There are several reasons for this change in perception. One is that the rise in house prices, always a lead indicator of UK consumption, seems to have shifted up a gear in the first months of this year. Another is the clear evidence of a rise in the rate of increase in earnings. And while the GDP figures do not give any cause for concern, some good work by Lehman Brothers demonstrates they may be wrong. Michael Dicks, an economist, looked at the way in which initial estimates of GDP have consistently understated what happened. Just as the statisticians always seem to find that invisible earnings are higher than they first thought, so there always seems to have been more economic activity than the numbers initially catch.
If this is right, the implication would seem to be that not only is the economy growing much faster than was previously thought: it is also closer to full capacity. It follows that the new government will have to tighten policy quickly and since it cannot, for political reasons, do much on the fiscal side, monetary policy will have to bear the burden.
There is a lot in this argument. The core of the problem is that while we have good figures on the manufacturing side of the economy, we have relatively poor ones on the service side. Yet services are much larger, and therefore more important. You can see some of the strain in service sector prices. Until this time last year the rise in UK service sector prices was much the same as those of France and Germany (left-hand graph). But since then UK service prices have climbed steadily. The rise in sterling, which is holding down the prices of manufactured goods, does not help much on the services side.
The rise in service sector inflation seems to have been associated with a surge in service sector demand. The best leading indicator of the pressure of demand in manufacturing is the purchasing managers index. Since last July the authorities have been collecting data for the service sector too and the first results were published a fortnight ago. The data is split into different categories but Goldman Sachs has calculated a single index from this and compared it with the manufacturing side (right-hand graph). As you can see, since the beginning of this year there seems to have been a sudden spurt in services, while manufacturing has remained flat. This is further evidence that growth is shifting up a gear at the moment.
So the figures are catching up with what is actually happening and the conclusion that interest rates will have to rise, maybe quite sharply, after the election is reinforced.
But that is the short-term, market response. Financial markets sleep, drink and eat interest rates. The wider issue, and it is one which we will hear much more of during the next five years, is whether our predominantly service economy can increase its output at a faster rate than in the past, and if so, how?
Capacity is probably a more flexible concept in services than in manufacturing. The lags are certainly shorter, for the unit size of investments, and hence the lead time in getting investment into service, is shorter. So it should be possible for service industries to respond to increased demand more easily than manufacturing. A rise in demand for services, too, is less likely to suck in imports, because a lot of services cannot be imported.
That means that if there is increased demand which cannot be met by increased output, it has to show in higher prices rather than increased imports. During the late 1980s boom, soaring imports acted as a buffer, checking the rise in prices of domestic goods that would otherwise have taken place. But that does not answer the capacity questions posed above. There almost certainly is no simple, single answer. It is possible, though, to sketch some parts of a complex one. We know that, provided we approach capacity slowly, we are more likely to be able to stretch it. A slow fall in unemployment is much less likely to lead to a sudden rise in pay rates than a rapid one.
We know too that it takes quite a long time for increased investment in service industries to lead to increased productivity.
One of the great puzzles of the past 10 years has been the tiny return that seems to have been generated by all the investment that companies have made in information technology. Part of the explanation may be that the investment has resulted in better quality of service and this is not caught in the output stats. But part of the explanation may simply be that it takes a while for new investment in any industry to improve productivity.
Our economy is going through a very rapid structural change. Whole new industries are springing up, while existing ones are transforming their way of business. I think it is perfectly possible that we will see radical improvements in productivity of service industries, which will enable some increase in the "natural" rate of growth of the economy.
But I doubt whether it is possible to do much to foster that growth, for that has to be a bottom-up, market-driven process. The best thing the authorities can do is to provide a stable macro-economic environment which comes back to that next rise in interest rates.