The scale of the rise in Irish unemployment over the past two years has been surprisingly modest, particularly over recent months. The unemployment rate was unchanged at 4.5 per cent over the first quarter of 2003, only marginally higher than the 4.3 per cent average in 2002 and still remarkably close to the lows of 3.6 per cent recorded in spring 2001.

Most forecasters had expected a more serious deterioration in the labour market and this still remains the case. The Central Bank, for example, is forecasting an average unemployment rate of 5.75 per cent for 2003 and the Department of Finance expects a 5.3 per cent figure.

The better-than-expected unemployment outturn in the year to date may also be a surprise to the public at large, given the frequency and scale of the job losses announced in the media over the winter months. Yet job losses are only part of the picture, as the unemployment figure is also affected by changes in the workforce as well as trends in employment.

Unfortunately information on these additional factors is not available on a timely basis so it is still not clear why the unemployment rate has stabilised in recent months, and whether this is temporary or not.

The unemployment rate is derived from the quarterly Household Survey data and is defined as the total unemployed as a percentage of the labour force. The latest figures available relate to December last year, which showed 84,000 unemployed and a labour force of 1.86 million. In the intervals between surveys, the monthly unemployment rate is calculated by using the trend in the live register to estimate the unemployment figure. So the live register does play a role in the unemployment calculation, even though it is a claimant count (that is, it is made up of persons claiming Benefit and Assistance and, therefore, includes some part-time workers) and "not designed to measure unemployment", in the words of the Central Statistics Office.

The monthly live register total is still rising but at a much slower pace than 2001 or the first half of 2002. Over the first three months of this year, the rise in the seasonally adjusted total was 4,800 or 1,600 a month, which compares with a 3,700 average monthly gain in the first quarter of 2002. The average weekly flow of new registrations has also slowed to 5,000 in March compared with 6,000 in the corresponding month of 2002.

Indeed the former is not too dissimilar to the 4,400 weekly average inflow recorded in March 2001, when the unemployment rate reached its cyclical low.

The issue then is whether this softening in the rate of new registrations reflects a diminution in the rate of job losses, a pick-up in employment or whether labour force growth has slowed further.

Certainly the published redundancy figures offer little comfort (redundancies in the first quarter were up 9 per cent on the 2002 figure) but the number of publicly announced job losses does seem to have declined in recent weeks. Moreover, the number of announced job gains appears to have risen, although one has to be careful about the small print in such announcements - all too often the headline figure reflects expected employment gains over a number of years. Employment in construction also appears to have stabilised, again after a period of decline, with February recording marginal gains on an annual basis.

It is also true that capital expenditure on computers and related investment in technology has risen in recent months in the US, breaking a two-year cycle of decline, which holds out the hope that the worst may be over for the Irish tech sector, which bore the brunt of the manufacturing job losses in the recent global downturn.

Yet the most recent data on income tax receipts offers little support for any notion that employment growth is accelerating (revenue in March was more than 6.5 per cent down on the same period of 2002) and other figures point to a fall in hours worked, so a more likely explanation for the relative stability in the unemployment rate is that the slower pace of growth in the labour force, evident from 2001, has continued. Some 30,000 people joined the labour force in 2002, down from 46,000 in the previous year, and the 2003 outcome may be lower still, at 25,000 or less, limiting the damage done to the labour market by the tepid scale of job gains.

The most likely scenario still remains that the unemployment rate will drift higher over the course of the year but probably not at the pace envisaged by most forecasters, who may have overestimated the scale of job losses and underestimated the growth in the labour force.

What is undoubtedly true is that the economy has weathered the international downturn remarkably well, as evidenced by the labour market. The unemployment rate was more than 10 per cent as recently as 1997 and the rise over the past two years still leaves the unemployment rate at the level recorded in early 1999 at the height of the economic boom.

Dr Dan McLaughlin is chief economist at Bank of Ireland