ECONOMICS: The Consumer Price Index is based on the average spending pattern and those who deviate from this often deem its results inaccurate, writes Jim O'Leary

Prices and inflation have been very much in the news in recent weeks. One detects a growing sense of unease about the inflation climate and a growing suspicion that the euro changeover has been exploited for profiteering.

One also detects growing scepticism about the accuracy of official measures of inflation, the most important of which is the Central Statistics Office's Consumer Price Index (CPI). Amongst economists, one increasingly hears the view that inflation of around 5 per cent has now become ingrained in the economy.

How valid are these perceptions? Well, probably the easiest one to deal with is the one about the accuracy of the CPI. Frankly, it is very difficult to believe that an index that is so comprehensive in coverage and so carefully constructed can be seriously inaccurate in terms of what it measures. But it is important to understand what this is: it is a measure of average price movements in respect of a representative basket of consumer goods and services.

To the extent that an individual's spending pattern deviates from the average, the CPI will not necessarily reflect the rate of inflation experienced by them. I suspect that this is the reason why the CPI can acquire a reputation for inaccuracy amongst individuals.

What about the impact of the euro changeover? Well, the evidence of profiteering suggested by the pattern of movements in the CPI is limited. Granted, at an aggregate level, prices have risen faster over the last six months for which data are available (November to May) than over the equivalent six-month period a year earlier.

Excluding the effect of mortgage interest rates, the figures are 3.6 per cent compared with 2.3 per cent. But most of this differential can be attributed to energy prices: between November 2000 and May 2001 energy prices fell by almost 7 per cent; between November 2001 and May 2002 they rose by almost 10 per cent.

Service sector inflation provides a bit more food for thought. Between November 2000 and May 2001 the average price level of services rose by 2.5 per cent. The corresponding increase over the period from November 2001 to May 2002 was 4.2 per cent.

The suspicion that this might be related to the euro changeover is strengthened by the fact that much of the increase was concentrated in the months of December and January.

Still, there have clearly been other factors at work, including the hike in insurance costs in the aftermath of September 11th. Moreover, while the euro changeover may have boosted service prices to some degree, the effect of this on the overall CPI has obviously been diluted.

What about the notion that 5 per cent inflation has now become ingrained in the system and will become correspondingly difficult to dislodge? Well, one's answer to this question is very much dependent on the model of Irish inflation that one works with. For me, the model that makes most sense starts by distinguishing between goods and services that are internationally traded and those sheltered from international competition.

For the first category, which comprises about 70 per cent of the CPI, inflation is largely determined by what's happening to prices in the rest of the world and by exchange rate movements. For the second category, comprising some 30 per cent of the CPI, inflation is largely determined by what's happening to domestic costs of production.

At the moment, the rate of inflation in respect of traded goods and services is running at about 3.5 per cent year on year, while the inflation rate in respect of the non-traded category is running at about 9.5 per cent, together yielding an overall rate (excluding mortgages) of 5.3 per cent.

Looking ahead, the likelihood is that the rate of inflation in the traded sector will decline.

As far as the sheltered sector is concerned, the critical influence will be the evolution of wage costs. Already, there is some evidence that private-sector wage inflation has started to decelerate in response to a loosening of labour market conditions. If this continues, expect the rate of price inflation in the non-traded areas of the economy to follow suit.

In this scheme of things, the notion of an inflation rate of 5 per cent becoming entrenched in the economy does not apply. Likewise, the notion of a wage-price spiral. However, none of this is to be taken as reason to be indifferent to inflation or to take the view that it simply doesn't matter. At the microeconomic level, it clearly matters a lot - to people on fixed incomes for example. At a macroeconomic level, the fact that real interest rates are substantially negative because of high inflation matters, and is almost certainly a contributory factor to overheating in the housing market.

And high rates of wage inflation matter because they carry with them the risk that a negative shock to the economy, in the shape of a very sharp exchange rate appreciation, for example, will cause greater loss of output and employment than would otherwise occur.

Jim O'Leary is lecturing in the economics department at NUI Maynooth.