House of Commons - Defence

Select Committee on Defence First Report

2 Cost basis for schemes

12. One of our central criticisms of the reviews has been the Government's insistence that any new proposals should cost no more than the schemes currently in place. As a result, the reviews have been, in the Government's own words, no more than "reshuffling the pack".[28] In our report last year, we concluded that "the pension review was hamstrung almost from the outset by the decision that the new proposals should be cost-neutral",[29] and that "the MoD needs to do a great deal more work on its proposals … taking best practice as its starting point, rather than cost-neutrality".[30] As the MoD's written evidence makes clear, however, and much to our disappointment, the Ministry's "prime objective" has remained "to produce a new scheme broadly equal in value to the current one".[31]

INTERPRETATION: MORTALITY ASSUMPTIONS

13. Moreover, the MoD's definition of 'cost neutrality' has changed, to the disadvantage of personnel.

14. In a memorandum in 2002, the MoD explained, in answer to our questions about cost-neutrality, that

the same actuarial assumptions have been used to calculate the costs of the existing and proposed schemes, except that, under the proposed arrangements allowance was made for the expected changes to exit rates close to the Immediate Pension point.[

The memorandum also explained what those assumptions were, stating that the allowances for pensioner mortality were "derived from standard tables", with the mortality of officers assumed to be lighter than that for other ranks.[33]

15. The Ministerial Statement of September 2003, however, states that "Overall, the changes will be broadly cost-neutral, taking account of the steps taken to cover increasing longevity costs."[34] The MoD has provided tables from the Government Actuary's Department (GAD), showing the demographic assumptions used, and the revised costs. The table below shows the GAD's estimate of how much the scheme, and elements of the scheme, will cost the MoD as an average notional percentage of pensionable salary for each member of the scheme:

AFPS—Final Scheme Design

Cost of scheme expressed as a level per cent of pensionable salary

Benefit	Existing Scheme	Existing Scheme with Revised Actuarial Assumptions⁽⁴⁾	New Scheme
Retirement: Pension Lump Sum	12.8 2.4	15.2 2.4	14.3 (see note 2)
Preserved Benefits: Pension Lump Sum	2.5 0.7	3.7 0.7	2.7 0.7
Ill-health Benefits	1.2	1.2	1.6
Spouse's Benefits	2.3	1.2	2.3
Death in service (lump sum)	0.1	0.1	0.4
Total	22.0	24.5	22.0

Source: Ev 45

NOTES:

(1) In calculating the above figures it has been assumed that Officers account for 25 per cent of total pensionable pay and Other Ranks comprise the remaining 75 per cent.

(2) Early retirement packages are still to be finalised. Cost shown is the amount provided for retirement pensions and lump sums and early retirement packages.

(3) A cost of 1 per cent of pensionable pay (as at 31^st March 2003) is roughly equivalent to a cost of £50 million.

(4) The effect of revised actuarial assumptions is an increase in the cost of the scheme due to the allowance for increased longevity. This increased longevity allowance reduces the value of spouse's benefits as it delays the point at which the member is assumed to die and the payment to their spouse commences.

16. The very important actuarial assumption on pensioner mortality has been changed in the course of the consultation process, to allow for "the cost of pensioners living longer".[35] From the table above, it can be seen that the expected cost of the existing scheme to the MoD in 30-40 years from now has been recalculated from 22 per cent to 24.5 per cent of pensionable salary, because of the changed demographic assumptions. The new scheme, however, will return the estimated cost to 22 per cent, over the same timescale, by when the Government will be starting to make payments under the new scheme to substantial numbers of personnel. Using the original actuarial assumptions, the cost of the new scheme has been calculated at 20.3 per cent—that is to say, the total value of the new scheme today is 1.7 per cent of pensionable salary less than the total value of the existing scheme.[36] In other words, the Government intends to use savings that it is making within the pension scheme not only to improve other benefits to personnel, but also to cover the estimated cost of personnel and their dependants living longer. This conflicts with what the Minister told us in evidence, that "all of any money that is saved … from one part of the scheme will be used in benefits on other parts of the scheme".[37]

17. We requested, and have received, further written information from the MoD on the revised mortality assumptions:

Comparison of the costs for existing and new schemes under revised and 1997 valuation mortality assumptions

	Cost of scheme expressed as a level % of pensionable salary
	1997 Valuation Mortality assumptions	Revised mortality assumptions
a) Existing scheme	22.0%	24.5%
b) Scheme proposed in original consultation exercise	22.0%	24.1%
c) Saving from deferral of pre IP preserved pension age from 60 to 65	0.9%	1.0%
d) Additional costs beyond b) for the following improvements: see Note (2) - 4 times death benefit - Partners pensions - Spouses' benefits increased by 25% see Note (3) Continued accrual over 35 years see Note (4)	0.1% 0.4% 1.0% <0.1%	0.1% 0.2% 0.6% <0.1%
e) Additional savings if benefits for exits between IP point and age 55 restructured	Assumed 2.0%	Assumed 2.0%
f) Total cost of revised scheme including restructured benefits below age 55.	20.3%	22.0%

Source: Ev 44

Notes:

(1) In calculating the figures it has been assumed that Officers account for 25% of total pensionable pay and Other Ranks comprise the remaining 75%

(2) The effects of these improvements are cumulative in the order shown i.e. as an example, spouses' benefits are increased by 25% after allowing for partners pensions. Totals are affected by rounding errors.

(3) Partners' benefits are also increased to 62.5% of member's pension.

(4) For members who continue in service, continued accrual up to 37 1/3 years will apply to spouses' benefits and partners' pension and up to 40 years for the member's normal pension and lump sum. Any enhancement formulae for death in service and ill-health will be limited to 35 years' prospective total service, so members who have completed 35 years service or more will not receive an enhancement.

(5) A cost of 1% of pensionable pay (as at 31^st March 2003) is roughly equivalent to a cost of £50 M.

18. This information reveals that the assumptions are based not on mortality rates now, but on projected mortality rates in the future, taking into account assumed further improvements in life expectancy up to 40 years ahead. The MoD has stated that

The GAD has indicated to the MoD that it should account to the Treasury for an increased sum of employer contribution[39] as soon as this can be factored into budget plans, as a result of these changed mortality assumptions.[40] The MoD is currently negotiating with the Treasury and the GAD how this should be reflected in the cost the MoD will pay for the accruing scheme liabilities.[41]

19. The MoD tells us that it is "reasonable that members should bear some of the burden of the increased costs given that they will benefit through longer enjoyment of a pension in retirement".[42] Since there is no direct contribution from members of the AFPS to the costs of the pension scheme (although the level of their pay is reduced, currently by 7 per cent, to take account of the value of the scheme), the MoD has not been able to take the route of increasing members' contributions that has been adopted in a number of cases. Other ways of passing on this cost are to reduce the benefits available under the scheme, or to make personnel work to a higher retirement age. One of the corollaries of people living for longer is that in the public services more widely they will be expected to work for longer. The Government has proposed that the retirement age for many public servants should be increased from 60 to 65. The option of working for longer in the Armed Forces, however, is rarely available to personnel. This is a circumstance particular to the manning needs of the Armed Forces. Thus the Government is proposing to reduce benefits to Armed Forces personnel on the basis that they will live for longer, without also giving them a real opportunity to work for longer within the public service. This is not reasonable.

20. Furthermore, only a minority of personnel can actually expect to "benefit through longer enjoyment of a pension in retirement", those personnel, mostly senior officers, who retire at or after the age of 55. This is because for the vast majority of personnel who leave the forces before 55, the age at which they will receive their pension is to be raised from 60 to 65,[43] matching, more or less, their increased life expectancy.

21. The MoD's budget is under pressure, and the costs of the revised mortality assumptions appear to have come as an unexpected extra pressure on this budget. To provide context, each percentage point of pensionable pay that the MoD pays annually under the pension scheme costs about £50 million. The full 2.5 per cent, if it had to be paid now, would cost about £125 million each year.

22. What the actuarial assumptions disguise, however, is that the MoD's gross pensions bill is actually expected to decrease in 30 to 40 years' time, as a result of the reduction in Service numbers from around 330,000 in 1975 to about 200,000 now.[44] What this means is that the MoD is currently making payments to pensioners drawn from a workforce much larger than that to which the Ministry is currently paying salaries. However, by the time the actuarial assumptions on life expectancy are realised, the MoD will be making payments to pensioners drawn from a workforce of roughly today's strength. The Government is basing its decision to reduce the cost of the pension scheme on the expectation that each individual member of the pension scheme will cost the MoD more in future because of increased longevity. However, because of the reduction in Service numbers over the last two decades, we would expect the MoD's total pensions bill to decrease just as the financial effects of increased life expectancy kick in. In terms of the total cost to the public purse, there is no need to adjust the pension scheme to reflect increased longevity in order to achieve 'cost neutrality'. It is only fair to point out that any cost increases as a result of pensioners living for longer are likely to be more than compensated for by the reduction in the total number of pensioners.

23. The MoD claims that "members should bear some of the burden of the increased costs" (our italics). However, the value of benefits available under the new pension scheme has been adjusted downwards to the full value (2.5 per cent of pensionable pay) of the expected cost of the new mortality assumptions. The MoD expects the cost to its budget of the new assumptions to be "somewhat less than" 2.5 per cent, with the rest of this cost likely to be borne by the Treasury.[45] As a matter of principle, we are not convinced that it is right that the financial cost of increased life expectancy should be borne by service personnel, particularly not through reduced benefits. But even if the MoD can justify its claim that personnel should bear some of the burden of the increased costs brought about by improved longevity, it should not have devised the new scheme as it has, on the basis that personnel in the scheme will bear the entirety of costs, particularly where it appears that these costs are almost certainly in excess of what the MoD itself will have to bear.

24. Mortality assumptions change over time. Presumably, the Government actuary has from time to time recalculated the cost of the existing pension scheme, and, on various occasions since 1973, when the last major changes were made to the scheme, the MoD's superannuation liability costs have increased as a result. But the benefits available under the scheme have not previously been altered to keep the cost of the scheme down.

25. Substantial numbers of claims under the new scheme will not be made for several decades. This makes it possible for the MoD to start afresh. Mortality assumptions are different for personnel of different ages, and cannot easily be applied to a scheme which is already in operation. This will be much less of a problem with the introduction of a new scheme. With the introduction of the new pension scheme, the MoD has changed its definition of 'cost neutrality' to seek to ensure that the future cost of an average member of the new scheme should not exceed the current cost of an average member of the existing scheme.

26. The new mortality assumptions will only marginally affect benefit entitlements under the existing scheme.[46] What this means is that the existing scheme will be substantially more valuable than the new scheme.[47] The MoD justifies this because those personnel in the existing scheme "cannot expect to benefit to the same extent from improving life expectancy" as future service personnel.[48] This has a certain logic, but in broad terms is likely to operate for many personnel as an incentive to remain in the existing scheme. We note, however, that the structure of benefits in the new scheme is quite different from the existing scheme, and that a comparison of the value of the schemes in broad terms will often not reflect how they would impact on individual personnel. A serving member of the forces may face a choice, for example, between a scheme which provides more valuable personal benefits in terms of Immediate Pension and preserved pension, and one which provides more valuable benefits to his spouse or partner in the event of his or her death. Nonetheless, personnel transferring to the new pension scheme will receive benefits of a lesser overall value than those remaining on the existing scheme. This would in many circumstances seem to be a fairly obvious disincentive to transfer.

27. We have always been unhappy that the Government in its review of Armed Forces pensions has imposed on itself the constraint of how much current benefits cost, rather than looking more generally at what Armed Forces personnel deserve. We are even less happy now that it emerges that the Government proposes to reduce the overall value of benefits under the new scheme, compared with the existing scheme. The goalposts of 'cost neutrality' have been shifted—to the disadvantage of future personnel.

INTERPRETATION: CHANGE IN THE PRESERVED PENSION AGE

28. The more general context within which the MoD has been working has changed since the initial stages of the review and consultation, with the publication of the Government's Green Paper, Simplicity, Security and Choice; Working and Saving for Retirement, published in December 2002,[49] and the later document, Simplicity, Security and Choice; Action on Occupational Pensions, published in June 2003.[50] The more general impact of these wider public pension policy changes is discussed below. But the proposal to raise the preserved pension age from 60 to 65 has implications for cost neutrality, which need to be addressed.

29. Preserved pensions are pensions payable to personnel who leave service before the age of 55. These personnel do not receive a pension as soon as they leave service (except where this is for reasons of ill-health), whereas personnel who stay on until age 55 do receive a pension on retirement. The consultation proposals envisaged the payment of preserved pensions at age 60. In line with more general Government policy, however, the new scheme envisages payment of these pensions at age 65, both for those who opt into the new scheme and for those who stay in the existing one (for future service).

30. We discuss below the implications of this change for Armed Forces personnel.[51] From the Government's point of view, however, it will generate cost savings. In about 30-40 years' time, a man of 60 who is a member of a public service pension scheme will, according to the Government actuary, live on average for about a further 26 years[52]—by not paying preserved pensions until age 65, the Government will save about one fifth of its total bill on preserved pensions. Although it will also mean making early departure payments for longer to personnel who leave service after the age of 40 (or 18 years' service), these payments are to be reduced (as we discuss below), and are already (in the form of the Immediate Pension) of lesser value year on year than the preserved pension.

31. These changes have not been factored into the Government's calculations of how much the existing pension scheme will cost, as there has been "no final decision by Government on how such a change might be applied to current public service pensions schemes and from what date any such change might be effective". However, savings are, according to the MoD, likely to be "low".[53] We note that this, however, will depend on how many personnel choose to transfer to the new scheme; the MoD's expectation seems to be that relatively few will. The Government should ensure that any money which would have been paid to pensioners in the existing scheme were it not for the rise in the preserved pension age is used to improve other pension benefits.

28 HC (2001-02) 666, Q 154 Back

29 HC (2001-02) 666, para 149 Back

30 HC (2001-02) 666, para 149 Back

31 Ev 39, para 1 Back

32 HC (2001-02) 666, Ev 39 Back

33 HC (2001-02) 666, Ev 53 Back

34 Ev 19 Back

35 Ev 19 Back

36 Ev 43 Back

37 Q 48 Back

38 Ev 39 Back

39 The Accrued Superannuation Liability Charge Back