The politics of pensions

Chapter by Tony Lynes for an unpublished book, April 1999.

Over the past half century, British governments have tried unsuccessfully to establish a clear and durable definition of the respective roles of pay-as-you-go social insurance and funded ‘private’ arrangements in pension provision. The latest attempt to achieve this goal is the Green Paper, A new contract for welfare: partnership in pensions , published in December 1998 by Tony Blair’s ‘New Labour’ government. It is based on the belief, set out by Blair in a Foreword, ‘that those who can save for their retirement have the responsibility to do so, and that the state must provide effective security for those who cannot’. The state would thus be left with a residual role, involving increasing reliance on means-tested supplements rather than contributory benefits. The government’s proposals, if implemented, will represent a decisive shift from public to private provision. Public spending on pensions will fall, it is claimed, from 5.4% of GDP to 4.5% by 2050; and over the same period ‘the proportion of pensioner incomes coming from the state, now 60%, will have fallen to 40%’.

Although the term ‘Green Paper’ implies that the proposals are open to revision in the light of public response, it is clear that the government regards the main lines of its policy as firmly settled: indeed, legislation enacting the proposals for a new type of ‘stakeholder’ pension schemes had already been laid before Parliament before 31 March 1999, the deadline for comments on the Green Paper. Moreover, the general direction in which the government is moving is not inconsistent with the policies pursued by the 1979-87 Conservative governments, and parliamentary opposition is therefore likely to be directed mainly to questions of practical implementation rather than of principle. Nevertheless, the Green Paper has been strongly criticised, not least by organisations representing today’s pensioners, whose needs it largely ignores and for whom the shift from pay-as-you-go social insurance to private funding and means-testing represents an obvious threat (the response of the National Pensioners Convention is, significantly, entitled The Unwanted Generation).

The growth of occupational pensions

For half a century from the start of state old age pensions in 1909, public and private pension provision in the UK developed independently of each other. Before 1948, state pensions, payable at a flat rate without regard to the individual’s previous earnings, were mainly restricted to manual workers, while occupational pension schemes were established by employers mainly for ‘white collar’ workers. From 1948, national insurance was extended to all employed and self-employed people, though married women were not obliged to contribute for pensions in their own right since they could receive a reduced pension on their husbands’ contributions. Around the same time, occupational schemes spread to cover increasing numbers of manual workers – especially those employed in industries nationalised by the post-war Labour government. In the post-war period of full employment, occupational pensions offered a means of both attracting employees and retaining them by penalising those who left before retirement age (most schemes made little or no provision for the preservation of pension rights of early leavers). By 1956, however, occupational schemes still covered only 43% of male and 21% of female employees (Government Actuary, 1994, p.4), with public sector employees much more likely to be covered than those in the private sector, and white collar workers than manual workers.

The vast majority of occupational pension schemes in the UK are set up for the employees of a single employer. The estimated number of schemes in the private sector in 1956 was between 35,000 and 40,000. In view of the uncertainties involved, it was regarded as axiomatic that such schemes must be ‘properly funded’, unlike the pay-as-you-go national insurance scheme – although precisely what constituted proper funding was far from clear, as the then President of the Institute of Actuaries explained in 1958 (and the explanation is no less valid today):

‘Pension schemes are not to be regarded as a rigid railway track to a fixed destination. The target is continually shifting, mainly because of changes in wage levels, but also because of changes in mortality, interest, expense and tax.

‘Pension schemes can therefore be more happily regarded as homing onto a distant and moving target under the guidance of the actuarial radar tracking system. Whether a scheme will be successful or not is only in part a question of where it is now: that is to say, its current degree of solvency. It is also largely a question of the power of its driving force to bring it curving on to track in due course. The main driving force is the ability of the employer to fulfil his obligations and to increase his contributions whenever necessary. Solvency is therefore often inextricably bound up with the resources of the employer.’ (Redington, 1959, pp. 5-6).

Very large numbers of occupational schemes, generally covering relatively small numbers of employees, were administered by insurance companies on the employer’s behalf. The larger schemes were more often ‘self-administered’, the best of them being based on the civil service superannuation scheme, which provided a lump sum on retirement of 3/80th of final salary for each year of service and a pension of 1/80th per year of service, roughly equivalent in total value to a pension of 1/60th per year of service. The adoption of this model was strongly encouraged by the rules regarding tax exemptions for pension scheme contributions and investment income, and it became increasingly prevalent from the 1960s on. Before then, however, most schemes in the private sector, and especially those for manual workers, offered pensions of either a fixed amount per year of service or a percentage of the member’s average pay over the period of membership, with no provision for revaluation to take account of price inflation or rises in average earnings.

Labour’s National Superannuation plan and the Conservative response

1957 is a crucial date in the political debate on pensions in Britain. In that year the Labour Party, then in opposition, published National Superannuation , a plan to raise the flat-rate national insurance pension to a more adequate level for those already in retirement and to provide additional earnings-related pensions for future pensioners. The authors of the plan pointed to the contrast between the majority of workers who could expect to receive on retirement only the flat-rate pension of about one-fifth of national average earnings, and the minority who would be entitled to an occupational pension in addition. The new National Superannuation scheme was designed to provide all employees, through the mechanism of social insurance, with pension rights of the same kind, if not at the same level, as those of the best occupational schemes. The aim was to produce for the average male employee a total pension of about half his pre-retirement earnings. While this might appear less generous than many occupational schemes, National Superannuation would have important advantages over them. In particular:

(1) The new earnings-related pension would be based not on final salary but on the employee’s pay for each year’s membership of the scheme, revalued up to retirement age in line with the increase in national average earnings – a much fairer formula, especially for manual workers whose earning capacity was likely to decline in the years preceding retirement.

(2) Pension rights would be fully preserved on a change of employment and revalued in line with average earnings up to pension age, just as they would be if there were no change of employment.

(3) Pensions in payment would be increased annually in line with a pensioners’ price index.

Despite these advantages, with about one-third of all employees covered by occupational schemes, it was hardly possible to propose an extension of national insurance into the territory occupied by them – the provision of earnings-related pensions – without, at the same time, proposing a mechanism by which state and private schemes could co-exist. The introduction of a compulsory, universal state earnings-related pension scheme, leading to many of the best occupational schemes being wound up, would have been strongly opposed by their members, as well as by the highly influential financial enterprises involved in them.

The solution proposed was that employees who were members of approved occupational schemes should be excluded from the state earnings-related pension unless, individually, they chose to ‘contract in’. To qualify for approval for this purpose, an occupational scheme would have to satisfy the condition that it ‘should not compare unfavourably’ with the state scheme; and radical changes would have been needed in most occupational schemes to enable them to satisfy that condition (Labour Party, 1957, pp. 41-42). Legislation to implement these proposals would, therefore, have encountered serious problems.

In the event, those problems were deferred by the victory of the Conservative Party in the 1959 general election and the enactment of a very different state earnings-related pension scheme, the ‘graduated pension scheme’, which remained in operation from 1961 until 1975. The main purpose of the graduated scheme was, as the government admitted, ‘to place the National Insurance scheme on a sound financial basis’ (Ministry of Pensions and National Insurance, 1958, p. 13). It achieved this by introducing a system of graduated (i.e. earnings-related) contributions based on a limited range of earnings, to produce an immediate increase in the income of the National Insurance Fund, while promising a very modest additional pension, based on the amount of graduated contributions paid but fixed in cash terms, with no provision for revaluation in line with either prices or earnings. An employer with an occupational pension scheme was allowed to ‘contract out’ of the graduated scheme, on condition that the occupational scheme offered pension rights equivalent to the maximum available from the graduated scheme; but as that condition could be satisfied by occupational schemes of very poor quality, wholly unprotected from inflation, it posed no threat to existing schemes. On the contrary, it encouraged insurance companies to market new occupational schemes offering small flat-rate pensions, designed solely to enable employers to save money by contracting out.

At the same time, however, occupational pensions based on ‘final salary’ were spreading rapidly. In the private sector, the proportion of pension scheme members in schemes of this kind rose from 23% in 1963 to 55% in 1967, 62% in 1971, 79% in 1975 and over 92% in 1979 (Government Actuary, 1968, p. 19; 1972, p. 26; 1978, p. 49; 1981, p. 42). Schemes based on average career earnings, rather than final salary, had been undermined by inflation in the post-war period. Despite the drawbacks of the final salary formula for manual workers, trade unions increasingly saw it as a desirable objective. And the insurance companies which had previously regarded final salary pensions as uninsurable were forced to revise their policy when faced with the threat of losing their share of the market (the Government Actuary noted the change in a report published in 1978: ‘Whereas, in the past, schemes insured with life offices have tended not to offer benefits depending on salaries of employees at or near retirement, this is now not the case and of the 20,350 schemes where all the contributions are paid to the life offices about 15,150 (74 per cent) offer final-salary benefits.’ (Government Actuary, 1978, p. 35)). Despite the improvement in the quality of occupational pension schemes, however, they still covered only about half the total number of employees. The problem of providing adequate pensions for the other half remained unsolved.

SERPS: a public-private partnership

By 1974, therefore, when a new Labour Government grasped the opportunity of replacing the graduated scheme with an earnings-related state pension scheme based on the party’s 1957 National Superannuation proposals (a previous Labour Government having narrowly failed to do so before being defeated in the 1969 election), the need to accommodate occupational schemes in its plans was more obvious than ever. The scheme which resulted, later known as SERPS (the state earnings-related pension scheme), embodied a new approach to the problem of allowing occupational schemes to contract out without compromising the standards of the state scheme.

The design of SERPS owed little to the example of funded occupational schemes, and much to that of pay-as-you-go social insurance schemes in other countries. The SERPS pension was to be a proportion of the individual’s earnings between a lower limit of about one-fifth of average earnings (the level of the basic pension) and an upper limit of about 1.5 times average earnings. At pension age, each year’s earnings would be revalued in line with average earnings. The pension would be based on the person’s best 20 years of revalued earnings between the lower and upper limits, at the rate of 1.25% of each year’s earnings, producing a pension (in addition to the flat-rate basic pension) of about 25% of earnings after 20 years of contributions – a formula which offered particularly good value to those reaching pension age in the first 20 years of the scheme. Widows were to inherit the whole of their husbands’ SERPS pension as an addition to their own entitlement, subject to a maximum, and the scheme also provided invalidity pensions.

Adding to these features the full preservation of SERPS pension rights on a change of employment and the fact that SERPS pensions in payment were to be fully protected from price inflation, it was clear that if occupational schemes were required to offer similar benefits in order to be allowed to contract out, none would be able to do so. The solution adopted was a partnership between state and occupational schemes, in which the occupational schemes would continue to provide the kinds of benefits for which they were designed, with the state filling the gap between those benefits and the target set by SERPS. Thus all employees with earnings above the lower limit would benefit to some extent from SERPS, even if they were ‘contracted out’. Far from threatening the continued existence of occupational schemes, SERPS would actually encourage it by making good some of their most obvious deficiencies.

To contract out of SERPS, an occupational scheme had to satisfy two conditions. The first, known as the ‘requisite benefits’ condition, was that it must provide a pension accruing at an annual rate of at least 1.25% of pensionable earnings, which compared favourably with SERPS for those with more than 20 years’ working life remaining. In practice, pensionable earnings, for this purpose, normally meant the earnings of a period of one or more years preceding pension age (‘final salary’).

The second condition was that the scheme must provide a ‘guaranteed minimum pension’ (GMP) similar to the full SERPS pension, but calculated on a year-by-year basis (i.e. not applying the ’20 best years’ formula). The GMP calculated at pension age was, however, to continue at the same rate regardless of any subsequent increase in prices. Moreover, the occupational scheme did not have to pay invalidity pensions, was required to provide a widow’s pension of only half the husband’s entitlement, and the pension rights of early leavers did not have to be fully protected. In all these respects, a contracted-out scheme was allowed to fall short of the SERPS standard, with the state making good the deficiency.

The effect of the two conditions was that employees could normally expect to derive some additional benefit from being contracted out of SERPS and, at worst, would not suffer any significant loss. Moreover, the reduction in national insurance contributions for contracted out employees and their employers (the ‘contracted-out rebate’) was deliberately set slightly above the estimated cost of the benefits that the occupational scheme had to provide. The 1975 Pensions Act, therefore, raised dramatically the standard of state pension provision while encouraging the further development of funded occupational schemes as an alternative to the state scheme.

The 1980s: Conservative cuts

It seemed at first that a political consensus had been reached on pension policy, but yet another change of direction was to follow, with the advent of another Conservative government in 1979. The main change affecting existing pensioners was the repeal in 1980 of the statutory requirement introduced by the Labour government that the basic pension should be increased annually in line with average earnings or prices, whichever had risen faster in the preceding year. In future, the requirement was to be limited to compensating for price increases. Although assurances were given that, in practice, pensioners would continue to share in rising national prosperity, subsequent annual increases were strictly in line with the price index, resulting in the basic pension falling between 1980 and 1997 from about 20% of average earnings to about 14%. In retrospect it seems probable that, at the Treasury if not at the Department of Social Security, this had always been the intended result of the repeal of the statutory link with earnings and was seen as the first stage in a deliberate and sustained shift from public to private pension provision and from pay-as-you-go to funding. Alarmist statements about the financial consequences of population aging in the 21st century helped to create a climate of opinion favourable to this process.

Substantial cuts in SERPS were to follow. A Green Paper published in 1985 proposed that the scheme should be abolished altogether. In its place, contributions of at least 4% of earnings would have to be made to an occupational or ‘personal’ pension scheme ( Reform of Social Security , 1985, p.6). The idea of allowing personal pension schemes as an alternative to occupational schemes was of fundamental importance. The assumption that the availability of private pension schemes must depend on the willingness of employers to provide them was always going to limit their potential coverage. Allowing the insurance companies to sell pension schemes on an individual basis seemed to offer the possibility of extending private provision to the whole employed population. Without the employer’s involvement, however, a pension scheme of the final salary type would not be financially viable. It was assumed, therefore, that all personal pension schemes and most new occupational schemes would be ‘defined contribution’ schemes operating on a ‘money purchase’ basis, in contrast to the defined benefit schemes which, alone, satisfied the conditions for contracting out of SERPS.

The proposal to abolish SERPS was soon dropped. The private pensions ‘industry’ was unenthusiastic about the prospect of having to provide for millions of low-paid and mobile workers, while the short-term financial consequences of abolishing SERPS, involving the disruption of the national insurance contribution system, were unattractive. Instead of outright abolition, therefore, the government introduced legislation (the Social Security Act 1986) substantially reducing the level of SERPS pensions for those becoming entitled in 2000 and after. The accrual rate was reduced, to produce pensions of 20% instead of 25% of earnings between the lower and upper limits; the ’20 best years’ formula was abolished, so that the pension would be based on earnings averaged over the person’s whole working life (excluding non-earning years due to child-rearing or sickness); and widows would, in addition, inherit only half their husbands’ SERPS pension instead of the whole of it. Figures produced by the Government Actuary showed that the combined effect of the SERPS cuts and the assumed continuation of the policy of raising the basic pension only in line with prices would halve the total cost of state retirement pensions by about 2035, compared with what it would otherwise have been (Government Actuary, 1986, pp. 10-11). On the same assumptions, national insurance contribution rates would be lower than in the 1980s, although the contributor/pensioner ratio was by then expected to have reached its lowest point.

Although SERPS was not abolished, the proposal to allow private money purchase schemes, personal or occupational, as a substitute for state provision was implemented by the 1986 Act. This was a fundamental change in the conditions for contracting out. A money purchase scheme could not be required to provide a ‘guaranteed minimum pension’, since the value of the pension would depend on future investment returns and annuity rates and no such guarantee could be given. Contracted out employees, therefore, could no longer be assured that their pension would be worth at least as much as if they had remained fully within the state scheme. In spite of this, personal pension schemes were sold over the next few years on a scale many times greater than the government had anticipated and in very large numbers of cases are now known to have been mis-sold to people for whom they were plainly unsuitable. The development of money-purchase occupational schemes was at first less dramatic, since employers with existing defined benefit schemes showed no great inclination to replace them with money-purchase schemes, but a gradual shift has occurred as employers setting up new schemes have tended to prefer the money purchase option.

Another package of economy measures enacted by the Conservative government in 1995 included the raising of the state pension age for women from 60 to 65, over the ten-year period 2010-20. It also carried to its logical conclusion the process of disconnecting SERPS from contracted-out occupational and personal schemes. The process described above, by which SERPS was to fill the gaps between the GMP and the benefits offered by the state to those contracting in, has been discontinued for pension rights accruing from April 1997 on. This change alone constituted a major reduction in long-term entitlements under the state scheme and paved the way for eventual abolition of SERPS.

On 5 March 1997, less than two months before the elections which ended 18 years of Conservative rule, the government produced a new plan misleadingly entitled ‘Basic Pension Plus’ (referred to below as BP+) which would have involved the abolition not only of SERPS but of the basic pension too for future entrants to the labour force, both being replaced by compulsory private schemes. It was a programme for the complete privatisation of pension provision. The abolition of the basic pension was to be achieved by diverting part of their national insurance contributions (£9 per week was suggested) into a money-purchase personal pension fund backed by a state guarantee: if the money-purchase pension turned out to be less than the basic pension would have been, the state would pay the difference. This enabled the government to claim, however implausibly, that, far from abolishing the basic pension, it was proposing to guarantee its future value. As the Minister explained: ‘The scheme is called the Basic Pension Plus because it builds on the present basic state pension. It is the basic pension – plus a personal fund, plus a state guarantee.’ A more accurate description would have been ‘a personal fund, plus a state guarantee – minus the basic pension’.

SERPS was, similarly, to be closed to new entrants, who were to be obliged to pay at least 5% of their earnings to an occupational or personal scheme – more than the 4% proposed in the 1985 Green Paper but still only about half the amount generally regarded as necessary to provide an adequate funded second pension.

An obvious objection to these proposals was the cost imposed on the new generation of workers, compelled to contribute to funded pension schemes for their own pensions while also contributing towards the current cost of pensions for those already retired. The government proposed to reduce the effect on public finances by altering the tax treatment of occupational and personal pension schemes, but this would have meant simply that the transitional cost would be met through higher taxes rather than through contributions.

The government did its best to conceal the inadequacy of the pensions that would result from implementation of BP+. A person with average earnings retiring around 2040, the minister claimed, could expect a total pension of £175 a week – about 50% of average earnings in 1996; but he was later forced to admit that, on his own assumption that earnings would rise by 2% in real terms, the pension would be only 18.1% of average earnings in 2040 for a man and 22.9% for a woman – assuming that they were in paid employment throughout their working lives, a particularly improbable assumption for a woman.

It is perhaps surprising that the BP+ plan should have involved replacing the basic pension with the money-purchase personal pension schemes which had caused the government so much embarrassment over the previous decade. The government argued that this time, since the new schemes would be compulsory, there would be no danger of mis-selling and the excessive costs involved in selling them on a voluntary basis would also be avoided. It could, however, equally well have been argued that with a captive market there would be less incentive to offer good value; and, while people would no longer have to be persuaded to buy, they would still have to be persuaded to deal with one insurance company rather than another. However low the costs of selling and administering such schemes might be, they would certainly be higher than those of a multi-member occupational scheme – and very much higher than the costs of a pay-as-you-go pension run by the Department of Social Security.

The 1997 Labour Government: towards private pensions for all

Although the change of government in May 1997 put an end to these plans, the Labour Party had made it clear, in a policy statement on pensions the previous year entitled Security in Retirement , that it did not propose a return to the policies of the 1974-79 Labour government. While not going as far as BP+ in the direction of privatisation, the statement nevertheless envisaged the continuation of the move from social insurance to funded pension schemes. Although the ‘over-arching objective’ was said to be ‘to ensure that all pensioners, today and tomorrow, share fairly in the increasing prosperity of the nation’, it pointedly avoided any commitment to reverse the changes in state pensions made since 1979. Regarding the basic pension, it said only: ‘The basic state pension is currently indexed to prices and Labour would not reduce this commitment …’ The absence of any commitment to restore the link with average earnings was underlined by the emphasis placed on plans to help the poorest pensioners by ensuring that they would receive the means-tested minimum income to which they were entitled.

The brief references to SERPS in Security in Retirement showed a remarkable lack of enthusiasm for a scheme which, on its introduction, had been regarded as one of the main achievements of the previous Labour government. It was to be retained ‘as an option for those who prefer to remain in it’, but:

‘… SERPS could not easily or sustainably be rebuilt in its original form. New funded pension schemes could produce better returns for the same contribution level for many people.’

In one respect, Security in Retirement did propose an extension of SERPS: a ‘citizenship pension’ for people, including low-paid workers and carers, who could not otherwise be expected to obtain an adequate second pension. For those in paid employment but without access to occupational pension schemes, however, new private funded schemes were proposed. These ‘stakeholder’ schemes would generally be large: unlike most occupational schemes they would not be limited to the employees of one firm but would be ‘multi-employer’, enabling them to operate more efficiently and at a much lower level of costs than personal pension schemes. SERPS, it was implied, would remain only as a last resort for those unable to make satisfactory arrangements in the private sector.

At the Labour Party’s annual conference in October 1996, serious concern was expressed about the direction in which it appeared to be moving and, although Security in Retirement was approved, it was agreed that a comprehensive review of pensions policy would be carried out after the expected change of government. The main proposals in Security in Retirement were repeated in the party’s manifesto for the May 1997 election, together with the promise of a review. In July 1997, organisations and individuals were invited to submit their views to the team of Department of Social Security officials carrying out the review. The government’s proposals were to be published, for further consultation, ‘in the first part of 1998’. In the event, the resulting Green Paper, Partnership in Pensions , appeared at the end of 1998.

The Green Paper contains few surprises. Its proposals are mainly concerned with provision for future pensioners and reflect the belief ‘that those who can should save for their retirement, and that the State should provide greater security for those who cannot’. The main groups who are assumed to be unable to save for their retirement are those earning less than £9,000 a year (around half average earnings) and those who ‘cannot work, through no fault of their own, for example because they are looking after young children or relatives with an illness or disability, or are disabled themselves’. For them, a new State Second Pension (SSP) is proposed. This would replace SERPS (contrary to the 1997 election manifesto’s promise that SERPS would be retained ‘as an option for those who prefer to remain in it’) and, after a transitional period, would provide only a flat-rate pension unrelated to the individual’s earnings. Thus, in the long run, earnings-related pensions would be provided only by occupational, stakeholder and personal pension schemes, not by the state.

The Green Paper has been strongly criticised by pensioners’ organisations for failing seriously to address the needs of existing pensioners, for most of whom the basic pension is the main source of income. Although the election manifesto promised that the basic pension would be increased ‘at least in line with prices’ and the terms of reference of the review included ‘all aspects of the basic pension and its value’, the Green Paper has little to say on the subject, other than that it will be increased ‘in line with prices’, omitting the words ‘at least’. To restore the basic pension in 1999-2000 to the same proportion of average earnings as in 1980, before the link was broken, would mean raising it from £66.75 to over £92 a week for a single pensioner, but the Green Paper does not mention the possibility of making good even a small part of this loss. The justification offered for refusing to restore the earnings link, even for future pension increases, is that by concentrating help on the poorest pensioners, the government will be able to do more for them than if the money were spent on raising the basic pension for ‘rich and poor pensioners alike’ – ignoring the fact that most pensioners, while far from rich, do not fall within the government’s definition of ‘the poorest’ as those entitled to means-tested income support. The measures proposed to help the poorest pensioners include a small increase in income support (to be known in future as the ‘minimum income guarantee’) in April 1999, the long-term aim of raising income support rates in line with average earnings, and efforts to reduce the number of pensioners entitled to income support who do not claim it. The fact that these measures will apply only to pensioners, not to younger recipients of income support, confirms the impression that the government is more concerned with deflecting demands for increases in the basic pension than with making income support more adequate.

The proposals relating to income support could, of course, be combined with the restoration of the earnings link. If short-term costs were the only consideration, that would certainly be affordable. A report by the Government Actuary published in January 1999 shows that by the end of the financial year 1999-2000 the National Insurance Fund is expected to have a surplus of nearly £6 billion over the level recommended by the Actuary as a working balance. Raising the basic pension in line with earnings rather than prices in April 1999 would have cost only about £0.4 billion. The long-term costs would be much greater but by no means unmanageable. The Green Paper seeks to dramatise the issue:

‘Some want the State to take on a far greater commitment to future pension payments, for example by raising the basic state pension in line with earnings. We . . . do not believe it would be right to commit enormous sums – over £30 billion – regardless of people’s needs and the nation’s ability to pay the bills in the future.’ ( A new contract for welfare: Partnership in Pensions , 1998, p. 7)

When asked how the figure of £30 billion had been calculated, the minister, Stephen Timms, explained that the cost would be ‘of the order of £33 billion’ in 2030-31, offset by savings of about £8 billion in means-tested benefits ( Hansard , 26.1.99, col. 234). He did not explain why the year 2030-31 had been chosen or why the cost for that year had been overstated by some £5 billion. It is difficult to avoid the conclusion that the £30 billion figure was inserted in the Green Paper simply to give the impression that restoring the earnings link would be too expensive to contemplate. The figure is, however, based on the assumption that average earnings will rise by 1.5% per annum up to 2030. As the National Pensioners Convention pointed out in its response to the Green Paper, if National Insurance contributions rise by the same percentage, ‘the income of the NI Fund from contributions will have risen by about £29 billion by 2030-31 – more than enough to meet the net cost of restoring the earnings link.’ (National Pensioners Convention, 1999, p. 8).

The government’s long-term proposals assume that, as now, the first tier of pension provision will consist of the basic state pension, while the second tier will be a combination of state pay-as-you-go and private funded schemes. With the basic pension falling over the next half century to only about 7% of average earnings, however, second pensions will become the main determinant of pensioners’ total incomes.

The Green Paper envisages two major changes in the types of second pension schemes available and in their coverage. The first of these is the introduction of stakeholder pensions, making it possible for a very large proportion of the employed population to contribute to a funded pension scheme on a collective basis – unlike the present situation in which the only private funded alternative to an occupational pension is a personal pension. It is claimed that ‘up to five million people stand to benefit from stakeholder pension schemes’. A combination of government regulations and advantages of scale will, it is intended, ensure that stakeholder pensions overcome two of the main objections to personal pension schemes: the high level of charges which makes them poor value for money, especially where the level of contributions is relatively low, and the penalties imposed on those contributing to a scheme for only a short period. Like personal pensions, however, stakeholder schemes will operate on a money purchase basis, involving a high degree of uncertainty as to the performance of the investment and annuity markets and, therefore, as to the value of the pension. Since employers will be obliged to offer access to a stakeholder scheme for any of their employees not covered by an occupational scheme, but will not be obliged to make any contribution to the stakeholder scheme beyond the amount they would otherwise contribute to the state scheme, many employers may prefer stakeholder schemes to defined benefit occupational schemes on grounds of cost, thus reinforcing the exsting trend in favour of money purchase schemes and reducing both the adequacy and the security of their employees’ pensions.

The second major change envisaged is the substitution of the State Second Pension (SSP) for SERPS. The intention is that, from about 2006, for those then aged under 45 and for new entrants, the SSP will provide pensions at twice the level that SERPS would have provided for employees earning (in 1999 terms) £9,000 a year or about £173 a week. The precise value of the pension, as a percentage of average earnings, will depend on the extent to which the lower earnings limit for contributions (tied to the price index) falls behind the growth of earnings, but the Green Paper estimates it (again in 1999 terms) at ‘almost £50 a week’ for a person who contributes to the scheme throughout their working life.

The SSP, the Green Paper explains, is intended to meet two priorities:

‘Our first priority is to give more help to those for whom private second pensions are not an option, either because they have low earnings or because they are, for a good reason, not working or seeking work. Our second priority is to help moderate earners to build up better second pensions, with as many as possible joining funded pension schemes.’ (A new contract for welfare: Partnership in Pensions, 1998, p. 39)

The underlying assumption is that stakeholder schemes will not be available to people who are not in paid work, and will not be suitable for those earning less than £9,000, who will therefore find it more advantageous to contribute to the SSP unless they are covered by an occupational scheme. Above £9,000, there will be a grey area in which some people may prefer the security and predictability of the SSP and others the possibility of obtaining a larger pension from a stakeholder scheme. The government’s aim is to make that grey area as narrow as possible and to encourage those in it to opt for stakeholder pensions, using the contracted-out rebate of national insurance contributions as an incentive. As the Secretary of State, Alistair Darling, told the House of Commons on the day of publication of the Green Paper:

‘… it is my intention to ensure that we amend the system further so that, if people stay in the state system, they will lose money.’ (Hansard, 15.12.98, col. 771).

The way in which this is to be done is plainly inequitable. At present, people who are contracted out of SERPS forgo an earnings-related pension from the state and are therefore granted a contribution rebate which is a percentage of their earnings. In future, the state pension forgone would no longer be earnings-related, but the contribution rebate would not only be earnings-related but would actually be more generous than now. The Pension Provision Group, set up by the government in 1997 to report on current levels of pension provision and likely future trends, has commented on this proposal:

‘We have doubts whether in practice it would be politically sustainable for the Government to give earnings-related rebates in return for people giving up flat-rate benefits. It would be seen as giving subsidies to higher earners.’ (Pension Provision Group, 1999, p 20).

If earnings-related rebates are not ‘politically sustainable’, however, the alternative is a flat-rate rebate which, if it fairly reflected the value of the SSP forgone, would not provide the desired incentive for higher-paid employees to contract out; while a flat-rate rebate which included an incentive element could cause difficulties for low earners, who would have to decide whether the size of the rebate outweighed the advantages of the SSP. A further complication is the fact that, for the first 20 years after the SSP becomes a flat-rate scheme for the majority of contributors, a minority of older contributors will continue to build up earnings-related pension rights in respect of their earnings over £9,000 a year. Thus, while flat-rate rebates may be appropriate for the majority, they will not be appropriate for the minority. Clearly there are serious problems here which the Green Paper does not attempt to resolve.

We have already noted the Green Paper’s estimate that, in terms of 1999 earnings, the SSP will be almost £50 a week for a person who is ‘contracted in’ throughout their working life. We have also noted the expectation that, on the assumptions made in the Green Paper as to the rate of increase in average earnings, the price-linked basic pension will fall over the next half century to about 7% of average earnings – about £30 a week in 1999 terms. Assuming that the aim of increasing the income support ‘minimum income guarantee’ in line with average earnings is achieved, the minimum income, for a single pensioner under 75, will remain at the earnings-related equivalent of its 1999 value of £75 a week (separate provision is made, at that level of income, for rent and council tax). A crude measure of the adequacy of the government’s proposals can be obtained by comparing the combined value of the basic and second pensions with the minimum income. To make such a comparison does not imply that the minimum is itself adequate, which pensioners’ organisations would strongly dispute. It is, however, the criterion that the Green Paper itself applies: ‘People who work all their lives should not have to rely on means tested benefits when they retire’. ( A new contract for welfare: Partnership in Pensions , 1998, p. 29)

According to the Green Paper, the basic pension and the SSP will, together, satisfy this criterion:

‘These reforms mean that anyone who works throughout their working life (including spells as a carer or off work through long-term illness or disability) will receive a total state pension above the rate of the minimum income guarantee. In this way, every bit that they save, however small and however infrequently, will count towards their final pension and will not need to be topped up by the new minimum income guarantee. This will reward both hard work and thrift.’ ( A new contract for welfare: Partnership in Pensions , 1998, p. 4)

In support of this statement, the final chapter concludes with a table which shows ‘broadly’ the combined amounts of pension that people retiring in 2050 can expect to receive:

Lifetime earnings (per week) Pension
£100 £78
£200 £81
£300 £89
£400 £94

Even those with earnings below £9,000 a year, it appears, will have a total pension (basic plus SSP) above the minimum income level. Replies to subsequent parliamentary questions, however, present a much less rosy picture. As many as one in four pensioner households are expected to be dependent on income support for a minimum income in 2050; and even if all those in work save an extra 5% of their earnings from now on, the proportion on income support will be about one in five ( Hansard , 12.1.99, col. 152) – a higher proportion than in 1999.

There are several reasons for this discrepancy. The table shows the initial value of the pension for those retiring in 2050, but the pensioner population of 2050 will consist mainly of people over retirement age whose pensions have, at best, risen in line with prices since they retired, while income support rates are not only assumed to have risen in line with earnings but also include additional ‘premiums’ for those aged 75 or over. These older pensioners, therefore, are more likely to qualify for income support than those just reaching pension age. Another reason why more pensioners are likely to need income support than the figures in the table suggest is that those figures assume that national insurance contributions have been paid or credited for the whole of the person’s working life, from age 16 to 65. Most people, even if they start work at 16, have gaps in their earnings record at some stage in their career. The government proposes a system of SSP credits for limited categories of carers and people who have been incapable of work for a ‘substantial’ period; but not for students, the unemployed, those incapable of work for an ‘insubstantial’ period or parents who fail to return to work as soon as their youngest child reaches school age. In practice, therefore, most people who rely solely on the SSP for a second pension are likely to need income support on or soon after retirement. Even a person earning £400 a week in 1999 terms (more than average earnings), who, according to the table, would receive a total pension (including a funded second pension) of £94 a week on retirement in 2050, would find that the pension had been overtaken by income support some 13 years later.

In short, it is not only low-paid employees, wholly dependent on state pensions, who can expect their future retirement income to be as inadequate as that of today’s pensioners. Those with earnings around the national average, opting for a private stakeholder pension instead of the State Second Pension and contributing at the minimum level required by the contracting-out rules, are likely to be only a few pounds a week better off on retirement than those earning half as much, and stand a good chance of ending their days on means-tested income support.

The end of the public-private partnership?

It is tempting to argue that private pension schemes can provide adequate pensions if contribution rates are high enough, and that all would be well if the government were willing to use compulsion to achieve that result. But the fact is that the government is not willing to add to employers’ costs in that way, whether directly by increasing employers’ contributions or indirectly by increasing employees’ contributions, thus generating pressure for higher wages. Moreover, the inadequacy of the levels of pension envisaged in 2050 is in large part a result of the failure to restore the earnings link for the basic pension. It would be absurd to try to compensate for that failure by compelling people to pay higher contributions to private money-purchase schemes, in pursuit of a doctrinaire belief in the virtues of funding.

If the earnings link were restored from now on, without any action to make good the ground lost between 1980 and 1999, the cumulative effect, by 2050, would be to increase each of the figures in the pension column of the table above by about £35, producing total pensions of between £113 and £129 a week – hardly generous by European standards but well above income support levels. And while increased contributions to funded schemes would do nothing for those already over pension age, increased National Insurance contributions (or even a decision to spend the surpluses currently accumulating in the NI Fund) would enable the basic pension to be increased immediately, with future increases in line with average earnings.

The development of pensions policy in Britain has hitherto been based on the recognition by successive governments of the complementary roles of public pay-as-you-go and private funded schemes. As other countries have shown, this is not the only way to provide adequate pensions; but there are no examples of the private sector alone successfully doing the job. The British public-private partnership reached its highest point with the introduction of SERPS in 1978. The basic pension, statutorily linked to average earnings (or to prices in any year when they had risen more than earnings), was to provide a solid foundation, while SERPS provided a minimum standard for earnings-related second pensions, encouraging rather than inhibiting the growth of private funded schemes but at the same time ensuring that contracted-out employees would continue to benefit from those features of the state scheme that private schemes could not fully replicate.

The partnership was gravely weakened by the Conservative government’s legislation of 1986 and 1995. Contracted-out schemes were in future to stand alone without the backing of the state scheme and were no longer required to provide defined benefits or a guaranteed minimum pension. Meanwhile, the role of the basic pension was shrinking as a result of the breaking of the earnings link.

The proposals in the December 1998 Green Paper would complete the dissolution of the partnership, making the basic pension increasingly irrelevant and terminating the state’s involvement in the provision of earning-related pensions. Only the SSP and the means-tested ‘minimum income guarantee’ would remain as an acknowledgement of the inability of the private sector alone to meet the retirement needs of the entire population.


Government Actuary (1968), Occupational Pension Schemes, Third Survey by the Government Actuary, London, HMSO.

Government Actuary (1972), Occupational Pension Schemes 1971, Fourth Survey by the Government Actuary , London, HMSO.

Government Actuary (1978), Occupational Pension Schemes 1975, Fifth Survey by the Government Actuary , London, HMSO.

Government Actuary (1981), Occupational Pension Schemes 1979, Sixth Survey by the Government Actuary , London, HMSO.

Government Actuary (1986), Social Security Bill 1986: Report by the Government Actuary on the Financial Effects of the Bill on the National Insurance Fund , Cmnd. 9711, London, HMSO.

Government Actuary (1994), Occupational Pension Schemes 1991, London, HMSO.

Labour Party (1957), National Superannuation , London, Labour Party.

Ministry of Pensions and National Insurance (1958), Provision for Old Age , Cmnd. 538, London, HMSO.

National Pensioners Convention (1999), The Unwanted Generation , London, National Pensioners Convention.

A new contract for welfare: Partnership in Pensions (1998), Cm 4179, London, The Stationery Office.

Pension Provision Group (1999), Response to the Pensions Green Paper , London, Department of Social Security.

F M Redington (1959), Presidential Address, in Journal of the Institute of Actuaries , Vol. 85, Pt. I.

Reform of Social Security (1985), Vol. 2, Programme for Change , Cmnd. 9518, London, HMSO.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s