Comment/Bill Roche: The first payments of the benchmarking awards to public service workers have stirred many of arguments and controversies first aired this time last year, with some new objections emerging from critics.

In assessing the merits of benchmarking, the following issues may deserve some consideration.

The broad approach to public service pay determination involved in benchmarking compares the pay of groups and categories of public service workers with workers in the private sector doing work that is comparable with respect to the demands, skills, training, responsibility etc involved.

In one way or another, this principle has become the settled basis for pay determination in public services internationally. Critics of this approach have failed to identify any plausible alternative basis for public service pay determination.

The superficially plausible alternative that has been widely canvassed in the Irish debate is to base the pay of individual public servants on their contribution to performance. Even if individual performance pay were to be introduced, there would remain the problem of how to determine basic rates of pay in the public service - the problem addressed by benchmarking.

But what advocates of individual performance pay fail to acknowledge, or have not taken the time to discover, is that the international experience with this form of pay in public services reveals it to be deeply problematic, commonly ineffective and not uncommonly counterproductive.

Perhaps the best illustration of this is the recommendation to the British government in 2000 in a report - the so-called Makinson Report - that says individual performance pay systems in the UK public service should be dispensed with and replaced by group performance incentives. Far from being the work of mandarin Whitehall insiders, this report was drafted by a senior executive from a major private sector company. Experience in other OECD countries is no more encouraging.

What other alternatives might there be to benchmarking? Hardly a return to the shambles of "special" pay claims, coupled with tightly interlocking internal relativities, that benchmarking attempted to address. If we take seriously the commonly avowed injunction that the Republic requires a modern, dynamic and highly trained public service, rates of pay in the public sector will need to be comparable to those in the private sector for work of comparable character and demands.

This is not to say that there are no problems inherent in the way benchmarking was implemented. It is standard in such exercises internationally to be almost ritualistic with respect to providing for openness and transparency in pay comparisons, and in the principles and methods on which they are based.

In the case of the State, the exercise followed the opposite extreme, and the reasons given for this - confidentiality with respect to the identities of private sector pay comparators e.t.c. - have been wholly unconvincing.

Systems for comparing public and private sector pay can also be designed with considerable flexibility in mind: for example, the parties could pre-agree that public service pay might be set somewhere between the top 25 per cent and the bottom 25 per cent of rates for comparable groups in the private sector - a concept known as the "inter-quartile range" and used in some parts of the British public service.

This leaves the parties free to negotiate between this floor and ceiling for different grades and categories, taking account of other principles like recruitment and retention, major restructuring, ability to pay and - if an effective system can be devised - performance.

The search for flexible and transparent mechanisms for public service pay determination is the major trend across developed economies. If benchmarking is to provide the basis for a lasting system of public service pay determination in Ireland, design issues will need to be addressed and, above all, transparency will need to be greatly improved.

A strong case could be made for the view that the real problem in public service pay determination in the Republic is not benchmarking but the dismal track record of public service management in harvesting the kinds of productivity changes expected in return for payment of the benchmarking awards.

Attempts under previous national agreements to conclude productivity-type agreements, linked to restructuring proposals of various kinds covering public service categories, commonly ended disastrously - pay awards dramatically overshooting budgeted targets, with little real gain in the quality or efficiency of public service delivery.

Nor are public service managers wholly to blame for this state of affairs.

The reality is that their political masters and mistresses have shown little real conviction with respect to public service modernisation, especially when determined action was required to forge major changes in work practices and related arrangements.

There is a presumption that many of the changes now being sought should be handled through the various workplace partnership arrangements that have developed across the public service.

If the parties agree that this should be a major channel for delivering change and modernisation, partnership in public service workplaces will be put to the test in earnest for the first time.

While solid foundations have been established in this area in local government and parts of the health service, progress in establishing partnerships in other areas has been patchy, and in some areas remains partnership at this level is virtually non-existent.

So the jury is out on whether a sea change will occur in the delivery of productivity and change. If it does not, many of the high-flown aspirations of public service modernisation will count for nothing.

Finally, some commentators have called for the non-payment of benchmarking awards in the current fiscal climate.

Leaving aside the widely respected principle of honouring pay deals, which counts in the private as well as the public sector, in anything other than conditions of deep fiscal crisis, such a posture is likely to result in widespread industrial conflict and disruption across the public service. It would also likely sink both the current social partnership agreement and the social partnership process more generally. The consequences of this would reverberate across the economy as a whole.

Bill Roche is professor of industrial relations and human resources at University College Dublin