The publication of the first National Employment Survey, providing details of pay rates, as of March 2003, in the private and public sectors, was bound to spark off controversy. All the more so because, by chance, this publication appeared at the precise moment when the current national agreement talks were beginning to address the issue of pay in both of these sectors, writes Garret FitzGerald
The detail in this publication that grabbed most headlines was that the average hourly pay rates of public sector workers was 40 per cent higher than the average figure for the private sector.
But that is quite misleading, for these averages do not compare like with like.
It might have been beyond the CSO's brief to have made this point. And given that the data relates to a period three years ago - hopefully its 2006 survey will appear closer to the time it is carried out - it may have seemed inappropriate to accompany the statistics with a press briefing. But the consequence is that the public is left with a false impression of the public-private pay relationship.
The fact is that twice as many public sector workers have third-level degrees than in the private sector, and more than four times as many - 55 per cent compared to 12 per cent - are classified as professional. Moreover, public sector workers have one-third longer employment experience than the average private sector worker.
Such an accumulation of differential factors clearly contributes to the higher rates of pay in the public sector. How much of the apparent 40 per cent differential is due to such factors, and how much reflects a bias in favour of public sector workers?
Light is thrown on this issue by an authoritative study by three NUI Maynooth economists, on the basis of sample data collected for the European Commission by the ESRI, and published two years ago in the ESRI's Quarterly Economic Commentary. The Maynooth economists' sophisticated analysis of public and private pay rates between 1994 and 2001 showed that when full account is taken of differences in the educational and social compositions of the two workforces, on a like-with-like basis, monthly earnings in 2001 in the public sector were on average about 13 per cent higher than in the private sector.
However, that 13 per cent pay differential of 2001 must subsequently have been substantially increased by the 2002 benchmarking process, which boosted public service pay by an average of 9 per cent.
This study also showed that this disparity varied considerably between different levels of pay. It was greatest - 17 per cent - in the case of lower-paid workers, and was least marked in the case of the top level of the public service. Indeed the pay rates of the top 5 per cent of public servants is at much the same level as for similar jobs in the private sector.
However, even the Maynooth economists' study does not tell the whole story, because it covered only monthly earnings - and last week's CSO data shows that, although men in the public service work slightly longer hours than men in the private sector, women in the public sector work shorter hours - probably because they make greater use of job-sharing. So in terms of pay per hour, the public/private pay differential is somewhat greater than suggested by the Maynooth study.
How did the 2001 monthly earnings differential of 13 per cent compare with the situation at around that time in other European countries? Notably, it was two to three times greater than the equivalent differential in the UK, France or Italy. This presumably reflects the exceptional strength of public service unions in Ireland and the relative weakness of unions in our private sector.
Public officials enjoy other benefits not available to private sector workers. These include much greater job security, and pensions linked to the current level of earnings of officials in the pensioner's retirement grade. By contrast, the occupational pensions enjoyed by a minority of private sector workers are often only defined-contribution schemes. Civil servants, but not other public service workers, also enjoy a reduced rate of PRSI.
It will be clear from this that on a like-with-like basis, public servants are in general better paid. But the 40 per cent differential suggested by the media is clearly misleading: the real figure is much lower than that.
If I had to put a figure on it, after taking into account these quite complex factors (including the benefits of benchmarking), and leaving aside the issues of security and pensions, public servants' hourly pay rates are probably around 20 per cent higher.
The question may well be asked: if, in 2001, on a like-for-like basis, public service monthly earnings were already some 13 per cent higher than in the private sector, why did the Government agree to initiate a benchmarking process designed to compensate public service workers for being less well paid than similarly placed private sector workers?
No one has ever tried to answer that question, nor has the basis for the public service pay increases awarded by that process ever been published.
The benchmarking report even failed to claim that its conclusions were based on public/private earnings differentials - although that, of course, is what it had been set up to evaluate!
All we know is that the economist member of the benchmarking team, (one of the authors of the Maynooth study), honourably resigned from it in protest at the non-transparent procedure being adopted to justify these increases.
Astonishingly, a second bench-marking of public service pay rates has now been initiated.
Unless this new exercise is based on a rigorous approach, akin to that adopted by the Maynooth economists, but also incorporating factors they did not include in their study (such as differences in hours worked and in the quite different character of the pension arrangements in the two sectors), this new exercise could merely aggravate further the marked disparity that already exists between the public and private sectors of our economy.