House of Commons - Defence

Select Committee on Defence Third Report

PENSION ARRANGEMENTS

Current Arrangements

19. All members of the Armed Forces automatically become members of the Armed Forces Pension Scheme (AFPS). It is a defined benefit scheme which means that the rules of the scheme determine how much pension the member will get, based on final salary, rather than on the amount of contributions. The main benefits of the current Scheme are—

Immediate pension for officers at age 38 or after 16 years' service and at age 40 for other ranks or after 22 years' service (whichever is later). Reckonable service begins at age 21 for officers and 18 for other ranks.
Full pension payable at age 55 based on 34 years' service for officers and 37 years for other ranks. A terminal grant of three times pension is paid on retirement as a lump sum.
Death in service lump sum for spouses of 1 to11/2 times pay, plus a pension of 50 per cent of the ill-health pension that would have been payable had the member retired at the date of death.
Death in retirement pension for spouses of 50 per cent of the pension the member was receiving at the time of death.
Ill-health retirement pension is payable dependent on rank and length of service. A terminal grant of three times the annual rate of pension is also payable.
Additional years can be purchased to provide additional survivor benefits, and additional pension benefits can be purchased through Additional Voluntary Contributions (AVCs) up to Inland Revenue limits, funded by the individual.

Forms of pay other than basic pay, such as specialist pay, are not pensionable. Examples of specialist pay include flying pay and deep diving pay. Pensions are increased annually from age 55 (or the date of ill-health retirement) in line with the Retail Prices Index.

Purpose of a pension scheme

20. The Armed Forces Overarching Personnel Strategy sets out the principles of the pensions policy as being—

to provide a suitable End of Service Benefits and Pensions package to enable the Armed Forces to retain Service personnel
to provide the foundations for secure living for personnel and their dependants once they have left the Armed Forces;
to provide a pension scheme for the Armed Forces that reflects modern standards and is consistent with the legitimate expectations of Service personnel;
to provide for those who are unable to remain in the Services through illness or injury and for the dependants of those who die.[27]

The MoD have told us that, where the Government is the employer, a pension scheme should—

... provide an income, after the normal retirement age for the scheme, that relates to the length of service given by the individual and provides incapacity benefits and protection for dependants ... provide benefits valued by the employees and the incentives should be geared towards the recruitment, retention and motivation needs of the particular employment ... encourage desirable behaviours, such as retaining people while they are particularly needed by that employment.'[28]

21. In evidence, it was emphasised that, for the Armed Forces, pensions are seen 'as part of a remuneration package designed to recruit and retain'.[29] The Minister repeated on a number of occasions that the pensions scheme 'is part of the overall manning strategy' and that both pensions and compensation arrangements were being addressed now because of recruitment and retention needs.[30] We accept the proposition that, for the MoD, as for any major employer, pensions are part of the pay package and therefore have a role as a management tool in recruitment and retention. In fulfilling this function, the Armed Forces Pension Scheme has a specific and unique element: it provides an Immediate Pension for those who leave the Services at a relatively early age and without completing a full career. This element is regarded by the MoD as a significant retention incentive. We discuss this in greater detail below (see paragraphs 58 to 70).

22. However, a pension scheme for the Armed Forces also needs to take proper account of the fact that the 'job' which the Armed Forces do, and the terms of service under which they do it, is fundamentally different to that of any other group of employees, in that they may be asked to kill or to be killed in the course of their duties. The Forces Pension Society takes the view that—

Government places unlimited demands on Armed Forces personnel of a quite different order of magnitude to those placed on any other public servants: moreover Service people have no form of independent representation. Government therefore bears a heavy moral duty to provide pension conditions for Service people which are at least broadly in line with those provided for other public servants, plus any special provisions to meet the particular requirements of military service. It is not acceptable that the requirements of military service should lead to benefits being provided that are worse than elsewhere.[31]

We agree and believe the MoD therefore has a responsibility to Service personnel which goes beyond that of a normal employer. Most people who join the Services are young, some as young as 17 or 18. It is highly unlikely that pensions will be high up their list of considerations in choosing a career. If the MoD is to be an effective and responsible employer it should therefore accept that it has a duty of care for the welfare and long term interests of these young people and that one aspect of this duty is to ensure that the pension provision in place for them when they leave the Services is a proper reward for the commitment they have made.

23. We believe that, in conducting the pension review, the MoD has placed too much emphasis on how the pensions system can be used to benefit the Armed Forces as employers, rather than on ensuring there is proper provision in place for Service personnel as employees, and that this has had a negative influence on the proposals it has brought forward.

Reasons for change

24. The Forces Pension Society believes that the AFPS—

... is well overdue for modernisation. It has many anomalies and shortcomings, and no independence of governance. It also fails to reflect modern best practice standards or the legitimate expectations of Service personnel and their dependants.'[32]

The MoD believes that: 'a number of features of the scheme are inconsistent with other public sector schemes and some provisions fall short of best practice in modern pensions schemes generally'.[33] An example is that the AFPS currently provides benefits linked to length of service and calculated on the basis of representative pay, depending on rank, rather than on actual pay at the time of retirement. This means that individuals retiring at the same time, at the same rank and with the same length of service will receive the same pension, even if their actual salaries at retirement are different.

25. Having decided that a review of the pension scheme was necessary, the MoD's overall objective was—

... to set a benchmark for new pensions arrangements which supported Armed Forces career structures by contributing to recruitment, retention and morale, and which provided a fair level of benefits for individuals while remaining cost effective and affordable.[34]

The stated aims of the exercise were to assess whether—

any parts of the scheme needed to be modernised in the light of good practice elsewhere
any benefits needed to be improved
the special features designed to meet the specific circumstances of the Armed Forces were still necessary and cost-effective[35]

The main areas which were examined were whether—

there was 'headroom' to increase benefits up to Inland Revenue limits if these were otherwise considered 'desirable and affordable'
pension should be linked to actual earnings, rather than representative pay
the age at which reckonable service begins should be the same for officers and other ranks
there should be an even rate of accrual of benefits
there were more flexible alternatives to the Immediate Pension
specialist pay (eg flying pay, submarine pay) should be pensionable
ill health benefits should be based on level of invalidity, rather than simply on rank and length of service
the death in service gratuity is adequate and whether widow(er)s' benefits should be withdrawn on remarriage.[36]

26. The last major improvements to the AFPS were made in 1973 and most people would therefore regard a review as timely and necessary. Our concern in examining the review's findings was to assess whether the reasons for proposing change, and the new proposals themselves, are in the best interest of the Service personnel who are the scheme members.

Proposals for a new pension scheme

27. The key proposals in the consultation document are that—

the scheme would remain unfunded (ie paid for from taxation)
pensions would be based on final salary on retirement, rather than on representative pay, and would accrue from the date of entry, at an even rate, offering a full pension after 35 years
no distinction would be made under the new scheme between officers and other ranks
immediate pensions would be payable from age 40 or after 18 years' service, with an option to choose instead a bonus payment and a later pension
a new three tier structure of ill health benefits would be established, with levels varying according to the level of disablement
dependants' benefits would be improved, including an increase in the death-in-service grant to three times pensionable salary.

Is the AFPS a generous scheme?

28. The Minister for the Armed Forces told us, at the beginning of his session of oral evidence—

I believe there are few who contest that the current pension and compensation schemes are generous—as indeed they are.[37]

He went on to say that the pension scheme was already 'substantially funded' and that pensions do not appear to be a key issue causing dissatisfaction amongst personnel.[38] The pension review was therefore conducted on the assumption that no additional funds were required. We wished to evaluate this assumption and assess whether it could be justified.

Cost to the MoD

29. The Armed Forces Pension Scheme is unfunded, which means that there is no pool of money building up from contributions and investment income to cover the costs of future pensions. Instead current benefits are paid from current taxation. Although the AFPS accounting officer is the Permanent Secretary of the Ministry of Defence, the Pension Scheme expenditure is a separate Vote from the main MoD expenditure (the Consolidated Departmental Resource Accounts). Pensions are paid from Annually Managed Expenditure. The cost is demand-led and there is no flexibility about what is paid out; it therefore falls outside the expenditure control regime of the MoD's Departmental Expenditure Limit which is borne by the MoD mainstream budget. In December 2000 the MoD estimated that the cost of paying Service pensions would rise from �2,257 million in 2000-01 to �2,727 million by 2004-05.[39] In 2000-01, the actual cost was �2,275 million.[40]

30. However, the MoD's main budget is charged a contribution, known as the Accruing Superannuation Liability Charge (ASLC), towards the cost of future pensions liabilities of the existing Armed Forces. This charge is calculated on the basis of 33.8 per cent of the total pensionable pay for officers and 18.2 per cent for other ranks, equating to an average of 22 per cent of Armed Forces pensionable pay.[41] It is paid out of the main MoD Vote and into the separate AFPS Vote to help offset the latter's gross expenditure. This ASLC contribution in the last financial year was �1,039 million.[42] The ASLC payments have to come from the MoD's Departmental Expenditure Limit.[43]

31. This means that, although the Treasury technically meets the costs of Armed Forces pensions paid to those who have retired from the forces, the MoD shares this burden. If a new pension scheme provided enhanced benefits at a greater cost, the MoD would either have to find the extra money from within its budget by reducing other expenditure, or ask the Treasury for an increase in its Departmental Expenditure Limit to accommodate this increase.

Abatement

32. The AFPS is a non-contributory scheme in that no contributions are explicitly deducted from salaries. However, the Armed Forces Pay Review Body (AFPRB) takes into account the deferred pay represented by pension provision in making its recommendations on Armed Forces pay levels. The benefits of the AFPS against those provided by civilian comparator pension schemes are evaluated by independent consultants on behalf of the AFPRB every five years. The AFPRB then applies an abatement to pay to reflect the enhanced benefits of the Armed Forces Pension Scheme over comparators. The level of abatement has been 7 per cent since 1997, and has decreased from 11 per cent since 1981. The Forces Pension Society believes that the decrease in the abatement figure demonstrates that the AFPS has failed to keep pace with modern good practice in pension provision.[44]

33. The abatement evaluation and calculation are complex processes. The most recent comparative valuation was carried out in 2000 and the findings published as an appendix to the AFPRB Report for 2001.[45] The 7 per cent abatement level has been retained, although the actuaries' research indicated that 6 per cent 'would be the most appropriate and fair adjustment for all members of the Armed Forces'.[46] If officers and other ranks are assessed separately, the figure for officers is 5.2 per cent and for other ranks 7.2 per cent, which reflect improvements in officer comparator schemes but a deterioration in other rank comparator schemes.[47]

34. Abatement of pay could be regarded as a means of 'clawing back' any additional expenditure on Armed Forces pensions, in that, if the relative value of the pension scheme increases because of an enhancement of benefits, it would be reasonable to assume that the AFPRB would then increase the abatement percentage to reflect this. The MoD told us that it would be a matter for the AFPRB to decide whether any change in pension benefits should be reflected in a change in the abatement rate, based on a wider assessment of Service pay levels. However the MoD accepted that—

Should the AFPRB decide to abate comparator pay to the full value of additional benefits, it is expected that the overall cost to the Department would be neutral,

—with the caveat that the Department would wish to take a view on increasing pensions at the possible expense of pay.[48] This allows at least the possibility, even within the constraint of cost neutrality, which we discuss below, for the MoD to achieve an overall uplift in pension benefits without contributing additional resources.

Should the AFPS be an overtly contributory scheme?

35. One of the key differences between abatement and an overt contributory scheme is that abatement is a deduction from pay before the gross pay of an individual is calculated. This contrasts with arrangements for employees in other pension schemes, where the pension is deducted from the gross pay as specified on their pay slip. When it comes to calculating the pension to be paid, the service man or woman's final salary on which the pension calculation is based will be a sum from which the 7 per cent abatement has been deducted. Employees in contributory schemes, however, will have their pension calculated on the basis of their gross pay before the pension contribution is deducted. Service personnel could therefore be considered to be losing out in terms of final pension received compared with other employees who might be making a seven per cent contribution to their pension. This disadvantage was highlighted to us in evidence from the Forces Pension Society as a key factor to be borne in mind when comparing the AFPS to other schemes.[49]

36. The MoD had not considered the possibility of introducing overt member contributions into the pension scheme as part of the review, as they believed this would be an 'additional complication'.[50] The Director General, Service Personnel Policy felt that a contributory scheme would have to be linked to a funded scheme, as contributions without a fund would mean that there was an 'element of artificiality'.[51] We find this a very weak argument. All of the public sector schemes listed in the table below, with the exception of the local government scheme, collect contributions from members while remaining unfunded.

37. There are certainly complications in making the AFPS an overtly contributory scheme, and it would affect the way the Armed Forces Pay Review Body calculates Service pay, particularly while the old and new pension schemes were operating. But there would be advantages to the employer, in that actual rates of pay to Service personnel would be higher, in that the 7 per cent abatement would not have been applied, and might therefore be more attractive to potential recruits. There would clearly be advantages to employees as their pensions would be calculated on a gross pay figure, rather than the abated figure. We believe the MoD has dismissed the possibility of contributions without properly considering it. There would inevitably be obstacles to overcome in introducing such a system, but there is no reason to assume, without first properly examining them, that these would be insuperable. We recommend that the MoD look again at the issue of making the Armed Forces Pension Scheme an overtly contributory scheme.

Comparison with other pension schemes

38. We asked the Minister on what evidence his assertion that the Armed Forces Pension Scheme was 'generous' was based. He told us—

Because we are not comparing, in many ways, like with like, it is very hard to take our scheme—existing and proposed—and set it against the wider trend within both the public and private sector, because people retire earlier and there are other matters relating to dependants' benefits and so on, which we have tried to correct. I think it can be tested, but it is only an opinion. It is for that reason that I have asked if there is a way in which this can be examined on an independent basis to make sure that that particular statement stands true examination ... I am fairly sure, in terms of the overall breadth and depth of the existing scheme, that it does rank high.[52]

The MoD set out in its initial memorandum the steps it has taken to compare the benefits offered by the AFPS with those offered by comparable civilian schemes—

The pensions review ... specifically examined the benefits offered under other public pension schemes, in terms of the value and nature of benefits that would be paid in certain circumstances. It also looked at the National Association of Pension Funds survey of private sector schemes.[53]

The Minister undertook to provide us with additional information on comparator schemes.[54] This largely reflects the information we had received from the Forces Pension Society, set out below.

Table 1

Comparison of the AFPS with other public sector pension schemes

Cost of benefits as % of pensionable pay

Member contributions

Cost to employer

Armed Forces Pension Scheme (current and proposed schemes)

22.0%

7.0% (through abatement)

15%

Fire Service

34.75%

11.0%

23.75%

Police

32.0%

11.0%

21.0%

NHS

20.0%

6.5%

13.5%

Teachers

18.5%

6.0%

12.5%

Civil Service (current scheme)

18.0%

1.5%*

16.5%

Local Government

17.0%

6.0%

11.0%

* Members' contributions for the new Civil Service scheme will be 3.5 per cent.

Source: Ev 20 and 120

In private sector schemes, the average member contribution is 4.7 per cent.[55]

39. The Forces Pension Society told us that the MoD's claim that the AFPS was the most expensive scheme in the public sector is based on 'very, very dodgy arithmetic' because it failed to take account of the abatement of pay figure. If the abatement is taken into account, then the AFPS is 'broadly around the average'.[56] The MoD believes that the AFPS compares well with all the main public schemes except the police and fire services, and told us that, as these services generally provide longer careers, it is 'more valid' to compare these schemes against Armed Forces officers' pensions rather than the Services as a whole.[57] We regard this reasoning as somewhat specious, given that abatement is the same for both officers and other ranks in the Armed Forces; moreover, the AFPRB's actuaries' assessment is that officers' pension benefits have reduced against comparator schemes while other ranks have improved.[58]

40. Inland Revenue limits allow for a maximum pension of two-thirds of pay: the AFPS currently pays 48.5 per cent, plus a lump sum equal to three times pension; this will increase under the new arrangements to 50 per cent , plus a lump sum, which will aggregate to 62.5 per cent of pay. The Review document acknowledges that the Inland Revenue limits would have allowed a greater increase in retirement benefits but believed that this would be 'very expensive indeed'.[59]

41. The Inland Revenue limit for spouses' benefits is two-thirds of the member's pension. The current AFPS provides a spouse's pension at 50 per cent of the member's pension; the review proposed no change because this would 'significantly add to expenditure on pensions'.[60] The Government Actuary points out that most private sector schemes calculate spouse's pension on the basis of the member's pension before commutation of any lump sum. But under the AFPS, the lump sum is separately calculated, so a spouse's pension is calculated on the basis of the member's pension net of the lump sum, giving a lower pension.[61] In the private sector, spouse's benefits are almost universally provided for life; the present AFPS, like the majority of public sector schemes, provides a spouse's pension until the spouse dies, remarries or cohabits. Under the new arrangements, Armed Forces spouses' pensions will be payable for life, even after remarriage.[62]

42. Inland Revenue limits for lump sum payments on death in service are four times pensionable pay. The AFPS currently provides a sum between one and one-and-a-half times pensionable pay; this will increase to three times pensionable pay under the new scheme. Half of private sector schemes provide a lump sum of four times salary, and 36 per cent provide three times salary.[63]

43. Occupational pensions have been in the news because of a spate of announcements from private sector firms about the closure of their final salary schemes to new entrants. In general, the companies were substituting money-purchase (defined contribution) for final salary schemes, and contributing substantially less in many cases. This is likely to mean deteriorating pension prospects for many in the future. According to the National Association of Pension Funds (NAPF) 2001 survey, average employer and member contribution rates to money purchase schemes are 6.0 per cent and 4.76 per cent of salary respectively, compared to 9.21 per cent and 7.38 per cent for final salary schemes.[64] One study suggests, for instance, that only under the most optimistic investment assumptions will 10 per cent of pay be sufficient to give someone a money-purchase pension equivalent to that in a good final-salary scheme, and then only after contributing for 40 years.[65]

44. While continuation of this trend in the private sector might affect the calculations of the abatement by the Armed Forces Pay Review Body, we do not believe it provides any argument for the MoD to reduce the quality of the package offered to the Armed Forces. The comparison should be with best practice, and good employers are continuing with final salary schemes because they believe they give security to employees. The John Lewis Partnership, for example, has recently announced that it is increasing its contribution to its final salary scheme, despite a fall in profits, because it sees it as a boost to recruiting and retaining staff.[66]

45. The Government Actuary's assessment of the current AFPS is that—

When compared to best practice in the private sector, AFPS provides some benefits which are more generous, and some which are less generous than best practice. Overall the scheme can be considered as generous and compares favourably with the majority of private sector arrangements. There are some areas where the AFPS may appear poor, in particular the benefits payable on death in service and under ill-health. These areas should be considered as part of the review process.[67]

The Review has attempted to address the deficiencies in death in service and ill health benefits but it has not increased benefits up to Inland Revenue limits, which was an option open to the MoD. It should be noted that the term 'generous' is qualified by the Government Actuary as relating to certain benefits and compared to the private sector; the Minister did not qualify his use of the term.

46. We would expect the AFPS to be generous in comparison with private sector schemes, given the very different commitment the members of the Armed Forces make and the very difficult and dangerous role they are often asked to perform. We are not convinced that the Armed Forces Pension Scheme, even under the proposed new arrangements, could be regarded as generous either by comparison with other public sector schemes, or with what is appropriate to the special status of the Armed Forces.

Cost neutrality

47. We had some difficulty establishing whether the MoD had conducted the pension review on the basis of cost neutrality or not. Cost neutrality would mean that any new scheme would not increase benefits overall; any improvement in one part of the scheme would be paid for by reduced benefits in another part of the scheme.

48. The Review document says that—

It is Government policy that improvements in public service pensions should generally be paid for either by savings elsewhere in the scheme, or by scheme members through contributions. The recommended new scheme would introduce important modernisation compared with the current arrangements, and a number of worthwhile improvements; resources have been provided for the improvements by adjusting some other provisions of the scheme. The recommendations are cost-neutral as a package.[68]

In written questions the Committee asked the MoD to set out what Government policy on public sector pensions is. Cost neutrality is not mentioned as part of overall policy, although the response does say—

In some contexts, such as the provision of benefits to unmarried partners, the Government has made it clear that it is prepared to consider changes in pension scheme rules to provide for the benefit improvements if the members want such changes and are prepared to meet the full cost of the improvement.[69]

The MoD was also asked specifically whether cost neutrality was taken as a fundamental basis for the review. The response was—

Initially, cost-effectiveness and affordability were specified rather than cost neutrality. The aim was that the new scheme should be well balanced to meet the evolving recruitment and retention needs of the Armed Forces and that, both in terms of individual measures and as a whole, the scheme should represent value for money. The review team considered whether improved pension benefits would assist recruitment and retention.[70]

The Review team decided that 'non-pension approaches' were more appropriate for dealing with recruitment and retention problems and—

The conclusion that the review should work within broadly cost-neutral terms was therefore reached at a relatively early stage, reflecting a judgement that a scheme that met the manning needs of the Forces could be delivered within that parameter, and at the same time recognising the high cost and value of the current scheme.[71]

49. We explored this in more detail with the Minister in oral evidence. He told us that—

... cost-neutrality was not originally and necessarily one of the ground rules for the pension review, although cost-effectiveness and affordability were specified. However, we could only justify more money if we could prove that the pensions area caused us real problems in recruitment and retention. Objectively we cannot do that.[72]

So although cost neutrality was not the review's 'original driver', the fact that pensions were not regarded as a major influence on recruitment and retention meant that the review concluded at an early stage that no additional money would be available, because 'a general increase in pension levels could not be a priority'.[73] The effect of that on the proposals for the new scheme was therefore, in the Minister's words, that—

... what we are doing is reshuffling the pack, so to speak; to take some elements and modify them to pay for enhancements and improvements elsewhere.[74]

The Forces Pension Society told us that they had been led to expect that the pensions review would be a fundamental examination of the existing scheme, with alternatives properly costed, before decisions on affordability and priorities were made. But because the constraint of cost neutrality was applied at an early stage, what resulted was simply 'a tinkering at the edges, a redistribution of resources within the existing cost parameter',[75] as indeed the Minister in effect admitted.

27 Armed Forces Overarching Personnel Strategy, Ministry of Defence, February 2000, p 55 Back

28 Ev 38 Back

29 Ev 40 Back

30 QQ 127, 131, 190 Back

31 Ev 17, paragraph 8 Back

32 Ev 25 Back

33 Pension Review document, op cit, paragraph 3.1 Back

34 Pension Review document, op cit, paragraph 3.9 Back

35 Pension Review document, op cit, Executive Summary Back

36 Pensions Review Document, op cit, Section 3 Back

37 Q 127 Back

38 QQ 127, 130, 162 Back

39 Second Report from the Defence Committee, Session 2000-01, Strategic Defence Review: Policy for People, HC 29-II, p 242 Back

40 Armed Forces Pension Scheme Resource Accounts 2000-01, p 10 Back

41 Ev 52. The basis for the calculation is 'that percentage of pensionable salary which, if paid throughout service, would be just sufficient to provide the promised benefits, assuming that future experience were to be line with the actuarial assumptions for the calculation'. The difference in the percentages for officers compared to other ranks reflects the higher proportion of officers who continue in post until attainment of early pension age. See Ev 119-120 Back

42 The MoD paid �1,042 million into the AFPS Vote, but nearly �4 million of this represented Additional Voluntary Contributions paid by Service personnel themselves. Back

43 The MoD's Vote is mostly Departmental Expenditure Limit, but temporarily about a third (approximately �11 billion) is regarded as Annually Managed Expenditure (cost of capital employed and depreciation) until government departments build up experience in estimating and controlling these notional costs. These elements will be counted as DEL in the 2002 Spending Review onwards. Back

44 Ev 18, paragraph 12 Back

45 Armed Forces Pay Review Body, Thirtieth Report 2001, Cm 4993, Session 2000-01, Appendix 4 Back

46 ibid, Appendix 4, paragraph 4.5 Back

47 ibid, Appendix 4, paragraphs 4.17 and 4.49 Back

48 Ev 113, paragraph 3 Back

49 Q 105 Back

50 Q 151 Back

51 Q 149 Back

52 Q 138 Back

53 Ev 40 Back

54 See Ev 113 and 119-126 Back

55 Ev 124 Back