Last year public service pay cost the taxpayer £5.8 billion, broadly equivalent to the total revenue from income taxation. It absorbed half the Government's day-to-day spending power. Its impact therefore on the achievement of economic and social objectives, and on the quality of life for everyone in Ireland, is enormous.

It is fully accepted that a proper balance should be struck between expenditure and the provision of services; and that public service employees are remunerated fairly.

But allowing for those concerns, there is a growing level of disquiet about public service pay. The rate of increase in the public service pay bill is not sustainable into the future. The bill has doubled since 1987, at an average increase of 8 per cent per annum. This compares with an average increase in manufacturing earnings of 4 per cent and in inflation of 2.5 per cent. Public service pay determination since 1987 suffers from two major defects. The first is a clear failure to maintain the initial downward pressure on numbers employed in the public service established in the PNR. The second goes to the heart of the system, namely the issue of relativities. On the numbers issue, the figures speak for themselves.

Nobody wants the public service to stand still. On the contrary, there is a constant demand for new and better services. But these new/better services should be productivity-driven on the basis of constant review of the existing portfolio of services; their continued relevance; the manner of their delivery; and the acceptance of change as an ongoing requirement.

The feeling persists that the development of services in the public sector is incremental; building on what is already there rather than by deciding on the most effective utilisation of existing resources.

The issue of relativities is, however, the fundamental problem. For as long as the tradition of "What you get I must have" remains the mantra of public service claims; and for as long as Government accepts this, any real change in public service pay determination will be minuscule. Staying with the period since 1987, however, the evidence is clear. The ongoing restructuring provisions of successive agreements were added to with impressive determination by powerful public service groups and "legitimised" by reference to relativities. In turn the same agreements have gone largely undefended by successive governments.

The scant regard shown by key public service groups for the use of agreed procedures and the blatant use of the threat, and in some instances the reality, of industrial action are reprehensible. This is all the more so in those areas providing essential services to the public.

The way public service pay is managed has important implications for the overall economy. The Irish system of national partnership agreements and, in particular, the trade-off between pay moderation and tax concessions has been a key element in our economic success.

Continued pay moderation is essential if we want to sustain, and hopefully improve, our competitiveness. Competitiveness is the engine which drives economic growth. Once we start paying ourselves more than our competitors, we risk throwing away all that we have gained in recent years.

Pay moderation is as important in the public service as in the private sector. There have already been some indications that the developments in public service pay may be affecting the pay expectations of private-sector employees, where pressure already exists in a number of areas because of labour market tightness.

And it would be unrealistic to expect private-sector employees to accept continued wage moderation as part of any new national programme, which provides public service colleagues with the same opportunities they currently appear to enjoy to operate outside the norms of the agreement.

In addition to the headline effect which large public service pay increases have on the private sector, we also have to remember that the direct cost of such increases has to be met from current public expenditure.

The rules of the game have changed as a result of EMU. A prudent approach to economic management is, therefore, even more important. We must position ourselves to deal with the slowdown in growth which we will inevitably experience at some stage.

Many of us will remember only too clearly the public expenditure cutbacks of the late 1980s caused by the crisis in our public finances. The last thing we need is a repeat of that experience. Yet that is precisely what we risk if we continue to build the cost of large public service pay increases into our public expenditure base.

Regarding equity of treatment and performance-related pay for public servants, the evidence suggests that the remuneration of the main public service groups will stand up well in any such exercise. The concept of performance-related pay has real merits in terms of value for money and motivation of staff. It does not, however, necessarily guarantee a long-term solution to the problems that have dogged public service pay determination for decades.

The key issue in regard to any new system of public service pay determination remains securing the abandonment of the "relativity" system. Even if these (relativity) problems can be resolved, an oversimplified approach to performance-related pay could pose serious concerns, not only in terms of the direct costs involved and the urgent need to control them, but also in terms of what new structures might do to pay determination across the economy.

A properly structured performance-related pay system for the public service has a great deal of merit. But solutions must look beyond the concept of performance, and consider how market forces can be brought to bear on public service pay in a real and effective way.

John Dunne is director-general of the Irish Business and Employers' Confederation (IBEC)

Series concluded