Monday, November 21, 2011

Public Sector Pay Once Again

This is old ground but somehow we keep having to go over it.  Yesterday’s Sunday Independent features an incredible article from James Fitzsimons, who “is an independent financial adviser specialising in tax and financial planning”.  It is incredible because it contains so many errors.  It is nearly difficult to know where to begin.

THE Government that lives in La-La Land has just created Disneyland for 300,000 public servants.

There are around 350,000 public servants in Ireland.  The breakdown of these is provided in Table A2 of the Earnings and Labour Costs Survey from the CSO (see page 15).

The 300,000 number that is commonly used is the number of full-time equivalent public servants.  Staff employed on a part-time basis and those engaged in job-sharing are only included on a pro-rata basis to get to the 300,000 figure.  The latest figure from the Department of Public Expenditure and Reform is that there were 302,769.7 full-time equivalents in the public sector in Q2 2011.

It is proposed to cut 37,500 jobs by 2015. This is 11.72 per cent of the 2008 figures when they were at their peak. Mr Howlin claims the gross public sector pay bill will fall by €2.5bn.

Based on CSO figures, the average earnings are €901.07 per week in the public sector. The savings for a cut of 37,500 jobs would be about €1.76bn a year. If 10 per cent of earnings constituted tax, the real saving would be less than €1.6bn.

When working out the reduction in the gross public sector pay bill compared to 2008, Fitzsimons focuses solely on the 37,500 reduction in headcount and completely overlooks  the average 7% pay cut for all public servants that was announced in the December 2009 Budget.  The reduction in the gross pay bill from this measure was almost €750 million.  Once this is included it is clear the the €2.5 billion claim from Brendan Howlin is not that wide of the mark.

Fitzsimons then goes on to calculate some measure of “net savings”.  This time he seems to omit the impact of the Public Sector Pension Levy introduced in April 2009.  This levy had no impact on gross pay but did reduce net pay (by around €1 billion).

It is incredible that someone “specialising in tax” would suggest that 10% of earnings constitute tax.  The average earnings figure he uses (which we will return to) is equivalent to an annual income of just under €47,000.  Using a fairly simple tax calculator it can be seen that the net pay for a public sector worker is €30,600.  Of the €16,400 of deductions €2,200 is a pension contribution so the tax on this average worker is €14,200.  That is 30% of gross pay.  Why is a figure of 10% used in the article?

According to the CSO average pay in the public sector is €901.07 compared to €611.88 in the private sector.  We have previously considered the impact our progressive tax system has on the 47% differential.

It is also important to note that the average of €901.07 provided by the CSO is based on around 400,000 public sector workers.  This is because it includes just over 50,000 employees in semi-state companies.  These are not paid from government resources and have not been subject to the pay cuts, levies and other changes introduced in the public sector.

I have asked the CSO to provide a breakdown of average wages in the public sector under the same headings used in Table A2 in the Earnings and Labour Costs release.  I have been told that this figure is not available.  The €901.07 average is likely higher because of pay in the semi-states.

The McCarthy Report on state assets reports (Table 4.4, page 23) that average pay in December 2009 for the 40,000 employees in the semi-states considered under the terms of reference of the report was €54,600 or €1,050 per week.  If we apply this to the CSO data it means that average weekly pay for the public servants in the total is around €880.

A 2% difference may not seem that significant it would be useful if the correct figure was available.  I have made further requests to the CSO but nothing has been forthcoming.

If the average pay in the public service were brought in line with the private sector, it would be cut by 30 per cent. This is without taking account of the cost of public sector pensions. It would achieve annual savings of €5bn to €6bn. We need this now, not in 2015 or 2020.

We can get some crude measures using The Analysis of Exchequer Pay and Pensions 2006-2011 which gives total pay and employment numbers across a number of categories.  The measure of gross pay given here includes salaries, employers’ PRSI and employers’ pension contributions.  The figures given here are for 270,000 public servants paid from the Exchequer.  The table excludes about 30,000 local authority employees.  The total public sector pay bill will be around 10% higher than the figures given here.

Exchequer Pay

In 2008 the gross Exchequer pay bill was €17.7 billion.  In 2011 it will be €16.2 billion.  There has been a drop of €1.5 billion or 8%.  Brendan Howlin is projecting a  further fall of €1 billion and a total drop of 14% on the 2008 level.

Exchequer net pay is calculated by subtracting employee pension contributions (€534 million in 2010), the public sector pension levy (€916 million) and some other minor appropriations-in-aid (€59 million).  Exchequer net pay has fallen from €17.1 billion in 2008 to €14.8 billion in 2011, a drop of 14%.  The projected fall to 2015 will be around 20%.

If a measure of net pay was provided to account for Income Tax, the changes to income tax would mean that the fall in net pay from the employees’ perspective would be greater than the 14% shown above.  We previously estimated this to be 17% for a public sector worker earning the average wage published by the CSO.

Fitzsimons argues that public sector pay should be cut by 30% now.  There is no evidence to support the claim that this will bring it “in line with the private sector”.  A 30% reduction in the 2011 gross pay bill would be €4.9 billion (or around €5.5 billion if local authorities are included).  It is claimed that this would “achieve annual savings of €5bn to €6bn”.

The actual saving would be much lower.  When questioning Howlin’s €2.5 billion figure at the start of the article Fitzsimons argues that the actual saving  would be lower because of lower tax revenue.  However when it comes to his own figure of €5-€6 billion he makes no such adjustments.

The cut in gross pay would reduce employee pension contributions (estimated c. €0.3 billion) , pension levy receipts (c. €0.3 billion) and income tax receipts (c. €1.25 billion).  Of course, there would also be loss of VAT and Excise Duty receipts as the pay reduction reduces consumption expenditure. 

The actual saving from a 30% cut in average pay in the public sector would actually be closer to €3 billion and not the €5-€6 billion claimed in the article.  If we went the whole hog and cut public sector pay by 100% the actual saving would be around €10 billion.  The deficit in 2011 will be €16 billion.  Public sector pay is not the only reason for the deficit so cannot be expected to be the only solution.

Public servants might have to suffer the same increases in VAT as the rest of us, but they have 50 per cent more income to cover the cost.

People pay VAT from net pay not gross pay.  Public servants have a higher gross pay than private sector workers but the effect of a progressive tax system and the Public Sector Pension Levy substantially narrow the gap.  The Pension Levy is not a pension contribution but is just a pay deduction.  This was made clear by Brendan Howlin in the Dail a few weeks ago.

The Trident report assumes that the pension related deduction, commonly called the public service pension levy, is a pension contribution. This is mistaken and the law could not be clearer. Section 7(2) of the Financial Emergency Measures in the Public Interest Act 2009 states: “(2) A deduction under section 2 is not a pension contribution for the purposes of the Pensions Act 1990”. The pension levy contribution is a misnomer. It was called that by the previous Government, but it is a levy on pay.

I hope it will not be a permanent feature, as I said to the unions when I met them. In my judgment, it is mistake for unions to characterise it as a pension contribution because the fear will be at a future date that it will be subsumed into the calculation of pension contributions. Under the auspices of the Financial Measures in the Public Interest Act 2009, it is not, by definition, a permanent measure. I hope it will not be a permanent measure, but, obviously for the foreseeable future, it is required.

Maybe it is too much to expect a tax specialist to know the difference between gross pay and net pay.  To finish here is the conclusion to my previous post on public sector pay.

This doesn’t mean that public sector pay should not continue to fall by more than private sector pay. A staggered wage cut from 0% for those at the bottom (with possibly even some increases) rising to maybe a cut of 12.5% through to the top of payscale (and maybe even more targeted than that), along with forthcoming general changes in income tax could bring the average fall in net pay in the public sector to 25% (with most of the latter burden shared by those above the average wage). 

A 25% cut in net pay would be a huge contribution from the public sector but what needs to be remembered is that with a 17% cut in net pay we are already two-thirds of the way there.  At times this seems to get forgotten in the ongoing pay debate.

And that is a 25% reduction in average net pay for those who are in the public sector.  By 2015 there will also have been a 10% reduction in the number of full-time equivalent employees in the public sector.


  1. It is clear where Mr Fitzsimons is getting the 300,000 PS employee numbers from, it is from the report on PS reform published by Brendan Howlin on 17th Nov 2011.

    "this year we are expecting a reduction of some 5,000 in staff numbers across the Public Service to bring the numbers below 300,000 by the end of 2011. New high level targets for each sector will be set out at Budget time. In total, we expect to achieve a planned reduction of 37,500 staff to 282,500 by 2015, from a peak of 320,000 in 2008."

  2. Another very interesting post Seamus.

    But I wonder about your proposal about a progressive cut in public service pay.

    Estimates of the public service pay premium in 2007 showed that it was highest at the lower end of the wage distribution and fell away to very little at the top:

    Both pension levy (2009) and pay cut (2020) have already been quite progressive so if anything I suspect this effect has been even more pronounced.

  3. Seamus,

    Just in relation to the pay figure, over 65% of Public Servants are now insurable and liable to Social Insurance at Class A. Even among Senior Management grades, the figure is now I gather over 40%. The Employer PRSI paid by the Public Service must be in the region of €700M at least by now. The Social Insurance fund has been heavily subsidised since 1995 by Public Service PRSI contributions as they are unlikely to be made unemployed and very few have reached retirement age. You don't appear to have factored in the E'er PRSI contributions in your calculations.

    The points made by Anon eile are also valid. The pay premium is substantial for clerical posts requiring little no specialist qualifications. However this is also an issue with other large employers of clerical staff, e.g. banks and insurance companies. I would suggest that the normal premium that applies to working in a trade union organised environment explains the discrepancy.

    Technical & professional staff within the Public Sector never struck me as excessively paid and to my knowledge it was hard to recruit for many positions until recently. Indeed in the most recent competitions into the Revenue approx two years ago, almost all of those recruited were female attracted by the "family friendly" work conditions.

    The only substantial group who seemed over rewarded were perhaps teachers and the Union premium certainly applies there, but of course not at Third level!

    Benchmarking Two proposed substantial pay rises (around 13%, if I remember correctly) for school principals and Nurse managers. I have in the past 12 months sat on 7 interview boards for school principal and deputy principal posts. In only one case was there any actual real competition for the post. In four cases there was only one applicant, which suggests that the rewards do not match the workload.

    I am advised that management level nursing posts are similarly hard to fill.

  4. I'm a little confused as to why you consider the pension levy a pay cut, when at the same time you compare PS weekly pay to private weekly pay, where one includes a DB pension and the other doesn't. Surely this distorts all comparisons

  5. Benchmarking Public Service Pay: A Flawed Comparison.

    Qualitative Benchmarking of Public Sector Employees with Private Sector Employees by comparing roles and responsibilities, competencies etc, provides credible comparisons, leading to the conclusion that similar pay scales are warranted.

    However, Quantitative Benchmarking of outputs of functions, processes, projects etc in the private sector with public equivalents would result in a different conclusion.

    Both Qualitative AND Quantitative Benchmarking are required to achieve a true comparison.

    Quantitative Benchmarking compares the macro pay cost for functions, processes, projects etc but when divided by the number of employees required in the Public Sector would set very different individual employee pay scales.

    This is the Engineering Economics approach rather than the simple Economics approach to Benchmarking.

    Applying Qualitative Benchmarking only, results in inflated macro public pay costs when aggregated from individual qualitative benchmarking, leading to too high a tax burden on the primary wealth creating private sector.

    Borrowing (€19 Billion pa) to bridge the difference over what would be a sustainable public service cost is daft!

    We now need €1000 car tax, €1500 house tax, €400 water tax and septic tank tax to pay for a wrongly Benchmarked Public Sector.

    We need properly benchmarked public service pay (Q&Q Benchmarked), consistent with the ability to pay of the private sector tax payer.

    The existing "Croke" Park agreement will probably have to be revisited.

  6. Seamus

    Agree with your final conclusion of a graduated pay cut but with a much more substantial cut at upper/top levels.

    Further I would add that this cut should not be 'saved'. It would be much better to use that saving to recruit at junior levels and specialist grades.
    My theory is that it would be better socially and economically and provide better levels of service to have more people working than fewer particularly as they would be paid from savings on higher remuneration levels.