in the purely domestic rate of inflation. The Bank of England worries about the impact of a fall in the value of sterling precisely because of the evidence that some of our success with low inflation is the result of "imported deflation" stemming from sterling's earlier unexpected strength.

Mr Kaletsky then argues that "the stagnation of real output is inconsistent with the steady level of unemployment, which is a much more reliable statistic than real GDP". Well, obviously, it's relatively easy to count the number of people out of work (although constant changes to the definition of unemployment do not inspire a lot of confidence). The level of unemployment, however, says very little about the level of economic activity. What matters is overall spending power and there is no particular reason to think that this should be affected only by the number of people out of work.

Go back a year or so. Look at the typical forecasts for UK economic activity. At the time, the unemployment forecasts were too high relatively to the subsequent outcome. But so, too, were the forecasts for average earnings (or, in common parlance, wages). In effect, a rise in unemployment has been avoided only because the UK labour force, en masse, has been prepared to accept significantly lower wage increases – adjusted for inflation – than in the recent past. Two dominant forces have been at work. Job losses have been in more highly paid industries than job gains. And, perhaps reflecting a sign of greater wage flexibility, lower bonuses this year have significantly curtailed the overall growth rate of real wages.

In other words, the labour market has weakened but not in the traditional way. Unemployment has remained low because wage growth has come down dramatically. The decline in wage growth, in turn, is simply a reflection of the self-same problems within the corporate sector that have led to falling investment. Companies have been in trouble and consumers have spent only because the Bank of England and the broader financial community have persuaded them to borrow despite deteriorating wage growth.

None of these arguments, of course, guarantees that the ONS has got the story right for the first quarter of this year. There is always a case for examining and re-examining the available evidence. But some of the arguments put forward to dispute the latest published data do not stand up to close scrutiny. Productivity growth has been poor, investment has stagnated, company profits have suffered and consumers have borrowed. Whichever way you look at it, the UK is still not in the best of health, whatever the final estimate of first quarter GDP turns out to be.

Stephen King is managing director of economics at HSBC.