[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries

State pensions and social solidarity

P roponents of pay-as-you-go (PAYGO)state pension systems often cite equity between generations, social solidarity, and social justice as advantages of such systems. The EU green paper on supplementary pension provision describes pillar one pensions as providing ‘solidarity within and between generations’ and ‘promoting social welfare’.

Realistic goals?
The provision of state PAYGO pensions involves one generation voting for pension promises which are to be funded by a generation which does not yet vote and may not yet be born. This is often described as an ‘intergenerational contract’, despite only one side being available to sign it! The fulfilment of those promises requires a stable demographic profile. But there is no control mechanism within the PAYGO system to ensure that the generation taking the decision will produce the requisite number of children. It seems difficult to conceive of a policy more likely to create social friction than one where a given generation votes pension costs on to the following generation without providing the means to meet those costs. Kessler (1996) asks what will happen if today’s younger generation decide that they do not want to have their standard of living fall as a result of pension promises which have been made to future generations? Will they express their dissatisfaction by non-political means as the elderly form a greater proportion of the electorate? Will the government be able to take the necessary action to move towards sustainable systems of funding, or will the transition costs make such a move beyond the bounds of political possibility?
In a system of private provision, conflicts are resolved by voluntary and enforceable contracts. In a state PAYGO system, resources are allocated by a process of competition between interest groups which try to influence the political system. This process is described as ‘public choice economics’. The outcome of this process may or may not lead to a more equitable distribution of income but it is not obviously more just or more liable to create solidarity in any real sense of the words; conflict is more likely. As one sees the friction between interest groups (for example pensioners and the unemployed) in countries with high PAYGO pension provision and high labour-market taxes, it is certainly not clear that such schemes produce solidarity.
The redistribution between generations in PAYGO systems is entirely arbitrary. Wealth does not flow automatically from rich generations to poor generations. Thomson (1992) has calculated that the welfare system in New Zealand has transferred 18 years of average pay from the generation born around 1955 (the lates) to that born around 1930 (the earlies). This has happened as a result of a combination of the following:
– the earlies receiving pension benefits to which they did not contribute;
– the lates meeting the transition costs from a PAYGO scheme; and
– the lates being insufficiently numerous to meet the costs of PAYGO pensions to the earlies without considerable increases in social security taxes.
There are also a number of non-pension issues, such as the earlies receiving free higher education and benefits for having children, which have since been withdrawn for today’s younger generation.
The implicit debts of PAYGO social security schemes are well documented. For example, according to Stein (1997) the accumulated liabilities of state pension schemes are 160% of GDP in Germany. These debts are genuine, off-balance sheet debts which represent the payments which have been promised to future generations of pensioners in terms of benefits already accrued in state systems, which are not represented by assets. They represent an intergenerational transfer which can never have been intended by those who set up the system.
It is impossible to foresee the redistributive consequences of PAYGO pension schemes between generations; it is therefore not possible to apply notions such as social solidarity to their operation. In the absence of the creation of any legal property rights, even the term ‘security’ in ‘social security’ is of dubious meaning.

PAYGO and funding: the fundamental difference
The mechanisms for providing state PAYGO pensions and privately funded pensions are fundamentally different. In the former case, it is hoped that there will be a sufficient number of children who then participate in the labour force to meet future pensions costs. A stable demographic profile is required, but there is no control mechanism in the system to ensure that the stable profile can be maintained. With funded pensions, saving takes place which leads to investment; the economic return from that investment enables consumption to be deferred from one generation to the next, without the following generation making a sacrifice. While many actuaries do not appear to recognise the role that the capital stock plays in the economy, it is a factor of production, just as is labour (although it must be accepted that a considerable change in the age profile of the population may lead to changes in consumption and saving patterns, which could lead to frictional problems in an economy and so to a requirement for relative price changes). Funding can also be regarded as a control mechanism which allows contributions to be increased to prevent deficits arising.
If a PAYGO system runs into financial difficulties because the demographic profile of the population changes, the control mechanism in a PAYGO system is to increase taxes. French social insurance taxes are equal to about 60% of payroll. This has a detrimental effect on labour force participation, thus worsening the problem. Continental European countries with a high level of unfunded provision have very high labour taxes and some of the lowest labour participation rates in the world.
Even the consequences for the distribution of wealth and income within a generation of PAYGO pension schemes are arbitrary, unknown, and, indeed, unknowable. It is difficult to ascertain the redistributive effects of a PAYGO scheme because contributions do not always relate to benefits; contributions are normally paid for longer by those who are poorer (because they spend less time in full-time education) and benefits are normally received for a shorter time by those who are poorer (because their life expectancy is less). While it is difficult to determine the redistributive consequences of PAYGO pension provision, there is no reason to assume that it will necessarily be progressive. The most authoritative study is probably that by Creedy, Disney, and Whitehouse quoted in Dilnot et al (1994). It finds that there is virtually no redistributive element in the UK state pension system (with SERPS included). Beach and Davis (1998) show a very ambiguous picture for the US. The ‘rate of return’ on social security pension contributions does appear to decrease as income increases, thus leading to a redistributive element. It should be noted that this study has been criticised by Myers (1998) as having a flawed actuarial methodology.

Is there another way forward?
All that may be achievable is the establishment of a sustainable framework of funded pension provision. Such a framework does not create absolute certainty as to the income an individual will get in old age but it is sustainable and secure (in the sense of not being self-destructive). Those in need can then be helped in other ways. Creating an overarching system of PAYGO pensions may be an attempt to achieve the impossible, which then undermines the economic security of those it is designed to help.
If the aim is to redistribute income within generations, this can be done through the provision of means-tested benefits. It could be said that means-tested benefit provision erodes both work and savings incentives. However, breaking the link between contributions and benefits has exactly the same economic effect as means-testing (people pay taxes on extra earnings but do not accrue benefit). Indeed, it is notable that the originators of the Beveridge system regarded the relationship between contributions and benefits as important, as ‘no means testing’ is often regarded now. Nevertheless, if there is a preference to avoid means-testing, contributions can be topped up for those on low earnings instead. With regard to redistributing income between generations, the proposition that state PAYGO pensions systematically redistribute money from rich to poor generations, thus creating ‘intergenerational solidarity’ is surely unsustainable.
The author would not wish to have a state pension system which was funded through investment in state-owned assets (such as is partially the case in Singapore). However, if the specified goal is to produce a system which is controlled by the state which incorporates mechanisms for income redistribution, a funded state system would seem more appropriate for achieving these than a PAYGO system, which can suffer from massive, unintended transfers of wealth between and within generations.