Unfunded pension promises don’t cost less to keep, but they’re more easy to break.

John Ralfe is right in saying that the pensions we as tax-payers have to and will have to pay to public sector employees are the same , whether they are funded for or whether they are paid by the Treasury out of tax revenues. That is fact and not open to challenge.

What we could  question is the friction that is created by building up a fund and paying pensions from that fund, rather than from money in the Treasury’s great big spending pot.

This  is what this little exchange between Mike Harrison and me is about.

Someone like Andy Young or Grant Ballantine or any of my readers who has worked on funded and unfunded pension costs at the Government Actuary’s Department will be able to tell me why there are tax-payer supported organisations as various as the Industrial Training Board, the Parliamentary Contributory Scheme and the Local Government Pension Scheme that are funded and why the NHS, Civil Service, Teachers , Fireworkers,  police and others are unfunded.

The late John Shuttleworth did a lot of work on the extra cost of pre-funding DB pensions at the end of the last century. The numbers are history but in my memory, the reason for pre-funding was never that clear. It costs quite a lot more to run a pension fund than to run a pension scheme that simply draws down from Government cash-flows. If you don’t trust that the cash will be there, you want a fund, but we trust the state to collect the revenues to pay the state pension , so why not all the other state organisations including LGPS etc. Anyway, there will be reasons and it is very hard to dismantle a fund (though the Government did it with Royal Mail, taking its assets onto our balance sheet and paying pensions from the central coffers. It can be done though the scale of an LGPS dwarfs anything done so far.

So why this argument with the IFS?

In macro-terms, the cost of funding in terms of operational issues, is not worth get bothered about, so why is John getting so hot under the collar about the IFS suggesting different discount rates for funded and unfunded public sector schemes?

It takes a bold man to take on the IFS on economic issues, but John Ralfe is a bold man. So let’s look at what he’s objecting to

it is sensible to include the rest of the last sentence which runs

Such a rate may also be more sensible when
estimating the ‘value’ to employees.

The IFS do not elaborate on this final point , but there is an argument that a funded pension is less easy to cut than an unfunded one. If there is a fund in place to meet future liabilities, then cutting those liabilities is politically more difficult (witness the ongoing battle over the McCloud judgement). I think the IFS are implying that a funded pension has more certainty of getting paid and therefore warrants a lower discount rate (implying higher value) than an unfunded one.

This begs a further question. It goes on to point out that pensions are skewing total reward in favour of public sector employees

When taking an estimate of employer pension contributions into account, there was a
raw difference of 21% between average public and private sector remuneration in
2021. When controlling for employee characteristics, we estimate the average
public–private remuneration differential to be around 6% (meaning that public
sector workers are paid roughly 6% more than their private sector counterparts on
average, once pensions are accounted for). This total remuneration differential has
fallen in recent years, but to a lesser extent than when considering pay alone, as a
result of the increasing relative generosity of public sector pension contributions

Flexing salary and pensions

And more contentiously, argues that the balance between present pay and deferred pay could be altered.

There is a strong case for rebalancing public sector remuneration away from
pensions and towards pay. A far greater share of overall public sector remuneration
is deferred, in the form of both employer and employee pension contributions,
compared with the private sector (20.1% versus 7.6% on average in 2021), and this
difference has been increasing over time. That means for a given level of
remuneration, take-home pay is lower in the public sector. One option, as a starting
point, would be to reduce employee pension contributions in the public sector,
alongside a commensurate decrease in pension generosity. That would increase
take-home pay for public sector employees with no change to the costs for their

To put it another way, if there is a deal to be done on the pay settlements for nurses and other striking groups of public sector employees, could it be based on such a rebalance?

The logic of the discussion on discount rates is that those in unfunded pensions should be feeling more vulnerable to cuts in future pensions.

When is a pension guarantee not a guarantee?

I think the answer to this will be fought out , not by lawyers , but by the key negotiators in the public sector pay battles to come.

The Government has made its position clear. For whatever reason (and blame needs to be taken out of this), there are insufficient public funds to meet the cost of living (plus) rises demanded by public sector workers.

I suspect that the highly flexible discount rates (the SCAPE rate +) used by Government to impute their future  pension costs for schemes like the NHS are recognising there is a little more “flex” in the promise, than elsewhere (for instance in the private sector DB promise).

And public sector pensions, unlike private sector pensions do not have a safety net called PPF, there is no Plan B for pensions, pension costs can only be reduced by negotiating benefit cuts or by cutting real pay.

Discount rates are a measure of pension value

Any of the 3000+ former members of the BSPS pension schemes know how discount rates effect the value of their pensions. High discount rates limit the value of the CETV and the value of compensation paid for poor advice.

People know that the higher the discount rate, the lower the value of the promise.

So what John is missing is that it is not the cost of funding that matters, but its value. The MPs have a funded pension promise, so long as the funds are there it is hard not to meet that promise.

Unfunded pensions have a promise that is less valuable – it may cost (nearly) the same to meet the promise, but it’s a promise that is easier to break.

I am not cleverer than John Ralfe, I just see things differently. In this case I think I see what the IFS are getting at and it’s different to what John’s saying!



About henry tapper

Founder of the Pension PlayPen,, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in pensions and tagged , , , , . Bookmark the permalink.

5 Responses to Unfunded pension promises don’t cost less to keep, but they’re more easy to break.

  1. John Mather says:

    When pension ministers talk about “unsustainable” and others talk of “unconditional obligation” one concern would be default

    The Great Demographic Rehearsal
    In 2020 a valuable analysis of the problem was produced by Charles Goodhart London School of Economics, London, UK Manoj Pradhan Talking Heads Macro, London, UK ISBN 978-3-030-42656-9e-ISBN 978-3-030-42657-6

  2. John Mather says:

    Reversal not rehearsal

  3. con keating says:

    The intrinsic cost of a pension is independent of the way in which it is financed. The argument for funding a scheme is simply that it defrays some or all of the cost of that provision. That is the primary argument for using the contractual accrual rate that Iain Clacher and I have advocated as the discount rate for its evaluation in the interim.
    The cheapest way to deliver is in fact unfunded and insured – as per the Swedish and German models.
    So, if you are government, with no need for insurance of your credit standing, unfunded is simply cheaper than funded.

  4. Allan Martin says:

    Unfunded public sector pension promises are based on the SCAPE discount rate. SCAPE is based on the long term growth (GDP) in the UK economy (the fund from which taxation is taken). Unfortunately the achieved GDP is not the CPI+2.4-3.5% pa assumed on this £2.1tn (WoGA 2020) of index linked pension promises (~gilts). This past service deficit is an inter-generational transfer, future tax payers pick up the tab, another arithmetic description would be a Ponzi scheme.

    The current CPI+2.4% long term GDP assumption was based on an OBR estimate, but for 2028 onwards, ignoring 2018-28 at 0.6% less. Some might question the “reasonableness” of that actuarial assumption and challenge the OBR for not highlighting the fiscal risk. (The Pensions Regulator doesn’t regulate this bit of occupational pensions for their staff or 5m others.)

    I suggest that next time time unattained UK GDP growth is quoted you do some private sector or funded pensions arithmetic, e.g. recession = 0% GDP for one year, £2.1tn x 2.5% = 5p on the basic rate of income tax! A failed FoI requests prompts the suggestion that this issue is in the HMT “too difficult box”, very brave decisions would be required. For this next financial crisis, His Majesty might ask “Why did no one see it coming?”

  5. ros altmann says:

    Allan Martin is quite right. SCAPE is hopelessly inadequate as a discounting measure for unfunded public sector pensions. Unlike the private sector funded DB schemes, a public sector unfunded scheme’s benefits, with full inflation-linking, cannot legally be changed. The benefit is fully protected in law, whereas state pensions are a benefit and private sector funded schemes have not only got limited price inflation uprating, but also no PPF reductions, the accumulated scheme benefits are protected in full. Even changing future benefit accruals does not change the costs already incurred for decades ahead. The biggest change the Government made was cpi instead of rpi, but even the Hutton reforms have ended up costing more rather than saving money so far, because of the way they were implemented. The McCloud situation gives the best of either past schemes or Hutton reformed benefits, so costs have still been rising. Every year for the past many years, the actual cost of public sector pensions has turned out far higher than was assumed in the fiscal figures.
    The bottom line is that public sector pensions cannot just be changed easily, both because of their legal protections but also because of the power of 6 million public sector workers who can bring the country to a halt.
    Of course, public sector workers deserve good pensions – the trouble is that they actually have superb pension arrangements, beyond the dreams of most private sector workers, but they don’t realise how valuable they are. The costs for each household in the country are enormous, but it is all hidden away behind a cloak of pension comlexity, so even those who benefit from these arrangements don’t usually realise just how much they are worth.

Leave a Reply