Hamish McRae: Numbers that don't add up may mask a UK downturn

The numbers for the UK economy don't fit together. And when statistics don't add up you have to ask whether it is the stats or your analysis that is wrong.

The numbers for the UK economy don't fit together. And when statistics don't add up you have to ask whether it is the stats or your analysis that is wrong.

The problem is this. Unemployment is not only falling but falling at the fastest rate for two years. Employment is rising and is now at the highest level ever. But some sort of economic slowdown has clearly taken place. The National Institute for Economic and Social Research does monthly estimates of GDP and in the three months to the end of February the rate of growth had come down to 0.2 per cent. You can see a similar slowdown in retail sales in both the official monthly figures and the British Retail Consortium numbers (see left-hand graph).

Were house prices to fall – and we just had a hint yesterday from the Chartered Surveyors that they had come down for the first time for more than two years – a sharp fall in consumption must surely take place. (More than 6 per cent of consumption is being funded by borrowings, often secured against housing, and this is clearly unsustainable.)

If you look at the areas where jobs are increasing the puzzle deepens. Manufacturing, as is widely known, is shedding jobs relentlessly, 144,000 down over the past year. Most of us thought that the main impetus for hiring was the public sector, the result of the surge in spending – and borrowing – under Gordon Brown. That is certainly a strong feature and public sector wage growth of 5 per cent is running ahead of the private sector. But service industries are apparently still hiring too – that and construction accounts for more of the growth than the public sector. What's up?

There seem to be at least three possible explanations – well, four if you think the figures are wrong. It is always worth asking whether the figures might be wrong but in this instance they should be all right. Employment numbers are hard to fudge unless there is a sudden shift in the balance between legal jobs and those in the "informal" economy; and there is no evidence of that. And the two different sources of retail sales confirm each other. So let's take it that the figures are right.

Explanation one is that there is a lag, or rather several lags. The employment figures reflect what has been happening over the past three months and reflect decisions taken before that. Once a firm has decided to cut its workforce it takes several months to carry out its plans. So it could be that the employment numbers are lagging indicators, reflecting what was happening in company boardrooms six months ago. It may also be that part of the rise in public spending is being spent on buying services from the private sector. So part of the rise in private sector employment is actually the result of the rise in public spending.

Explanation two is that there is not as much of a slowdown in the economy as people in London and the South-east think. There is some evidence that London is in recession, with three of its big industries, tourism, finance and communications, going though hard times. But the rest of the country may still be showing decent growth and may continue to do so. It is quite possible that manufacturing may stop shedding jobs for a while. There is a close relationship between the changes in employment in manufacturing and the level of sterling (see right-hand graph), so as the pound has eased back, so we can expect that source of employment to steady itself.

Explanation three is that many employers are making the judgement that it is right to maintain their payrolls in the expectation that we have come most of the way through the downswing of this cycle and that faster growth is ahead. That would be quite rational, for the costs of hiring and firing are huge. The idea that companies can fine-tune the size of their workforce sounds neat but the real world is not like that.

There will be something in all these explanations but my instinct is that the first is the most important. Lags are tremendously important in economics but they are devils to understand. But as a general rule, economic responses seem to take longer than expected and then happen more violently. Nothing happens for ages and then everything moves at once.

The test comes next month with the Budget and the increases in National Insurance and council taxes. If they make a serious dent in both consumption and employment then we will have to think again about the hopes for growth this year. Last year the Treasury thought that growth in 2002 would have a two as the big number. Now we know it was a one: growth was 1.6 per cent. This year they still expect another two, though this may be shaded down when the new forecast is released at the time of the Budget next month. If it turns out to be one-point-something, the first thing that will go will be the Budget arithmetic. The next, I fear, will be employment levels.

So relish those employment (and unemployment) figures for they do reflect an economic success story. But they may turn out to be the high point of this cycle. Eventually growth and employment will recover but first there may well be a dip in both.