Radical
Statistics
Pension Myth-selling and Gender
Jay Ginn
In order to justify cutting state pension provision, both Conservative and New Labour governments have resorted to Pension Myth-selling. Myths range from those suggesting that state pensions are becoming less important to those arguing that the ageing of populations requires the privatisation of pensions. All these myths are gendered in one way or another. And each is made possible by the complexity of the area of pensions. I shall discuss eight myths in turn.
Myth 1 'Most people have a private pension in retirement'
This myth has been highlighted before by Radical Statistics (Ginn, 1993; Ginn and Dugard, 1994). By citing figures only for men, politicians in the early 1990s gave the misleading impression that 70 per cent of older people have a private (occupational or personal) pension. Alternatively, figures are quoted which refer to the proportion of couples and non-married individuals. In both cases, this use of statistics, by making women's low rate of private pension receipt invisible, boosts the apparent rate of receipt. The figures were widely repeated in the media and served to justify past and planned cutbacks in state pensions.
Recent data shows that among women pensioners aged over 65, only a third have any private pension income (see Table 1). Moreover, the amounts are often small and may bring no financial gain, due to the widening pension poverty trap - the gap between the basic pension and the means-tested Minimum Income Guarantee.
Table 1: Percentage with a private pension and average amounts, age 65+, 1993-4, click HERE
The proportion of older women with a private pension is rising. Yet gender inequality of income from private pensions, among those with income from this source, increased over the eight years from 1986 to 1994. At the median, women's income from private pensions fell as a proportion of men's from 65 per cent to 56 per cent; at the upper quartile it fell from 58 per cent to 54 per cent and at the lower quartile from 75 per cent to 65 per cent (Ginn and Arber, 1999a). This suggests that, as a wider range of women have access to occupational pensions, more who are low paid or have only short membership are included. Much of the income inequality between women and men in later life arises from the gender difference in private pension income. It is no surprise therefore that older women's median personal income as a proportion of older men's fell from 71 per cent in 1985 to 62 per cent in 1993-4 (Ginn and Arber, 1999a).
Because it has been indexed only to prices since 1980 the value of the basic state pension has declined relative to national average earnings, from 20 per cent in 1980 to 15 per cent in 2000 and is set to decline to 7.5 per cent by 2050. Although men, especially those in non-manual well-paid occupations, have been able to compensate for the decline in the basic pension by occupational pension income which has increased with each successive cohort, this is not so for most women and other groups disadvantaged in the labour market.
Thus the policy of reducing the share of pension provision from the state, initiated by the Conservatives, has resulted in widening inequality in pensioners' incomes, structured by gender, class and ethnicity. New Labour's policy shares a similar aim, planning to reverse the shares of pension provision by the state and private sector, from 60:40 in 1998 to 40:60 by 2050 (DSS, 1998).
Myth 2 'Pensioners nowadays are well off'
The myth that the majority of pensioners are well-off and less in need of state pensions than in the past has been used since the 1980s to justify moving away from universal benefits towards means-testing (targetting in New Labour-speak). Since National Insurance (NI) pensions are the largest single item of social security spending (around £40 billion per annum), it is not surprising they attract the attention of politicians determined to cut state spending. Yet NI, in contrast to other parts of social security spending such as Income Support, is a contributory scheme to transfer resources from workers to pensioners in accord with the established intergenerational contract on which all such social insurance pensions are based (see Myth 8).
Undoubtedly, pensioners as a whole saw an improvement in their position relative to the rest of the population over the twentieth century. Their share of total personal disposable income increased between the 1950s and the 1990s, after allowing for their increase as a proportion of the population (CSO, 1986). However, there are important qualifications to this picture of rising prosperity among older people.
First, pensioner incomes were extremely low in the past relative to the rest of the population - the only way was up! Pensioners' per capita disposable income was estimated as only 40 per cent of that of non-pensioners in 1951, rising to 70 per cent in 1985/6 (CSO, 1986) - a substantial rise. But anything less would have been unacceptable by any standards.
Second, recent estimates by the Women's Unit show that in terms of gross individual income pensioners are still much less well-off than wage-earners (see Table 2). The distribution of pensioners' incomes, like that of the working age population, has become more dispersed over time (Hancock and Weir, 1994) so that the average pension income often quoted by politicians has become increasingly unrepresentative of the mass of pensioners. Clearly tax and other deductions from pay modify the income relationship between pensioners and wage-earners, but this is only so because most pensioners have incomes too low to pay much tax.
Table 2: Average gross individual income by family type, men and women, 1996/7 , click HERE
Third, income inequalities by gender and family type are stark (see Tables 1 and 2). Married men had the highest income among pensioners, yet their median gross income was only 41 per cent of average gross male full time earnings; the corresponding figure for married women was 16 per cent. Unfortunately, the Women's Unit provided no analysis of older women's income by marital status. Other research has shown that among ever-married women, widows often 'inherit' private pension income but divorced women, whose employment has usually also been restricted by childrearing, are more likely to face poverty in later life (Ginn and Arber, 1991; 1993; 1999a) and see Table 1.
Falkingham and Victor (1991) investigated the 'myth of the WOOPIE' (Well-off Older Person), which was popular with Conservative politicians and the press in the 1980s. They showed that WOOPIES, defined as those older people in the top quintile of the total income distribution, had not increased between 1974 and 1985. WOOPIES are a small minority, typically male, aged under 75, from the professional classes and receiving private pension income.
Myth 3 'The main problem with personal pensions has been mis-selling to those with occupational pensions'
The commission-driven sales of personal pensions to individuals in good occupational pension schemes was a major fraud and has seriously reduced the pension income those individuals can expect. The finance companies responsible have dragged their feet for years rather than pay compensation; by July 1997 compensation had been paid to only 12,000 out of the 570,000 cases identified since 1994 (Scotsman, 1997).
But the issue of mis-selling to those who had no access to an occupational pension scheme and for whom the choice was between the State Earnings Related Pension Scheme (SERPS) and a personal pension has been virtually ignored. Compared with those who have access to an occupational pension scheme, this group is lower paid and otherwise less advantaged in the labour market.
A personal pension is expected to provide a better return on contributions than SERPS if, in the long run, the rate of interest exceeds the growth in national earnings. Contributing to a personal pension when young and switching back into SERPS between age 30 and 40 is considered the optimal strategy on average (Dilnot et al., 1994). Because of the disproportionate effect of charges on the contributions made by the lower paid, experts have estimated that opting out of SERPS into a personal pension would not be advantageous for those earning below £200 per week (in 1993) (Durham, 1994). Applying these criteria, those who in 1993-4 were earning less than £200 per week or were aged over 40 were probably ill-advised to contract out of SERPS into a personal pension. Table 3 shows the proportions of men and women contributing to a personal pension who were too low paid or too old to benefit from a personal pension, compared with staying in SERPS.
Table 3: Percentage of personal pension contributors (aged 20-59) likely to receive poorer value than if they remained in SERPS*, by hours of work for women, click HERE.
In all, 68 per cent of those who were contributing to a personal pension plan and had no access to an occupational pension scheme were apparently ill-advised to contract out of SERPS, 58 per cent of men and 83 per cent of women. Among women, it was mainly their low pay which made their decision questionable. Only 10 per cent of the women part timers who contributed to a personal pension earned enough for this to be worthwhile and only 4 per cent of part time contributors fulfilled the criteria for both age and earnings.
This estimate of mis-selling to those who would have remained in SERPS is inevitably crude, since the outcome will depend on the specific plan's charges and performance, the level and continuity of the individual's future earnings and on macro-economic developments far into the future. No one can be certain whose personal pension will turn out to provide better value than SERPS and whose will not. But the figures suggest that this type of mis-selling is extensive. It is arguably more serious than mis-selling to those who would otherwise have belonged to an occupational pension, since the ex-SERPS members are a larger proportion of employees and are less advantaged than the group awaiting compensation; 12 per cent opted out of SERPS for a personal pension, compared with only 5 per cent who rejected an occupational pension in favour of a personal pension (Ginn and Arber, 2000a).
There is no plan to compensate individuals who were mis-sold a personal pension instead of SERPS, as the amounts lost have been deemed too small. Whether those on low incomes would agree with this verdict or even knew of it is not recorded.
Myth 4 'Stakeholder pensions will avoid the drawbacks of personal pensions'
High charges in personal pensions, the regressive effects of flat rate charges and penalties for gaps in contributions, all of which are more serious for women than for men, are well known drawbacks. Pension analysts have shown that charges and fees in such individual money purchase pension plans are even higher than generally thought. If charges arising from administration and investment management, from alteration to the contract and from annuitisation are included, these typically reduce the value of the pension by up to 45 per cent (Murthi et al, 1999).
The new Stakeholder Pensions (SPs) have been presented as low charge pensions suitable for the moderately paid (roughly those earning £10-20,000 per annum) lacking access to an occupational pension scheme. Since SPs will replace SERPS as the only earnings related pension available for the moderately paid, many women are expected to opt for this form of pension. Regulations will ensure that SPs have no front-loaded charges, no penalties for gaps or transfers and a maximum charge of 1 per cent. These limits on charges are welcome but is the 1 per cent charge as modest as it sounds? Do people understand how much their pension will be reduced by a charge of 1 per cent? Or what it means?
The effects of a 1% per annum charge on the total fund and a 1% per annum charge on contributions are very different. A simplified example illustrates this. Suppose an individual with a salary of £20,000 per annum pays contributions to a SP for 20 years at 5 per cent of earnings (£1000 per annum, slightly more than she would have contributed to SERPS). For simplicity, we first assume the invested fund has only grown to £20,000 at retirement (with no increase in contributions nor interest on investment).
Table 4: Comparing charges on contributions or on the fund, click HERE
These examples are not realistic because earnings and hence contributions will tend to rise over 20 years. But the point is that the impact of a 1 per cent charge can be much higher than one might at first imagine and compares with zero charges (and very low administration costs) in state pensions.
Myth 5 'The demographic time bomb requires privatisation of pensions'
Alarmist language has pervaded writing on the implications of population ageing for the sustainability and equity of public pension provision. For example, the World Bank has predicted 'a looming old age crisis'. In the World Bank's view, public pension systems will face unmanageable liabilities and must either reduce benefits or extract unacceptably high contributions from future generations of workers (World Bank, 1994). Since the recommended reduction of state pensions will generally hit women hardest, there are gender implications.
Although some European public pension schemes have been over-generous to early retirees and could be improved in this respect, the arguments for switching to private funded pensions have been challenged by both economists and social policy analysts (Mabbett, 1997; Toporowski, 2000; Hills, 1995; Lloyd-Sherlock and Johnson, 1996; Quadagno, 1996).
First, pensions must be paid from the production of the worker generation, irrespective of whether their pensions are public Pay-As-You-Go (PAYG) or private funded schemes. Population ageing affects the viability of funded pensions as well as PAYG schemes, since the return on capital will tend to diminish (Mabbett, 1997). A financial market dominated by pensions funds, economists argue, generates capital market inflation. An ever-increasing inflow of new contributors to private pensions will be required if they are to maintain past rates of return on investment - the chain letter scenario, or Ponzi finance (Toporowski, 2000). Second, the generation in which the transition to private funded pensions takes place has to pay twice; once for its own future private pensions and again for the existing public pension scheme until its last liabilities are fulfilled (Hills, 1995). The double burden of pension contributions during the transitional period is likely to reduce economic performance (Mabbett, 1997). Hence introducing new funded pension schemes exacerbates any problems due to population ageing and unfairly burdens the generation experiencing the transition.
In some countries, increases in the population aged over 65 will be less dramatic in the future than many mature welfare states have coped with in the past (Glennerster, 1999). Paradoxically, Britain faces a less steep decline in the ratio of working age to pension age population than other European countries, yet has gone furthest in reducing state pensions and promoting the private sector of pension provision, mainly at the expense of women (Ginn and Arber, 1999b). There is enormous variation in the level of employers' contributions to social security schemes in the European Union, yet this bears no systematic relation to the age structure of populations. Clearly, political factors rather than demographic ones influence statutory requirements on employers and pension policy generally.
An emotive theme concerning population ageing has been the supposed withdrawal of funds from child welfare. Yet there is no evidence of an inverse relationship between the amount of spending on older people and on children. Comparison of the balance of state spending on young and old suggests that, among seven major Organization for Economic Cooperation and Development economies, only Germany's spending had become biased towards older people between 1959 and 1986 (McDaniel, 1997). An associated argument that, if the population is ageing, welfare states create inequity between generations, relies on generational accounting techniques. These measure (or project) each cohort's financial inputs to the welfare state, in taxes and social insurance contributions, and receipts from it in terms of state welfare (Kotlikoff, 1992).
Such methods are flawed as measures of intergenerational equity (Ginn and Arber, 1999b). Among numerous other problems and omissions, generational accounting is insensitive to gender differences in the balance of paid and unpaid work and changes in this between cohorts. Unpaid care for both the younger and the older generation is a vital form of transferred resources between generations, incurring substantial pension costs for working age carers, mainly women (Ginn and Arber, 2000b). Because such care is socially constructed as non-work (Grace, 1998) it is ignored in generational accounts.
Myth 6. 'Private pensions are "actuarially fair"; they don't redistribute across the membership'
An argument sometimes used against state pensions is that redistribution across the population is not a proper function of pension schemes. The basic state pension redistributes explicitly towards the low paid, since the pension is flat rate (depending only on years of contributions), while NI contributions, which pay for both the basic pension and SERPS, are related to pensionable earnings (i.e. earnings between the Lower and Upper Earnings Limits).
Redistribution in private (occupational) pension schemes also occurs, although it is less well-publicised. As in state pension schemes, the non-married subsidise the married, irrespective of parental roles (see Table 5). For example, occupational pensions provide survivor benefits for married members. This increases the liabilities of the scheme while married members pay the same rate of contributions as the unmarried. Historically, a subsidy to marriage in the form of widows' benefits could be justified as fair to women, in that having children and being excluded from employment was closely associated with marriage for women, while lone parenthood was relatively rare. This is no longer the case, as the association between marriage, motherhood and non-employment has unravelled. The cost of redistribution towards married scheme members (including childless individuals whose spouse is also employed) is borne by non-married members (including lone and cohabiting parents). While cohabitees may receive sSurvivor benefits at the discretion of trustees of occupational pension schemes, this is far from universal and is not guaranteed. In contrast, personal pensions do not ostensibly redistribute towards the married, since annuity rates reflect marital status.
Table 5: Derived entitlements in state and private pension schemes, click HERE
Personal pensions depart from actuarial fairness in their treatment of women. The fund built from rebate-only contributions buys a sex-neutral annuity but any portion of the fund arising from contributions above this amount buys a lower annuity for women than men, based on greater average longevity of women. While this may seem fair, equally large mortality differentials due to race and class are ignored in setting annuity rates (Ginn and Dugard, 1994).
As is well-known, early leavers from occupational pension schemes (often women leaving for childbearing) subsidise the pensions of those staying in the scheme until retirement (mainly men and childless women). Moreover, those with a rising earnings profile (mainly non-manual men and childless women) gain far more from a Final Salary scheme than those with flat or declining earnings (manual workers and women whose career has been stalled by childbearing gaps and part time employment). Thus there are considerable cross-subsidies in occupational pension schemes, generally towards those who are advantaged in the labour market; mainly married men in non-manual occupations.
Myth 7 The new State Second Pension will help the low paid, especially women
The new State Second Pension (S2P) will treat the low paid, and certain carers, as if they had an income of £9500 per annum. On the face of it, this is a woman-friendly reform. However, its effectiveness in helping those with low earnings and periods of caring to escape poverty is undermined by price-linking the basic NI pension. By 2050, the basic pension (by then worth only 7.5 per cent of average earnings) plus the maximum amount of S2P will barely exceed the Minimum Income Guarantee (MIG), which is intended to be earnings-linked (Rake et al., 1999). Moreover, in payment the S2P, like the basic pension, will be indexed to prices and therefore fall relative to the MIG, catching pensioners in a savings and pensions poverty trap. Those with minor additional income from other sources or with modest savings may find themselves living at poverty level yet disqualified from claiming the means-tested MIG and associated benefits.
Many women will not qualify for a full S2P, which requires 49 years of contributions or credits. Carer credits for childcare are restricted to those with a child under age 6, unlike Home Responsibilities Protection (HRP) in the basic pension, which applies to those with children under age 16 or in full-time education.
Substantial reliance on means-tested benefits will therefore continue under New Labour's reformed pension system and women who have raised children will predominate among the poor because of their obstacles in obtaining private pensions.
Even worse, the new pension system will be regressive, placing disproportionate burdens on the low paid. Opting out of the S2P into a Stakeholder Pension (SP) will attract NI rebates, as is currently the case for those opting out of SERPS into a personal or occupational pension. But the rebates will be higher than the amount of SERPS/S2P saved; every £1 of SERPS rebate handed over to the personal pension providers costs 22p in public subsidy (PPG, 1998: Table 7.1). This means that those remaining in the S2P will effectively be penalised. In order to finance over-generous rebates for the higher paid who opt out, they will have to pay higher NI contributions than otherwise needed for a given level of benefits. Since only the low paid will belong to the S2P, any redistribution in favour of carers and the very low paid will come from those who are themselves low paid. This brings us to the final myth.
Myth 8 Taxpayers subsidise state pensions while private pensions are self-financing
It is often implied that contributions to private funded insurance are 'responsible', while Pay-As-You-Go (PAYG) social insurance represents a 'burden' on the economy. Yet PAYG schemes, like funded schemes, are mechanisms for transferring resources across the life-course and receive little or no input from general taxation. Moreover, there is no tax relief on NI contributions, while private pensions attract substantial tax relief. Due to cuts in NI benefits and the fact that over time contributions rise with earnings while NI benefits are only price-linked, the NI fund now has a substantial surplus. A recent report from the Government Actuary (2000) recommends that the NI fund should have a reserve balance of 1/6 of annual benefit expenditure. For the year 1999-2000, estimated benefit expenditure is £46.26 billion, the required reserve £7.71 billion. But the surplus in December 1999 was £13.77 billion, or £6.08 billion more than required. The expected surplus at December 2000 is £16.66 billion. Thus the NI fund, far from being a burden on taxpayers, is generating a surplus far greater than prudent management requires, by underpaying those entitled to benefits.
In contrast, public spending on private pensions escalated during the Conservative administration. Tax relief for private pensions grew from £1.200 million in 1979 to £8.200 million in 1991 (Wilkinson 1993), while the total net cost to the NI fund of incentives to transfer from SERPS to personal pensions was estimated as £6 billion (1988 prices) over the period from 1988-93 (National Audit Office 1990). In 1994/5, tax expenditures, rebates and incentives to occupational and personal pensions totaled £20 billion (Glennerster and Hills, 1998).
The social security budget has a much higher profile than spending on tax reliefs and rebates for private pensions. The little known public subsidy to private pensions has been branded as 'reverse targetting' (Sinfield 1993: 39). It siphons off resources which could have been used to maintain the value of state pensions, and provides a substantial benefit to those with private pensions, especially where earnings are high. Moreover, increasing private pension coverage, facilitated by tax spending, removes the sharp elbows of the middle classes from defending state pensions and contributes to the crisis of legitimacy surrounding public provision (ISSA, 1995). New Labour, despite a stated intention to target benefits on the poorest, is continuing with a policy which directs public funds towards the better-off and further weakens state welfare provision.
Conclusion
The last Conservative government, in keeping with neo-liberal ideology, shifted resources from public to private pensions. Yet New Labour, despite their stance in opposition, has reinforced the privatisation of pensions. In order to legitimate their pension policy, both Conservative and Labour governments resorted to myth-selling concerning pensions, capitalising on widespread ignorance and confusion about this very complex area of policy. It is women whose economic security in later life is most adversely affected by the power of these myths to make socially divisive pension policies appear acceptable.
The private finance sector has, not surprisingly, been a willing partner in myth-selling. For example:
Anyone who is naïve enough to think that the State will be able to provide for their retirement is seriously mistaken. We have an ageing population and a declining birthrate - the mathematics say the money just won't be there. The clear message from the Government is to start making your own pension provision NOW! (Tesco Personal Finance leaflet, February 2000).
As this article has indicated, the supposed economic-demographic imperative is a social construction. Better state pensions are affordable. For example, the net cost of raising the basic pension by £20 per week (and a corresponding increase for pensioner couples) has been estimated as £5.8 billion per annum, or £2.1 billion per annum if the Upper Earnings Limit on contributions were removed (Hancock and Sutherland, 1997). These are well within the NI fund surplus. A more modest increase in the basic pension, to the level of the Minimum Income Guarantee (i.e. from £67 to £78 per week in 2000 for a non-married pensioner) which was recommended by the Paying for Age inquiry (Birch et al., 1999) would cost £3 billion per annum. On present policies, the NI contribution rate required to fund state pensions is set to fall (Birch et al., 1999). The issue is whether resources should be increasingly directed into the private sector at the expense of the public. The privatisation of pensions and of other aspects of the welfare state can be seen in most neo-liberal states and in countries under their economic domination through the World Bank and the International Monetary Fund. These policies can only be understood as stemming from the 'public burden' model of welfare and the individualistic ideology which have characterised the thinking of policy-makers in Britain since the 1980s. Women are the chief losers from this way of thinking.
REFERENCES
Birch, R., Hancock, R., LeGrys, D. and Roberts, R. (1999), Paying for Age in the 21st Century. The Millennium Papers, London: Age Concern England.
Dilnot, A., Disney, R., Johnson, P. and Whitehouse, E. (1994), Pensions Policy in the UK. An economic analysis, London: Institute of Fiscal Studies.
Durham, P. (1994), 'Millions will lose money on private pensions', Independent, March 28.
Department of Social Security (1998), Partnership in Pensions, Cm 4179, The Stationery Office.
Falkingham, J. and Victor, C. (1991), The Myth of the Woopie? Incomes, the elderly and targetting welfare, WSP/55, London: STICERD.
Ginn, J. (1993), 'Pension Statistics Ignored', Radical Statistics, No. 53: 13-15.
Ginn, J. and Arber, S. (1991), 'Gender, Class and Income Inequalities in Later Life', British Journal of Sociology, 42(3): 369-396.
Ginn, J. and Arber, S. (1993), 'Pension Penalties: the gendered division of occupational welfare', in Work, Employment and Society, 7(1): 47-70.
Ginn, J. and Arber, S. (1999a), 'Changing patterns of pension inequality: the shift from state to private sources', Ageing and Society, 19(3): 319-42).
Ginn, J. and Arber, S. (1999b), 'Gender, the generational contract and pension privatisation', chapter 8 in S. Arber and C. Attias-Donfut (eds) The Myth of Intergenerational Conflict: the state and family across cultures, London: Routledge.
Ginn, J. and Arber, S. (2000a), 'Personal pension take-up in the 1990s in relation to position in the labour market', Journal of Social Policy.
Ginn, J. and Arber, S. (2000b), 'The Pension Costs of Caring', Benefits.
Ginn, J. and Dugard, P. (1994), 'Statistics: a gendered agenda', Radical Statistics, 58: 2-14.
Glennerster, H. and Hills, J. (1998), The State of Welfare: the economics of social spending, Oxford University Press, Oxford.
Glennerster, H. (1999), 'Which welfare states are likely to survive?' International Journal of Social Welfare, 8: 2-13.
Government Actuary Department (2000), Cm 4587, Feb, London: The Stationery Office.
Grace, M. (1998), 'The work of caring for young children: priceless or worthless?', Women's Studies International Forum, 21(4): 401-13
Hancock, R. and Sutherland, H. (1997), Costs and Distributional Effects of Increasing the Basic State Pension, London: Age Concern England.
Hancock, R. and Weir, P. (1994), Pensioners' Incomes 1979-1989: Detailed analyses of the General Household and Family Expenditure Surveys, London: ACIOG.
Hills, J. (1995), 'The welfare state and redistribution between generations', chapter 3 in J. Falkingham and J. Hills (eds), The Dynamic of Welfare. Social policy and the life cycle, Hemel Hempstead: Harvester Wheatsheaf.
International Social Security Association (1995), Social Security Tomorrow: Permanence and change, Geneva: ISSA.
Lloyd-Sherlock, P. and Johnson, P. (1996), Ageing and social policy. Global comparisons, London: Suntory-Toyota International Centre for Economics and Related Disciplines.
Kotlikoff, L. (1992), Generational Accounting: Knowing who pays and when for what we spend, New York: Free Press.
Mabbett, D. (1997), Pension Funding: Economic Imperative or Political Strategy? , Uxbridge: Brunel University.
McDaniel, S. (1997), 'Intergenerational transfers, social solidarity and social policy: unanswered questions and policy challenges', Canadian Journal on Ageing/Canadian Public Policy, Supplement, Spring: 1-21
Murthi, M., Orszag, M. and Orszag, P. (1999), 'Administrative costs under a decentralised approach to individual accounts: Lessons from the United Kingdom', paper presented to conference on 'New Ideas about Old Age Security', The World Bank, September.
National Audit Office (1990), The Elderly: Information requirements for supporting the implications of personal pensions for the National Insurance Fund, HC 55. HMSO, London.
Pensions Provision Group (1998), We All Need Pensions - the prospects for pension provision, London: The Stationery Office.
Quadagno, J. (1996), 'Social Security and the Myth of the Entitlement Crisis', The Gerontologist 36:391-399.
Rake, K., Falkingham, J. and Evans, M. (1999), Tightropes and Tripwires: New Labour's proposals and Means-testing in Old Age, Casepaper 23, London: STICERD.
Sinfield, A. (1993), Reverse targetting and upside down benefits - how perverse policies perpetuate poverty, pp. 39-48 in A. Sinfield (ed.) Poverty, Inequality and Justice, Edinburgh: Edinburgh University Press.
Toporowski, J. (2000), The End of Finance, London: Routledge.
Wilkinson, M. (1993), British tax policy 1979-90: Equity and efficiency. Policy and Politics, 213, 207-17.
World Bank (1994), Averting the Old Age Crisis, New York: Oxford University Press.
Jay Ginn
Sociology Dept
University of Surrey
Guildford GU2 5XH
E-mail: j.ginn@surrey.ac.uk
