Economics: While jobs in manufacturing have been disappearing, employment in service industries continues to rise, writes Jim O'Leary.

Job losses have been hitting the headlines of late, accentuating the gloom that has come to pervade commentary on the economy.

In one recent week, almost 1,000 jobs were declared lost in two manufacturing operations (3Com and Schneider), which really set off the alarm bells.

But, with all due sympathy to those affected, job losses are nothing new. Even in the halcyon days of the late 1990s, some companies were closing down and others cutting back.

In 1999, for example, the year in which the Celtic Tiger produced the unprecedented rate of 11 per cent growth in GDP, almost 10,000 jobs were lost in IDA-supported firms alone, and God knows how many more elsewhere in the economy.

Job losses per se are not the issue, at least in macro terms. At one level of abstraction, they are simply a manifestation of structural change in a dynamic economy.

The issue at the moment is the rate at which jobs are being shed. In this regard, two things distinguish the past couple of years from the 1990s.

First, the rate of job loss has accelerated significantly because of a combination of weaker demand for Irish output and diminished competitiveness of Irish producers.

Second, for similar reasons, the rate at which new jobs are being created has slowed sharply. Thus, whereas in the 1990s job gains exceeded job losses, resulting in a net employment increase, since 2001 the position has reversed, with the number of jobs exceeding the job gains and resulting in a net employment fall.

However, this is true only of the manufacturing industry (and agriculture). Over the past two years, according to the Central Statistics Office's (CSO's) Quarterly National Household Survey, manufacturing employment has declined by 15,000, or around 5 per cent, but employment in construction, and in every area of the services sector for which data are published, has continued to rise.

Granted, the rates of increase in the private sector have slowed appreciably, but some of them have been quite respectable: employment in hotels and restaurants is up 10 per cent over the two-year period, while financial and other business services are up 4 per cent.

What's been happening to employment in manufacturing has not been representative of the economy as a whole.

In this connection it is worth making a point of wider relevance. Manufacturing gets disproportionate attention from analysts and commentators, and there is a tendency to use the manufacturing sector as a barometer of economy-wide conditions.

One reason for this is that there are far more indicator data published for manufacturing (which accounts for just 17 per cent of total employment in the economy) than for the services sector (66 per cent of total employment).

Therefore, the CSO publishes detailed data on industrial production, industrial turnover and producer prices in industry on a monthly basis; but there are no equivalent data published for the services sector.

Also on a monthly basis, the CSO publishes detailed data on merchandise trade, almost all of which is trade in manufactured products and none of which relates to services.

However, it is clear from what information is available, and not just the employment data, that economic activity has remained a good deal stronger in the services sector than in manufacturing in recent times, and that it would have been highly misleading to extrapolate from manufacturing to the economy as a whole.

At the highest level of aggregation, for example, the latest national accounts data show that services sector output grew by 2.7-3.9 per cent in the 12-month period to the first quarter of this year while industrial output grew by just 1 per cent.

Almost certainly the latter figure seriously exaggerates the strength of manufacturing (i) because it includes construction and (ii) because of the well-known measurement issues that arise from transfer pricing by multinationals.

Other, arguably more meaningful data, better illustrate the contrasting fortunes of manufacturing and private sector services.

The Purchasing Managers' Indices published by NCB show a pronounced dichotomy between the two. In the case of manufacturing, the overall index of activity has been below 50 (indicating decline) every month for the past 12 months except September; in the case of services, the equivalent index has been above 50 (indicating expansion) every month for the past 12 months except March and April.

The latest balance of payments data, released earlier this week, are especially revealing of the divergence between manufacturing and services. They show that the value of merchandise exports (almost all manufacturing) in the first half of this year was down 20 per cent on the first half of 2002, but that the value of services exports increased by 6 per cent over the same period.

In summary, the available evidence suggests that the manufacturing and private services sectors of the economy have been on divergent paths for the past couple of years, with the former contracting and the latter expanding at a modest but respectable pace. It might be argued that this cannot be sustained on the grounds that the services sector provides inputs into manufacturing and, as such, requires a buoyant manufacturing sector to prosper.

However, this sort of argument derives from a naive and anachronistic model of economic activity. I can think of no good reason why the Republic cannot, over time, transform itself from a mainly goods-exporting economy into a mainly services-exporting economy, in which case manufacturing will certainly shrink further in relative terms (and may well shrink further in absolute terms too).

Indeed, given our endowment of human capital and the level to which our cost base has risen, this is what must happen if the economy is to achieve its productive potential over the long run.

Of course, recent trends in economic activity have a strong cyclical element, but it is also useful to place them in this sort of long-term perspective.

Jim O'Leary is currently lecturing in economics at NUI-Maynooth. He can be contacted at jim.oleary@may.ie