You are here: Home > About Smurfit > News > Faculty Views

Cutting public sector pay: an alternative view

Date: 06 Nov 2009

Published: Industrial Relations News

November 4th 2009

There is no single, best measure of the “public sector wage premium”, which varies over time and across occupations. This means that the basis on which public sector pay might be cut “is not as sound as some people claim on the basis of headline figures”, according to UCD Professor, John Geary and Oxford University economist, Dr Anthony Murphy.

The issue of public sector pay and its determination is a matter of considerable controversy. The government has stated that large cuts in public expenditure, including public sector pay, are necessary to stabilize the government deficit. The justification for pay cuts in the public sector is primarily based on the belief that public sector workers enjoy substantially superior wages in comparison to their counterparts in the private sector.

Hard evidence for a public sector “wage premium” is found in recent research from the ESRI. ESRI researchers have found that in October 2006 full time permanent public sector workers aged 25 to 59 enjoyed a “wage premium” of about 26% on average over their counterparts in the private sector, an increase of 12 percentage points on the average March 2003 premium of 14%. In the current difficult economic circumstances, this report has been followed by calls from economists both within the ESRI and elsewhere for across the board public sector wage cuts of the order of 5% to 10%. We are inclined to be more circumspect and cautious about across-the-board public sector wage cuts. Alongside the public service pension levy which was introduced in April 2009, public sector wages could fall by 17% on average in money terms, without any corresponding large fall in private sector wages.

We make this contribution not as “hurlers on the ditch”, but as researchers who have also tried to measure the size of the public sector wage premium. Indeed, the ESRI use the same statistical procedures as we have employed in recent years together with the same data sets made available by the CSO. Our analysis reveals broadly similar findings – on average, public sector workers are paid more than private sector employees, but not 26% more. The latest CSO National Employment data suggest that the average public sector premium in October 2007 was about 17%, but this average masks a huge amount of variation by gender and occupation, etc.

LIKE WITH LIKE?

In reply to the ESRI’s findings, the public sector trade unions have argued that the ESRI’s research does not take proper account of the differences in the work undertaken by public and private sector workers. They claim like is not being compared with like. For example, they ask, what is the equivalent to a primary school teacher in the private sector, and how might one reasonably compare the work of An Garda Síochána with the work performed by private sector security firms?

In an effort to respond to these objections, the ESRI has used even more sophisticated statistical techniques (propensity score matching) and has claimed that it is possible to compare “like-with-like”.

We feel their confidence is misplaced. While it is our view that the CSO data does provide an accurate portrayal of wage levels and wage movements in the economy, to claim that the use of propensity score matching and more detailed (2 digit) occupation codes allows one to make precise comparisons between jobs of ostensibly similar skill profiles and responsibilities (i.e. “like-with-like”) is not correct. The job descriptors used by the CSO are simply not rich and refined enough to permit such an analysis across all occupations, nor could they reasonably be expected to be, given the form of survey research.

In other countries, public sector wage rates are established on the basis of analyses conducted by independent pay review bodies. Earnings surveys, of the type conducted by the CSO, submissions from stakeholders, consultations with unions and employers as well as job evaluation surveys form part of the assessment process.

ROSE AND FELL

The econometric modelling which we have conducted shows that the public sector earnings premium or discount depends on the choice of comparison group and varies a lot both across occupation groups and over time. For example, the average estimated premium is reduced by about 5% if the earnings of public sector employees are compared to those of private sector employees in large firms. The average estimated premium falls by about 3% if employees in personal and protective services (e.g. gardaí and prison officers) are excluded from the analysis. If semi-state bodies are excluded from the public sector, the average estimated premium falls by at least 2%. Moreover, as noted above, the estimated average public sector premium rose substantially between 2003 and 2006, and then fell between 2006 and 2007. This indicates that, with just four years survey data on individual earnings in the public and private sector, one needs to be careful about inferring medium term trends in the overall public sector premium.

Our point is therefore this: the CSO datasets are an important resource and should be used to inform public policy making, but not to the exclusion of other research methodologies and inputs.

A second reason for caution is that the ESRI’s research is necessarily limited. It does not examine pay movements in the wider economy and, as such, its analysis of changes in the public sector premium is incomplete. For example, consider the large inflow of migrant workers from the new member states (NMS) and elsewhere, many of whom are likely to return home in the next few years. Between 2002 and 2006 the number of NMS nationals increased from 20,000 to 126,000. The vast bulk of these workers gained employment in the private sector, particularly in manufacturing, construction, wholesale/retail trade and hotels and catering. Relatively few gained employment in the public sector.

IMPACT OF MIGRANTS

What was the effect of this massive influx of migrants on the labour market and wage patterns? We have tried to estimate the size of this effect on private sector wage levels. This is not an easy exercise, nor can it be said to be without limitations given the available data. However, to the extent we have been able to pin down this effect, we found that wage rates did not increase as quickly in those parts of the private sector where migrants were employed in significant numbers. Further, our preliminary results suggest that a rise in the share of migrants in an occupation group is associated with reduced earnings. Indeed, a 5 percentage point rise in the share of migrants in a given sector may have reduced earnings by up to one and a quarter percent. Thus it would seem that where migrant workers gained employment in large numbers in the private sector they acted as a brake on wage growth.

It is not that we decry such developments for they might be fairly claimed to be necessary to improve Irish industry’s competitiveness, as long as decent standards of employment are maintained and fair wages are paid. However, there is exploitation of workers, many of whom are migrants. Such exploitation is difficult to capture in surveys of earnings administered to employers, yet it is apparent from evidence arising from workplace inspections carried out by the National Employment Rights Authority. Breaches of employment law were found to highest in hotel and catering, retail, construction, contract cleaning and security. Such exploitation is less likely in the public sector, not only because of the protection offered by trade unions, but because of the state’s ethos of maintaining a model of good employment.

We do not suggest that proponents of large across-the-board wage cuts in the public sector are seeking to drive down standards of employment. But a full discussion of wage levels and movements in the Irish economy, does beg certain questions which will not be easy to resolve: by what standards will wages be set, at what level will ‘harmonisation’ be sought, and by what means will wages be determined?

The latter question is critically important. The danger, as we see it, is that if the ESRI’s results are taken as sacrosanct and as legitimate grounds for imposing large wage cuts in the public sector, there will be considerable industrial strife.

PARTNERSHIP AT LOW EBB

At the time of writing, the prospect of reaching an agreed and peaceful solution is poor. For over twenty years, social partnership has provided a forum for dialogue on wage negotiations between the social partners. One important outcome of these deliberations has been the significant decline in levels of industrial conflict, such as strikes, to record-low levels. But the shine has gone out of partnership, not only for economists - many of whom identify partnership as being part of the problem and therefore cannot now be part of any viable solution - but also for many trade unionists. Many unionist leaders, who were proponents of partnership, have grown increasingly disillusioned with the failure, as they see, of the government to engage in meaningful talks. They argue that there is little to which they might point to as success, and by which they might defend partnership, so as to prevail upon their colleagues to be patient and to strive for a solution through “jaw-jaw” rather than “war-war”. Critics of social partnership have gained significant ground.

The risks could not be more serious for all sides. If the government insists on imposing large cuts in public wage on the basis of broad brush, headline figures, and in the absence of meaningful dialogue and negotiation, it is certain there will be industrial conflict of a form and bitterness we have not witnessed for many decades. Membership disaffection is running high. Witness, for example, the mandate received by union leaders in the largest public sector union, IMPACT, to pursue industrial action – 86% of the membership has voted in favour. Compare this to the failure of the same union’s leadership to receive a mandate to pursue industrial action in opposition to the public service pension levy in the Spring. Other unions have also gained majority support from their members to pursue industrial action.

UNION UNITY AT RISK

But there are enormous risks for unions as well. Despite the best efforts of union representatives to portray the government’s response to the economic crisis as an attack on all workers – private and public – it is clear that such unity will be very difficult to maintain. There is also a danger of divisions between public sector unions. The 24/7 Alliance group of unions (gardaí, nurses, ambulance drivers etc., who provide round-the-clock public services) are seeking to preserve their pay, a large portion of which is made up of allowance and premia payments. Other public sector workers do not rely on such additional payments. If the government seeks to preserve base pay but reduce allowance payments, the former group risk significant reductions in their standards of living. In these circumstances, it will thus be exceedingly difficult for senior union leaders to navigate towards a solution while preserving union unity.

These are very delicate moments in Irish industrial relations. What is needed is a voice of fairness and judgment. But one thing is certainly clear, a move on public sector pay which involves government by fiat and justified on the basis of headline research findings which are very open to debate, will result in a return to industrial strife and sectionalism, with huge costs both in terms of the country’s economic fortunes and social cohesion.

In summary, we argue:

The risk of imposing wage cuts across the board in the public sector is that it will result in industrial strife with huge costs both in terms of the country’s economic fortunes and social cohesion.

There is no single, best measure of the public sector wage premium. The estimated premium varies over time and across occupations. It also varies across the income distribution and at the upper end of the distribution, is often a discount. It is also difficult to find good "like-for-like" comparison groups for some public sector occupations. Thus, the basis on which public sector pay might be cut is not as sound as some people claim on the basis of headline figures.

Setting the correct ‘bar’ for public sector pay is a difficult one, and can be influenced by many factors including poor standards of employment in the private sector – so we have to find some way of factoring this into our assessment.

John Geary is Professor of Industrial Relations at UCD Business School and Dr Anthony Murphy is an economist and Fellow of Hertford College, University of Oxford.

back to Faculty Views

Cutting public sector pay: an alternative view

MEET OUR STUDENTS

MEET THE PROGRAMME COORDINATOR