Making sense of Ireland s public-private pay gap

Transcription

1 Making sense of Ireland s public-private pay gap IMPACT Communications Unit 27th August 2014

2 Introduction The gap between average earnings in the public and private sectors is back in the news, but the claims and counter-claims are baffling to most. For example, yesterday s Irish Times claimed that the pay gap was widening, with public servants now earning even more than their private sector colleagues. But IMPACT said the public-private pay gap had narrowed. Who was right? The answer is both. Working from the same set of CSO statistics, the Irish Times found a widening of the gap because it selected figures from one three-month period, while IMPACT identified a narrowing of the gap by looking at the full set of CSO figures for a year. This example demonstrates the perils of taking the headlines at face value. The reality is that economists, statisticians and other researchers disagree about both the size and the significance of the public-private pay gap. Media headlines always exaggerate the size of the gap and the articles below them seldom attempt to explain the reasons why it exists. In the recent past, this has fuelled calls for public service pay cuts. These days, it s being used to argue against pay restoration not just in the public service but across the economy. IMPACT believes that the difference between pay in the public and private sector is an important matter of public and policy interest. It s not rocket science - but, equally, it s not as simple as the media headlines always suggest. This short briefing looks at some factors at play. We hope it will help policy-makers and the public make sense of the public-private pay gap - and see that it s not nearly as significant as the headlines would usually have us believe.

3 Is there a public-private pay gap in Ireland? The simple answer is yes, just as there s a gap between average pay in the IT and retail sectors. But there are good explanations for them both. Comparing public and private sector pay and earnings is not straightforward. Economists and others disagree on how to measure differences across the two sectors. When assessing comparisons between public and private sector pay, care is required to understand exactly what is - and is not - being compared. Virtually all studies and statistics quoted in this contentious debate (including the CSO s quarterly earnings figures) are comparisons of average earnings in two highly varied sectors. The CSO s quarterly earnings figures are not weighted in any way to take account of the actual jobs that people do, qualifications, educational attainment, age or other factors that explain differences in earnings. In fairness to the CSO, they make that crystal clear in their publications, but this is seldom referenced in media coverage and commentary. Studies that dig beneath the raw data to find explanations for the gap in average public and private earnings - taking account of qualifications, job type, experience and other factors invariably reduce the pay gap substantially. But, again, economists disagree on what factors should be taken into account, and this inevitably effects their conclusions on the size and significance of the gap. Explaining the pay gap Imagine a hospital located next to a department store. Both employ, say, 200 people. It s likely that 80% or more of the hospital staff are graduate professionals with high levels of personal and professional responsibility at work - doctors, nurses, physiotherapists, radi-ographers, pharmacists, social workers and so on. These are relatively well-qualified well-paid professions. The staff in the department store do important work too, and carry it out professionally. But only a small proportion of them are likely to be professionally qualified or even graduates. It s likely that there will be a very wide gap between the average earnings in the hospital and the department store. But there s a good explanation for it. The same would be true of a secondary school located near a hotel, or a forensic science lab across the road from a meat factory. Conversely, staff in a private sector pharmaceutical laboratory would be paid far more on average than the workers in the nearby (public service) street cleaning depot. But across the economy public servants are - on average - far more likely than private sector workers to be in professional roles with graduate qualifications. One major explanation for the pay gap. The most comprehensive analysis of Irish public and private sector earnings 1 in recent years was produced by the CSO in October The study acknowledged that economists disagree on how to measure differences across the two sectors. So it published results from a variety of models and tests. It found a very broad range of estimates, all of which demonstrate a measurable reduction in the pay gap regardless of the methodology used. 1 Central Statistics Office, National Employment Survey 2009 and 2010 Supplementary Analysis, 12th October This is the most recent CSO analysis of public and private sector earnings. It is understood that a similar analysis may be carried out in 2014.

4 Depending on how the gap is measured, the CSO report found that the pre-pension levy public service pay premium ranged from 6.1% to 18.9%, but that the gap was significantly different for men and women. The pre-pension levy pay premium for men ranged from 2.3% to 16%, depending on how the gap is measured. The pay premium was wider for women at between 9.2% and 21.5%. Factoring in an average pension levy of 7.5%, the net public service premium gap ranged from minus 1.4% to plus 11.4% overall. For men, it was minus 5.2% to plus 8.5%. For women it ranged from plus 1.7% to plus 14%. There are virtually no studies based on direct comparisons of the pay of people doing the same or similar work (or jobs of a similar size in terms of roles, responsibilities, value, and educational or professional qualification) in the respective sectors. This is partly because wide differences in pay and reward structures in different organisations (public and private) make such direct comparisons extremely difficult. Instead, most comparisons are of average public-private wage differentials. These can also vary dramatically as a result of (i) the method of analysis (ii) the data sample chosen for analysis (iii) how researchers decide to adjust for bias in data samples and (iv) whether and how researchers account for relevant pay determinants like qualifications, educational attainment, experience, age, or size of enterprises. Factoring in an average pension levy of 7.5%, the net public service premium gap ranges from minus 1.4% to plus 11.4% overall. For men, it was minus 5.2% to plus 8.5%. For women it ranged from plus 1.7% to plus 14%. One of the most contentious questions in the debate has been whether or not to weight pay data to take account of the size of the workplace. This is significant because workers in most large employments - whether public or private - generally earn significantly more than workers in small or medium enterprises. And it makes a huge difference to the outcome of public-private pay comparisons in Ireland, where most private sector workers work in small enterprises and most public servants work in large organisations.

5 The public service pension levy The most-quoted figures in the public debate - the CSO s quarterly earnings figures - do not take account of the so-called pension levy, which cut public service earnings by 7.5% on average. Clearly, the pay gap is significantly reduced if you take account of this additional tax on public servants, which was introduced in 2009 on top of the normal pension contributions of 6.5% that public servants pay. Pay and pay movements It is also important to distinguish comparisons of pay rates (how much people are paid) from comparisons in pay movements (how much their pay has increased or decreased) over a given period. Both are of interest, but they are different things. For example, if worker A earns half as much as comparable worker B, and their pay increased by 5% more than worker B s last year, they are still underpaid in relation to worker B. Like pay rates, the significance of pay movements should be assessed in light of the time periods under comparison. For example, a comparison of pay movements in the year to January 2013 will not capture the fact that public service pay rates were reduced in January 2010 (or that pay for new entrants was reduced again in 2011) as these pay movements will have been washed out of the figures. Neither will it capture additional pay cuts for some public servants which came into force in July Pay structures Pay structures can have similar effects. Incremental pay scales are far more common in the public than the private sector. As a result, statistics can show pay increases even after pay cuts have been imposed. If my pay was cut substantially in January 2010 and I received a small increment in February 2011, statistics will show that my pay rose in 2011, even if it remained far below the January 2010 rate. Increases and decreases in overtime and premium rates, and the availability of overtime work, can also have a significant impact on earnings in both sectors. The payment of bonuses and performance-related rewards, which are far more common in the private than the public sector, can have similar effects. International comparisons Unsurprisingly, international comparisons are even more difficult to make as pay and reward structures, the size and content of jobs (including jobs which, on the face of it, seem directly comparable), and tax and benefits systems vary even more widely. Crucially, most international studies do not take account of the cost of living which, in Ireland, remains 15% higher than the EU average.

6 Research conducted for IMPACT also shows that certain large groups of Irish public servants - including clerical officers and primary teachers - are paid less than their German counterparts at every stage of their careers, even though the cost of living was 17% higher here at the time the research was conducted 2. This study directly compared the pay of certain public service grades in Ireland and Germany. It did not take account of an agreement that has since increased German public service wages by 6.3%. In September 2012, the IMF reported that the Irish public service pay bill was equivalent to 11.2% of GDP compared to an average of 11.1% for OECD countries who are members of the EU 3. These figures do not take account of the so-called pension levy which reduces Irish public servants pay by an average of 7.5%. In other words, these figures show that the Irish public service pay bill is roughly in line with comparable EU countries as a percentage of GDP even before you deduct the so-called pension levy. The OECD s Government at a Glance, 2011, which compares international public sector labour costs, found only two groups in Ireland - hospital consultants and top central government managers - were paid above international standards 4. Both have experienced significant pay reductions since the study was published. In general, the OECD says that the cost of employing Irish public servants is about average. For further information contact info@impact.ie 2 Ciarán Lyng, A comparison of public service pay rates in Germany and Ireland, IMPACT, August Note, this relates to the total pay bill and is not a measure of pay rates alone. The pay bill is determined by the number of public servants and by their pay rates. 4 Government at a Glance 2011, OECD, 24th June 2011.This is the most recent issue of the OECD s Government at a Glance that includes data for Ireland. Pay for Irish hospital consultants and senior civil servants has fallen again since this survey was undertaken.