Most markets work when allowed to and labour markets are no exception. Interference in the working of markets therefore, needs to be justified by compelling evidence of market failure. Where such evidence does not exist, markets should be left to work freely; where it does, policy should be directed towards the removal of the causes of market failure where possible.

When national wage agreements started again in the late 1980s there was compelling evidence of failure in Irish labour markets. The unemployment rate was 17.5 per cent in 1987, the year that launched the Programme for National Recovery. The proximate reason for this was that wages had been growing too fast.

Accordingly, moderation in the rate of growth in wages was needed to solve unemployment. Many commentators believe that social partnership facilitated that moderation, although there are grounds for disputing this proposition.

Whatever the contribution of social partnership in the period since 1987, circumstances have changed enormously. The unemployment rate has fallen to 5 per cent. Evidence of generalised labour-market failure has greatly diminished and, with it the justification for interference in the wage bargaining process.

Although unemployment can fall a good deal further (several states and large metropolitan areas in the US have jobless rates as low as 2 per cent), the predominant problem today is a shortage of labour, not excess supply.

Where there is a shortage of labour, the rate of wage increase typically accelerates. This process has been under way in the Irish economy for a year or more.

In the first half of last year, average hourly earnings were already rising by 5.25 per cent in manufacturing industry and indicators point to broadly similar rates of increase in other sectors.

The deal unveiled earlier this week does little more than validate the recent rate of wage inflation. It provides for a cumulative increase of 15.75 per cent over the coming 33 months, equivalent to an annual average rise of 5.5 per cent. Will the deal be inflationary? The answer is that it will be no more inflationary than wage increases would be in its absence, if the labour remained as tight as it were now, or tightened.

Indeed, wage increases have already started to exert unmistakable upward pressure on consumer prices. The services component of the CPI is inflating at a rate of almost 6 per cent year-on-year, a clear reflection of labour cost growth.

Will the deal erode the competitiveness of the economy? The answer here is the same as above. The competitiveness of the economy is already being eroded by labour shortages, although this is being masked by currency depreciation.

The new wage deal will not quicken the pace of competitiveness loss relative to what would have happened in its absence, again provided that the labour market remains tight.

In any case the loss of competitiveness is not a prospect that should be contemplated with the same dread at this juncture as it was when the unemployment rate was in double digits.

Losing competitiveness now is part of the process whereby the economy slows down to a pace that is sustainable over the medium to long term.

Wage increases of 5.5 per cent-plus per annum are entirely appropriate to an economy operating as close to full capacity as the Irish economy currently is. But they would not be appropriate if the economy were to be hit by a sharp external shock that abruptly cut the growth of output and employment. For example a shock such as an international recession or a big drop in the value of sterling.

This is a risk that is inherent in multi-annual wage agreements. I hope the employers have insisted on a clause being inserted into the agreement that offers some insulation against this risk.

All my remarks so far have related to the private sector. The public sector is a different animal. The problem here is not the size of the overall wage and salary bill or the fact that, under the terms of the new pay deal, it will grow at least as fast over the coming three years as it has in the recent past.

After all, if we are to have education, health and public administration systems that are capable of delivering high standards, teachers, doctors, nurses and public servants generally must be well paid. Moreover, the requirement that they be well paid is all the more pressing when labour markets are tight and pay levels in the private sector are rising strongly.

But, while attractive levels of pay are necessary for the delivery of high standards of services across the public sector, they are not sufficient. Public-sector workers, no less than their private-sector counterparts, need a system of sanctions and incentives if the highest standards of performance are to be maintained. In this regard, it has to be said that the operation of public-sector labour markets is so heavily circumscribed that they can scarcely be described as markets at all. At the point of hiring and firing, and at all points in between, the efficiency of publicsector labour markets is greatly diminished by regulations and restrictions. This cannot but reduce standards of service. It is difficult to see who wins from all of this. It is clearly not the taxpayer.

Probably the greatest single challenge now facing policy-makers is the reform of public-sector labour markets to make them more efficient. This has got to involve much more flexibility in respect of recruitment, tenure and dismissal.

It should also involve performance measurement and a system that links performance to remuneration. Some modest progress has been made in these areas in recent years, and some more will probably follow after this week's deal. For me, that's where the merit of the new agreement stands or falls.

Jim O'Leary is chief economist with Davy Stockbrokers