My Photo

Blog powered by Typepad


« Shock! Horror! McCarthy Vindicates Public Sector Efficiency! 20th July: The Recession Diaries | Main | Opinion = Fact = The Whole Truth?: 24th July The Recession Diaries »

July 22, 2009



Sorry Michael -

To my mind, it is the enterprise-by-size comparison.

but this is nonsense. Salaries at Microsoft aren't related to salaries at Walmart, salaries in my local newsagent aren't related to salaries at the Eclectica hedge fund.

You can't just pick an arbitrary feature, totally unrelated to worker productivity (number of employees) and decide to benchmark against it just because it helps produce the statistical result you desire. At a minimum, it just doesn't make any sense - in truth it's a statistical trick.


I think this is your strongest argument

Cutting public sector wages will have almost no effect on our public finances

The corollary is that raising taxes will have almost no effect on the size of our public finances. (Opposite sides of the same coin).

Though surely they will have an affect on the public finances - the amount of debt the government will have to take on would be less in absolute terms? It's just the size of the deficit and national debt as a percentage of GDP that barely change.

Can we borrow our way out of this recession?

Or is it tax raises versus pay cuts?

Michael Taft

Mack –the points you’ve made seem to me to confirm the point. Rather than putting all the numbers together and start comparing, its best to examine, first, within sectors and then define that range which makes for the best of comparison. That’s why you can’t compare local newsagent salaries with those paid in hedge funds. Nor with those paid in the public sector. Nor, even with larger enterprise units. Take managerial structures. The local newsaget’s is fairly minimal. In larger enterprises you can have a range of managerial functions: Area, Store, Assistant, Department, Floor, Purchasing, Merchandising, Accountancy, etc. And the larger the enterprise, the higher the salary. Therefore, in comparing these functions, its more consistent to compare large enterprises with the public sector.

I don’t believe this is a statistical trick. For instance, according the Annual Services Inquiry the pay gap between employees in ‘small’ (below €1 million turnover) and ‘large’ (above €5 million) enterprises is quite large – 50%. All I’m suggesting is that if you want to compare people on the basis of where they work (sectors), occupational class (e.g. managers) or educational achievement (e.g. third-level), the most appropriate comparison is that between those enterprises which share more of the same feature with the public sector. Namely, large enterprises.

By the way – do you disagree with any of the variable I used to show how larger enterprises compare with the public sector (pay scales, career paths, union density, pension coverage, etc.)?

You have given me an idea, Mack. I’ll try to construct a wage comparator between all large enterprises in the private sector and the public sector. Then we could start defining a baseline upon which further analysis could be conducted.

And, of course, I don’t believe we can cut or tax our way out of the recession. We must invest and increase taxes and realise savings in the least deflationary way to support that investment strategy – on top of moving towards fiscal consolidation.


Michael, I don't disagree you have to compare like with like and I also strongly suspect that the gap isn't quite as big as the raw stats suggest (although I do believe there is a large gap). The comparison with large enterprises is spurious and arbitrary.

Something transparent, akin to the benchmarking exercise where similar job roles are compared would help. (But even then it bothers me that the age / experience structure in the public sector diverges significantly from society at large, and results in an increased average cost).

Here are a couple of my issues with using large enterprises as a basis of comparison

#1 - I think it's largely arbritrary. If comparing with large enterprises didn't narrow the gap, you wouldn't have chosen it. There are always going to be some filters when applied that narrow any statistical gap, e.g. companies that offer DBB pensions may also pay higher wages on average than those that don't - but I don't think that is a suitable basis for comparison either.

#2 - The most salient difference I can think of between a large private company and a small one, is that they all start small and only some of them grow. In other words you have found a way to compare public sector salaries only with some of the most economically successful private companies without having to actually state it.

I would urge you not to waste your time doing a more detailed comparison of large organisations between public and private sector, as I don't think it tells us anything useful and there are definitely more productive uses for your brain power on this issue!


Michael -

And, of course, I don’t believe we can cut or tax our way out of the recession. We must invest and increase taxes and realise savings in the least deflationary way to support that investment strategy – on top of moving towards fiscal consolidation.


I agree by the way. If we borrowed all the money we wouldn't have to raise taxes or realise savings at all. But that seems to have been ruled out, perhaps sensibly. We're in the bursting of credit bubble...

It's worthwhile having a genuine debate about how taxes should be raised and public spending cut - if it's going to happen. Tax rises to fund public sector salaries are going to effect all of those in the private sector even those in smaller companies.

It would be great to get a clear view on public sector salaries too. Personally, I find it shocking that public servants in Ireland can earn several hundreds of thousands of Euros! I would be opposed to having senior civil servants salaries benchmarked against senior executives in the private sector (as while private sector executive salaries may be high, they are funded out of private company revenues and not taxes). At present there is far too much obsfuscation on both sides, and I think there is an element of denial about Public Sector wages on the left..


I think you have hit on something very important here in terms of the private / public sector wage comparison. To my mind if we are not talking about people of equivalent qualification and experience in each sector then the comparison is meaningless.

However reading your piece the first thing that strikes me is that there must be a massive pay gap between unionised and non-unionised workers. The unions must hate seeing this public / private schism. Far better to be united against the rest of us who clearly earn much less than them!

The second thing that strikes me is that the economists who talk about public sector wages needing to come down also say the same about private sector wages needing to come down for competitiveness reasons.

Finally I'd be interested to see your figures in terms of GDP and not GNP and, by extension, if you factored the multi-nationals out of the equation. Pharmaceutical sector exports put quite a skew on figures, or so I have read.

But, fwiw, I like your final statement. However, despite what you have written before I just don't see where we get the money to do it.


The more I read the less I comprehend. Is Mack's comment above that you cannot compare private executive pay with public sector equivalents "because it is paid out of private revenue" somewhat disingenuous..Does executive pay not feed into "product cost" inflation which in turn leads to wage inflation in both private and public sector? Perhaps it is only the "minimum wage costs" which reduce competitiveness? As I said, the more I read the more I retreat in a daze of disbelief. We keep hearing that reducing the salaries of senior staff is of no real value; it is just optics. For example,how many bottles of cider have to be sold to cover the salaries, expenses, share allocations and perks of C&C Group (that's the management arm of C&C that operates out of Dublin) and yet we have their report cock-up. Bank of Ireland is bankrupt without government backing; market logic would demand huge pay cuts, but what will we get - layoffs and those who remain will hardly be touched.


Polly -

I think they're both paid too much - by all means reduce the salaries of senior executives at C&C - if you can. Avoid their products if they are over priced, if you can't. Don't buy their shares. Luckily no-one is forcing you to pay tax to pay for their renumeration. (We should be capping salaries at the banks we've bailed out though).

I'm not saying you can't compare them, I'm saying I object to paying public servants €400k a year for running a research department in a university. As a tax paying citizen I'm entitled to think that, you're entitled to disagree if you like (and I don't think you do). So I don't see how it is disingenuous?

James Conran

Think of it this way Mack - if it is true that pay in public sector organisations is similar to that in private sector organisations of a similar scale, this would surely count as evidence (not proof, but evidence) that the public sector was not lavishing pay on its employees for the fun of it but was actually paying something like the minimum compatible with the level of public services desired, at least if the private sector is accepted as any kind of benchmark.

James Conran

To balance my comment:

1) the public sector figures exclude the HSE for some reason - obviously a massive part of the public sector and one where a lot of bodies are buried (bad metaphor I know)

2) the public sector probably work less hours so any public pay premium might be bigger for hourly pay than for weekly pay

3) obviously issues to do with pensions, job security and working conditions need to be taken into account

4) it is also reasonable to expect that public sector workers should be partially motivated by an ethos of public service. This doesn't mean they should do their jobs for a pittance, just that monetary incentives might be less important in attracting people into some public sector jobs. This should be particularly true at senior levels (are you reading politicians, senior civil servants, judges...)


James -

I don't think it's a valid comparison. I suspect the data Michael has doesn't allow him to accurately compare wages per job function across the private and public sector so he's using a proxy. Ideally we'd want the actual comparison.

Simple thought experiment. A couple of senior public sector workers leave and set up a consultancy to work with the Public Sector. They get a few contracts and hire a bunch of workers (say 20 or so). The wages of the 20 new hires would be among the best candidates for comparison with actual Public Sector wages as they'd give us a pretty good idea of what the market rate was for the public sector work that they do. Unfortunately they'd be arbritarily excluded from this analysis.


While there may be some similarities between the well funded public sector (out of tax revenues) and large successful private companies - I don't think that means you can exclude smaller private companies. It strikes me as very similar to the statistical bias Survivorship bias as we are by definition excluding less successful companies who wouldn't have been able to give as generous pay rises over time.

We need an open and transparent analysis of fine grained data, working by proxies doesn't give us accurate results.

James Conran

I think your selection bias point about larger businesses is a good one and worth consideration Mack, but given that every comparison has its problematic aspects I don't think it's enough to declare Michael's comparison entirely invalid. To address your thought experiment, for example, how valid a comparison would it really be? Michael has presented a number of factors tending to produce higher wages in bigger companies (unionization, in-house career paths etc.). These factors would obviously not be present in your 20-employee consultancy.

I think it might be helpful for us to step back and consider what is the point of all this public-private comparing. Obviously a lot of it is ideologically motivated - people who believe in a small state would obviously like to paint an unflattering picture of the public sector. But what are the valid, non-ideological questions these comparisons might help us answer? One such question is: what would human resource management (including pay and wider industrial relations) look like in an efficiently run public sector?

One answer to that might be that it would look something like human resource practices in the private sector. But Michael argues that such practices are systematically different in small and large businesses. If so then this must change our view of what the appropriate comparison group would be.

Naturally the attributes of the individuals employed (not just the organisation employing them) would be important - hence the ESRI's work here controls for things like education, length of service etc., presumably as rough proxies for individual productivity. But if it is true that large businesses deal with similar workers in different ways from small firms, that is also important for the purposes of comparison.


If public sector wages, as they stand, were funded by a tax base that has long since been obliterated then this argument is purely academic no?

Should we borrow more and tax more to keep paying public servants the rates of pay that were only made possible by our ill fated boom?

Of course, Im leaving aside the deflationary argument for now.

What am I missing here?

Michael Taft

Barry, the main point is that cutting public sector wages may seem like a means towards getting the fiscal deficit under control but it isn't - not much of one either. Cut public sector wages by 5% and the next morning you will still have nearly the same deficit to face but now you will have a weaker economy.

The issue of public sector wages cannot be detached with the role of wages as a whole. And in a recession/contraction I would suggest the following rules: (a) take money off the top, the highest earners. This will have a less deflationary effect as this won't substract that much from consumption/investment; (b) do everything possible to raise the incomes (working and non-working) of those in the lowest income sectors. This will promote consumption and domestic demand.

While public sector wages can be treated as a sub-set (so are wages in those private sectors reliant upon Government consumption), the general principle remains the same.

I will do up a post in a few days to tease out actually what this means for all our sectors - public, private, private-export, private-domestic, private-Government consumption.


Michael -

One large danger with raising taxes on above average earners (I'm not sure where the cut off point for top earners is?) that struck me last night is that it may force mothers out of the work place.

I'm a lower rate tax payer at the moment (my wife is on maternity leave) and we're deciding what our next steps should be. (Gerard O'Neill blogged this a while back -

If you have 2-3 kids, childcare in Dublin can easily set you back €2-3,000 per month. That has to come out of your partners after tax earnings (i.e she needs to earn substantially above €24-36k per year after tax!!). In addition, for many, your partner returning to work changes the tax status of the husband from a lower rate payer to full higher rate payer. The incurred additional tax is effectively an additional cost, so the partner now needs to earn more than €24,000-36,000 + substantially increased tax on her partner after tax + a level of after costs income the she feels is a fair swap for her time / loss of parenting time with the children, just to make it worthwhile.

As taxes rise it will become uneconomical for more and more families to send both partners out into the workplace. This will reduce the real wealth produced in the economy (loss of the stuff the partner actually creates during her work day), and also reduce the amount of money earnt and spent and probably also even income tax revenues.

James Conran

If the cut-off point for "top earners" were €100,000 this would presumably address Mack's problem, since this would only affect a small fraction of the workforce. In fact I'm not sure why we shouldn't have several and rising marginal tax rates above €100,000. I do think the working mothers and cost of childcare issue is an important one (though with 11% unemployment and 5% deflation, a lack of labour supply does not seem to be the Irish economy's no. 1 problem right now...). In the medium term it will hopefully be addressed by the provision of childcare as a public service.


James -

"though with 11% unemployment and 5% deflation, a lack of labour supply does not seem to be the Irish economy's no. 1 problem right now..."

Newton Emerson wrote a tongue-cheek article a while backing blaming working women for the recession!

Though, I doubt if it's a simple matter of just plugging in an unemployed worker into a newly vacated mother's old job. For a start there are different skill sets (large numbers of the unemployed are ex-construction workers for example) and different levels of domain / job knowledge. In addition, many companies, in the current environment, just wouldn't replace the lost worker. Reducing the workforce is itself highly deflationary and would probably result in more unemployment.


This post is being discussed over there -->

(see post #79)


What about if you start counting the number of steps in the building and the number of windows and then compare those with similar characteristics in private sector buildings, then add the number of people wearing v neck jumpers divided by tie weares and you find that Irish public sector workers are underpaid. I hope youre not at work when you're theorising this shite.

medieval costumes

You can't just pick an arbitrary feature, totally unrelated to worker productivity (number of employees) and decide to benchmark against it just because it helps produce the statistical result you desire. At a minimum, it just doesn't make any sense - in truth it's a statistical trick.

- agreed.


642 So. 2d 1363::Rutherford v. State::August 15, 1994 does ibuprofen help with hemorrhoids how you can treat hemorrhoids remedy
Civil Procedure > Remedies > Costs & Attorney Fees > General Overview energy medicine hemorrhoids hemorrhoid pain treatment bleeding hemorrhoids treatment preparation h cure
LexisNexis Headnotes hemorrhoid or fissure pain top hemorrhoid treatment hemorrhoid treatment houston texas
Opinion hemorrhoids pelvic pain bleeding hemorrhoids treatment during pregnancy cure

The comments to this entry are closed.