Social security and the challenge of demographic change


From 29 November to 4 December 2010, the International Social Security Association (ISSA) will meet in Cape Town, Republic of South Africa, to mark the event of the ISSA World Social Security Forum. The Forum provides a unique opportunity for decision-makers from all regions to share knowledge, recognize good practices and discuss key policy challenges as these relate to the design and delivery of national social security programmes. One key policy challenge identified by the ISSA's worldwide membership is demographic change. For this important reason, among the events planned for the Cape Town Forum, a plenary will focus specifically on demography.

To coincide with the preparations for the World Forum, and to complement the wider and longer-term endeavours of the ISSA to promote knowledge sharing, the International Social Security Review has chosen to produce this double special issue on “Social security and the challenge of demographic change”.1 The expectation is that this set of papers will make a contribution to supporting social security policy-makers, practitioners, analysts and researchers in all countries as they work towards developing and implementing tailored policy responses to the multifaceted challenge of demographic change.

The demographic context

The world is undergoing demographic change, and population ageing is a central element of that. Population ageing, defined as the growth of older age groups (aged 60 or older; aged 80 or older) as shares of the total population, is occurring in all regions of the world. In particular, Figures 1-3 show the percentage of population aged 60 or older by region for the period 1950 to 2050; the percentage of population aged 80 or older by region for the period 1950 to 2050; and life expectancy by region for the period 1950 to 2050.2 Overall, the number of individuals aged 60 or older is projected to reach 1 billion by 2020 and will approach 2 billion by 2050, representing 22 per cent of the world's population. The number of the “oldest-old” (aged 80 or older) is projected to rise from around 90 million today to more than 400 million in 2050, representing 4 per cent of the global population by 2050. Remarkably, global life expectancy has increased from 47 years in the early 1950s to 68 years in 2010, and is expected to reach 75 years by 2045. This process of population ageing is historically unprecedented; however, the phenomenon is not uniform across countries. The extent of ageing varies most between the more- and less-developed regions.3 In the most-developed countries, one-fifth of the population is already aged 60 or older, with the expectation that this share will rise to more than 30 per cent by 2050. In many developing countries, less than 10 per cent of the population is aged 60 or older. Although the pace at which different populations will age also varies, the proportion of those aged 60 or older in developing countries is projected to at least double in the coming four decades. For example, by 2050, those aged 60 or older will represent 20 per cent and 30 per cent of the populations of India and the People's Republic of China, respectively. For all countries, more- and less-developed alike, the recognized phenomenon of elderly females outnumbering elderly males will continue, albeit in a less pronounced manner than in the past. As a consequence of improvements in health at all ages, the male death rate is expected to decline.4

Figure 1.

Share of population aged 60 or older, by region
Source:ESA (2009).

Figure 2.

Share of population aged 80 or older, by region
Source:ESA (2009).

Figure 3.

Life expectancy at birth, by region
Source:ESA (2009).

Three main factors serve to explain past and projected changes in the share of the population aged 60 or older. Declining fertility rates is the first, because this change reduces the proportion of young people in relation to the elderly. An increase in life expectancy is a second factor, and is seen both in developed and developing countries. On a global scale, this trend prevails despite the negative impact of the HIV and AIDS pandemic in some low- and middle-income countries and also despite a decrease in life expectancy, particularly among males, in some countries of Central and Eastern Europe and the former Soviet Union during the 1990s. The third main factor is past variations in birth and death rates. The impact of the ageing of the post-Second World War “baby boom” generation on high-income country population pyramids is an excellent example in this regard. Similarly, advances in primary and maternity care in developing countries that contribute to improved maternal and child survival and subsequent temporary increases in fertility rates will inevitably lead to a bulge in the numbers of the elderly as these cohorts age.

It is important to note that these outlined demographic changes are taking place in the context of existing social security laws, institutional structures and benefits programmes. Although the scale and scope of social security provision can vary greatly between countries, the demographic changes already in play will affect nearly all countries significantly, with ramifications for national labour markets. That the most predominant mechanism for financing social security worldwide remains social insurance, wherein covered workers and their employers are mandated to pay contributions, demographic change that leads to a decline in the size of the formal labour market is likely to have a negative impact on the financing of social security provision. This is especially so for pay-as-you-go (PAYG) social insurance programmes, where current contribution revenue from the active population (workers and employers) is the main source of income used to finance current benefit expenditure destined for the non-active population. In the worst-case scenario, if the ratio of active contributors to current beneficiaries declines to a point where the value of expenditures on beneficiaries exceeds the value of income from contributors, then, in the absence of sufficient other income from reserve funds, returns on investments or State subsidies, the PAYG programme will be unable to meet its commitment to finance current benefits. However, such a possible eventuality is not restricted to the impact of demographic change alone. Financial disequilibrium can also result from contribution evasion and underpayment, in particular if this coincides with structural changes to the labour market that lead to high levels of unemployment and/or a serious decline in the size of the covered active population. For example, in Serbia and Bosnia and Herzegovina, large-scale informal employment has contributed to the current very low ratio of active social security contributors to pensioners (1.25:1). As a comparative measure, this current ratio is lower than the projected ratio for the countries of the European Union (EU) once the baby boom generation has retired (1.4:1) (ISSA, 2010a).

Important questions for national social security programmes

Financial disequilibria of the kind found in Serbia and Bosnia and Herzegovina should be viewed as extreme events. More generally, to explain the common challenges for social security programmes that derive from current global patterns of demographic change, three main elements must be considered: increased longevity, falling rates of fertility, and evolving migration flows. The papers in the issue collectively draw attention to the potential impact of these elements on the functioning of national social security programmes. In turn, this necessarily leads to a series of important questions, the responses to which will vary according to the country or particular type of programme in question.

  • • To what degree will demographic change negatively impact the capacity of societies to adequately finance existing social security programmes through risk-pooling social insurance contributions, the allocation of tax revenues by government, and mandated individual savings?
  • • To what degree will demographic change negatively impact capacities to adequately finance the implementation of new social security programmes, not only for the vast majority of the global population who are currently inadequately protected but to address the evolving and new income protection, health and long-term care needs of populations in ageing societies?
  • • To what degree can existing social security programmes, which were largely designed for the family structures, social needs and lifecycle risks of populations of the second-half of the twentieth century, be considered appropriate and equitable for the family structures, social needs and lifecycle risks faced by modern, increasingly mobile, populations?
  • • To what degree might the ageing of national workforces, including the ageing of public servants, impact the effective day-to-day administration and delivery of social security programmes?
  • • To what degree might demographic change negatively impact political commitment and public attitudes towards social solidarity, the core value underpinning social security programmes?

There is a significant element of uncertainty surrounding the answers to all the above questions. For many Member States of the Organisation for Economic Co-operation and Development (OECD) in particular, such uncertainty as regards how to adequately finance social security programmes in the coming decades is not a new one. Notably, this question was broached by the International Labour Office (ILO) in relation to Europe as early as 1987, with findings subsequently published in 1989 in a volume entitled From pyramid to pillar: Population change and social security in Europe (ILO, 1989). The ILO reported that an awareness of the challenges posed by demographic ageing trends for European social security programmes had arisen initially in the 1950s. The ILO underlined, however, that the “concern” (ILO, 1989, p. 85) voiced in the 1950s about population ageing had not resulted in policy action, particularly the reform of pension systems. Across Europe, successive incumbent governments appeared unwilling to tackle the tricky issue of pension reform, especially when the problem to be addressed lay decades in the future. It would not be until after the oil shocks of the 1970s, a period marked by recession and high levels of unemployment, that the question of the cost of social security would become important and the general public would become truly aware of the challenges of demography for social security financing (ILO, 1989, p. 86). It is plausible to suggest that despite this rise in awareness in the 1970s, some governments nevertheless opted to continue to side-step the demographic challenge — a challenge that could still be comfortably ignored and left to a future generation of politicians.

In more recent years, it is possible to discern a similar pattern of events. A full five years after the ILO released From pyramid to pillar, the World Bank published its influential policy research report Averting the old age crisis (1994). Challenging the policy vision adhered to by the ILO,5 the Bank's promotional message that accompanied the release of the report was deliberately sensationalist and headline-grabbing: “existing systems of financial security for old people are headed toward collapse”. While the Bank's policy conclusions have been, and remain, the focus of serious contention, Averting made a major positive contribution to taking discussions of the demographic challenge out of highly specialized technical meetings and placing them, more firmly than ever, in the public realm. Since 1994, the different responses offered by the ILO and the World Bank to the challenge of demographic change and to the social policy needs of older populations have continued to evolve, including greater policy focus by both of them on poverty reduction. At the national level over recent years, reforms to strengthen the longer-term health of pension system finances have been commonplace. In spite of all this, and remaining mindful of the fact that the recession years of the 1970s did not result in significant structural change in a majority of countries' social security systems, it remains to be seen in what manner the most recent serious macroeconomic shock may provide the necessary catalyst to transform heightened political awareness of demographic change into active and sustained political will.

It also remains to be seen whether increasing pressure for greater fiscal control will act to limit what is financially possible for political actors in terms of feasible reform options. For instance, it is often argued that structural pension system reforms involving the introduction of mandatory individual accounts (MIAs) may be one way to address the predicted financial challenge posed by population ageing for the sustainability of conventional defined benefit PAYG pension systems. Such reform, resulting in the creation of a multi-pillar pension system,6 is aimed at reducing the role of the PAYG pension “pillar”. However, it also involves high transition costs over several decades. An important element of the transition costs are the public subsidies that must be allocated to the “old” PAYG pillar. Specifically, with a portion of pension contributions diverted to the “new” MIA system (which implies a commensurate fall in contribution revenue for the PAYG system), public subsidies are necessary to finance adequately the defined benefits of current pensioners and, in future years to come, of older workers (usually, those aged 40 or older at the time of the reform) who are excluded from the MIA system. It is not yet clear whether, in the current generalized context of reduced fiscal leeway, these transition costs will challenge arguments in support of the MIA option and make reforms aimed at strengthening the financial health of existing defined benefit PAYG programmes more fiscally and, indeed, politically attractive.7

Certainty and uncertainty regarding population ageing and social security systems

Population ageing is now a reality in most societies, and thus it is no longer merely a looming problem on the longer-term political horizon that may be conveniently or easily ignored. As a consequence of post-2008 public spending cuts and, in some countries of Europe, moves towards fiscal austerity, the time may finally have come for sustained and meaningful political action to address the challenge demographic change presents to different national social security programmes and systems. But this is far from certain, given the last sixty years of political debate on this issue. One thing is sure, however. If politicians do not act at least as quickly as one would hope, it will not be because the trends of demographic projections can be doubted.

It is sometimes suggested that projecting demographic change is, at best, an inexact science, wherein even the smallest of changes in initial input variables necessarily demand that projections require revision. This view is certainly not wrong, but it seriously overestimates the degree of error found in demographic projections. To the contrary, accurately projecting the broad direction of demographic trends may be considered more straightforward and sure than is the case for projecting other social and economic changes. For instance, making projections based on the limited number of key parameters of demographic change (e.g. the death rate, the fertility rate and migration flows) can be more confidently done, and consistently so, than is this case for projections anticipating changes to, for example, the myriad interwoven factors that contribute to the growth rates and productivity of national economies. When the subprime crisis occurred in the United States in 2007, few would have been able to confidently project the subsequent negative social and economic effects that have since impacted the global economy and, even more so, specific national economies.

Accordingly, and despite the sensitivity of demographic projections to even small changes in variables, they nonetheless enable a degree of certainty in projections about population changes. For example, from a global perspective, it can be stated with certainty that the absolute and relative number of older individuals will increase. It can be stated with certainty that within the global population, the relative number of children will decrease. It is also certain that, increasingly, a greater proportion of the world's population will live in cities.

However, as regards the impact of demographic change on the future nature and roles of social security programmes there is a relative degree of uncertainty. Already at the level of social security institutions, an important challenge is often a lack of robust data profiling all scheme members and their dependants. This is particularly true for developing countries where institutional capacities to collect contributions and disburse benefits in an effective and efficient manner may also be weak. All things being equal, these challenges have a direct bearing on efforts to ensure that benefit programmes remain adequately financed. However, when demographic change must be factored in, this task becomes doubly complicated.

As previously mentioned, most national social security programmes are financed from current national revenue. Specifically, a certain portion of national revenue, normally collected through a combination of mandated social security contributions and taxes, is redistributed to finance benefits and services. Broadly speaking, projections about how much revenue should be collected, and to whom and in what amounts it should be disbursed, are based on assessments of the ratio of working-age population to the total non-active or dependant population (both young and old). It is clear that major changes to population age structures, which will significantly alter the “young” and “old”, but not necessarily the total, dependency ratio on a longer-term basis, demand that the manner in which social security programmes are financed, and the decisions about how revenue is disbursed, be revisited. As evidenced since at least the 1950s, the challenge for social security programmes lies with the observation that the political economy of social security reform is often out of step with evolving demographic structures and the pace of population dynamics.

Posing a further challenge to policy-makers is the uncertainty regarding possible changes in individual behaviour. For example, despite the certainty that populations in nearly all countries are ageing, there is less certainty about how many “older” citizens are likely to remain active in the labour market if given the opportunity to do so.8 To address this uncertainty, OECD countries have employed various policy tools to keep older workers working. These tools include raising the pensionable age, increasing the number of years of work required to be eligible for a full pension, and tightening the eligibility conditions for receiving an early pension. Linking benefit entitlements more closely to personal contribution records, either through the use of defined contribution plans or notional defined contribution systems, ostensibly with the aim of encouraging individuals to work longer to earn higher benefits, is another. As the evidence from mandatory individual accounts in Latin America and elsewhere has revealed,9 defined contribution plans have not met expectations about their ability to encourage workers to contribute more to social security. In addition to unsatisfactory investment returns and high administration costs, low levels of contributions have contributed to the fact that many mandatory individual account systems provide inadequate benefits. This problem is particularly difficult for low-income and irregular earners, groups that often include a disproportionately high number of women (ISSA, 2010b). More generally, and despite these and other approaches and adjustments introduced by countries, older workers often use alternative “social security” routes to exit early from the workforce: e.g. disability and unemployment benefits.

In many countries the average age at which workers exit employment is much younger than the legal pensionable age. This behaviour pattern became rooted in the 1970s when it was encouraged in many countries as a means of responding to high levels of unemployment. In a positive development for social security finances, between 2001 and 2007, the general trend in the EU reversed, and saw a rise in the average age at which workers exit employment, from age 59.9 to age 61.2 (EC, 2009, Table 9). However this is still lower than age 65, the most common pensionable age for women and men in EU Member States. Worldwide, a majority of countries apply the same pensionable age for women and men, and others such as Australia, Hungary, Latvia, Romania, Slovakia and the United Kingdom are gradually raising that of women to equal that of men (SSA and ISSA, various years). Changes to family structures, cultural values, employment patterns and women's access to education as well as improvements in life expectancy all bring into question the practice of women retiring earlier than men. Lower retirement ages for women, but longer life expectancy in retirement (and, possibly, widowhood), combined with women's often more fragmented and less-well-paid employment histories, and thus lower levels of social security contributions, all add up to a greater risk of women living alone in poverty in old age (ISSA, 2010a).

If the EU trend in the rise in the average age at which workers exit employment is followed more generally, this will be important for at least two reasons. First, the continuing activity of older workers is likely to defer, and possibly reduce, their need for financial support through social security systems. If engaged in paid employment, they will also continue to contribute to social security. Second, a higher level of labour force activity will bolster national economic output, thus strengthening fiscal receipts that in turn will support public expenditures including those directed to tax-financed social security programmes. On these grounds, it becomes important to better gauge the likely success of policy initiatives that, on one hand, seek to remove the legal and cultural barriers that often prevent older workers from remaining active and, on the other hand, that seek to influence individual behaviour by encouraging older workers to defer retirement. A further related question is the degree to which improvements in health indicators and a greater emphasis on preventive primary health initiatives will enable healthier “active” ageing (which would be abetted by a health and safety framework and re-modelled work processes tailored to the needs of older workers). As labour markets become ever-more globalized, a final consideration is the kind of work that will be available for younger and older workers alike. Even prior to the global downturn, labour market trends were moving towards more precarious and vulnerable employment, with increases in temporary and part-time work and informality. For all countries, the longer-term impacts of these developments are likely to be to the detriment of social security and public finances and the adequacy of individuals' contributory and tax-financed benefits.

Of course, regardless of how these developments evolve, sustaining national economic output — which has a positive spin-off for social security financing — depends on more than labour force effort. Other factors that contribute to economic output include the level and type of investment made by owners of productive assets; the time frame over which companies seek to see returns on investment; the level of health and education of the population; the level, type, and incidence of taxes; government interest rate policy; exchange rate policy; innovation in technologies and organizational processes; and, of course, the international economic environment. This list of intervening elements acts to remind us that in seeking feasible policy solutions to the challenges of demographic change, all policy, including social security policy, must be designed strategically as part of a coherent whole.

Demographic change implies not only a need to ensure adequate financing in future, but may also require a political renegotiation of the social contract. For the success of the latter, the nature and strength of political will, and the importance to be accorded to truly inclusive and participatory social dialogue, may be determinative. Renegotiating the social contract poses a number of questions, such as how might evolving demographic structures, involving a relative increase in the older population compared to that of the younger population, impact political decisions regarding the relative distribution through social security programmes of national income across generations? In ageing societies, will there be less need to invest in family benefits programmes? Or, might some countries favour greater investment in family programmes as part of orchestrated efforts to support families with young children? Although there is little evidence to suggest that generous family benefit programmes alone lead to higher rates of fertility (ISSA, 2010c), the need to bolster human capital formation to support future higher national economic output and, in some countries, the imperative to allay geo-political concerns over population decline may strengthen political support for family benefits programmes. Another question is how might the demographic shift towards older societies impact the future role of social solidarity vis-à-vis individual responsibility in the design of social security systems? For most countries, all reforms that may lead social security to reduce the role of social solidarity imply a deliberate policy choice. However, it remains to be seen whether such a decision might hinge as much on the reduced financial capacities of States to ensure continuing levels of public provision as it might on the strength of any ideological arguments to promote individual responsibility. Again, there is uncertainty. Clearly, the significant degree of uncertainty in many countries about the rate at which the global economy will grow in the short to medium term, which will impact the space for fiscal manoeuvring by States, will have a bearing on all of the above. This is of particular relevance for developing countries that will most probably confront the challenge of growing old before growing rich. Significantly, by 2040, projections suggest that 75 per cent of the global population aged 65 or older will live in developing countries (ESA, 2009). However, despite these uncertainties, one assertion that can be made with a degree of assurance is that all countries will develop social security solutions that are specific to, and shaped by, their own institutional history, culture and values.

Identifying and addressing the challenges of demographic change

A first objective of this special issue is to present substantive papers that discuss all of the above as well as broader, related topics with the aim of highlighting the challenges of demographic change for social security that are already arising and anticipating others that will become prominent soon.

All countries have some form of institutional social security. In practice, the scale and scope of these social security programmes vary. In OECD countries social security provision is typically comprehensive (with provisions for old age, disability, survivorship, maternity, work injury and occupational diseases, unemployment, family allowances, medical benefits, and for a growing number of countries, long-term care) and universal (ostensibly, covering 100 per cent of the target population). In contrast, social security coverage in lower-income countries in sub-Saharan Africa and South Asia may extend to less than 10 per cent of the population and may provide coverage against a limited number of contingencies (e.g. old age, disability, survivorship, and work injury). Between these two extremes, a majority of countries have intermediate levels of coverage in terms of the risks that are covered and the proportion of the population that has access to such protection. On the basis of global estimates, it is thought that only 20 per cent of the world's working-age population has access to adequate social security and about one-third is without access to any health-care coverage (ILO, 2010, p. 32; p. 43).

Across countries, the institutional structures and mandates of social security administration vary as well. In some countries, social security provisions are highly centralized and managed by the State. In others, social security is institutionally fragmented according to the covered risk or geographic region and the State may be but one stakeholder in a context that includes civil-society and market-based provision. Alongside these institutional factors, public attitudes and expectations and political choices about the aims of social security programmes, and not least the relative balance of the roles of social solidarity and individual responsibility in pursuing these aims, also differ.

These differences and preferences, even if only of a relative nature, all act to define the characteristic features of national social security provision. Nonetheless, and despite often significant national specificities, social security programmes worldwide are confronted by a number of common challenges. Of these, the demographic shift to ageing societies is one of the most important. The identification of common policy challenges facing social security programmes is important, because it may permit the development and sharing of possible common policy lessons and the identification of good practices to inform possible responses. Nevertheless, these steps alone are rarely sufficient to effect change, which typically requires some degree of political negotiation and compromise. As the historic 2010 health care reform in the United States firmly underlines, the political economy of social security reform is often slow and difficult.

The observation that the reform of social security programmes is seldom, if ever, a “quick-fix” presents a paradox: a common assertion found across the papers in this issue is one of a need for a greater sense of urgency in policy reform. This is not done to provoke panic or to scaremonger. Rather, it is based on the matter-of-fact assessment that the “world is entering substantially uncharted waters regarding the size of its elderly population” (Bloom, Canning and Fink, 2010, p. 297). By entering into such uncharted waters, there is a pressing need for new thinking that can inspire policy-making and practice. Although population ageing is the product of declining fertility as well as increased longevity, combined with the dynamic evolution of past variations in birth and death rates — and is thus a phenomenon that impacts all generations — public old-age pension systems remain the cornerstones of most national social security systems. For these reasons, a recurrent theme — which is addressed directly through two national case studies — is the priority given to the need for new thinking to guide efforts to better ensure adequate income security for the elderly.

Of course, in any search for new policy thinking, a first challenge is to acknowledge that there is indeed a problem. In this regard, the case of the Russian Federation presented by Nicholas Eberstadt and Hans Groth is illustrative: Russia's future difficulties in financing all aspects of social security remain widely unrecognized. In their search for responses, Eberstadt and Groth suggest that Russia pushes the challenges of financing and delivering social security to the “frontier”, since all the current policy palliatives proposed for dealing with ageing societies do not sit easily in the Russian context. Eberstadt and Groth focus on Russia's demographic crisis and the implications it holds for the ability of the Russian government (or the Russian people through their own efforts) to generate enough funds to provide a reasonable level of old-age economic security. One factor that adds significantly to the problem of Russia having a moderately ageing society is that Russians are disproportionately less healthy than residents of other countries at similar — and even at much lower — levels of income. Thus, a given ratio of working-age to non-working-age people will not be accompanied by as high a share of workers as in other countries. The authors' analysis also concludes that revenues from oil production will be inadequate to finance pensions for Russia's elderly. Although the arguments presented focus on pensions, the authors point out that the same factors that will make it difficult to deliver adequate pensions also mean that other aspects of social protection will also be difficult to fulfil. Within the scope of this issue, Eberstadt and Groth can only begin to address the possible policy options: reducing the mortality rate of working-age individuals, finding a way to strengthen higher education, and fundamentally reforming institutions and economic policies. The Russian case is salutary in that it discusses issues of the urgency, as well as of the policy complexity, of responding to the challenges for social security systems brought about by demographic change.

In the second national case study, David Bloom, Ajay Mahal, Larry Rosenberg, and Jaypee Sevilla discuss ageing and economic security issues in the context of India. They first review the various factors that are affecting the ability of India's elderly to rely on traditional family ties to ensure their well-being, and the fact that the fraying of such ties, in conjunction with increased longevity, has led to an increased risk of the elderly falling into poverty. They highlight the particular risks endured by low-income workers (including a majority of the informal sector), migrants, and women, with respect to both pension coverage and health insurance. After explaining the array of types of contributory and tax-financed (mandatory and voluntary) pension and other social programmes countries can adopt, they discuss these from a holistic social protection perspective with particular reference to India. They also look for lessons India's experience may hold for other countries. Most saliently, an assessment is provided of India's experiences in implementing social protection programmes, with important consideration given to the efficiency gains to be gleaned from appropriate public-private partnerships and capital investments in information and communication technology (ICT) platforms. ICT is perceived as vital to enabling programmes to improve the quality of benefits and services to better meet the needs of individuals within heterogeneous population groups. However, India still faces a number of difficult challenges with regard to programme implementation, not least of which is the human resource challenge of a shortage of suitably qualified staff in a number of key positions.

This last issue — the human resource challenge — is more fully investigated by a contribution addressing the ageing of public-sector workforces in OECD Member States, a process that will see the imminent departure into retirement of significant proportions of staff of all grades in public-sector institutions, including social security administrations. Roddy McKinnon calls attention to this urgent problem: the public-sector workforce, including that of social security agencies, is ageing rapidly, and insufficient efforts are being undertaken to replace workers who will be retiring soon. A compounding factor is that large numbers of employees whose experience would ordinarily make them the prime candidates for moving into the senior positions that are soon to be vacated by long-standing employees are themselves roughly as old as their superiors; so they, too, will be retiring soon. In this connection, alternative policy responses such as the mentoring of lower-level employees may be useful in helping train staff to fill upcoming vacancies. McKinnon uses data from Ireland and Belgium to concretize the case that the public-sector workforce will lose not only top managers but public servants of all grades in the coming years, including the human resource management staff tasked to address this major staffing challenge. Building on practices from New Zealand, he suggests a broad approach to the hiring, retention and development of staff that adds flexibility to labour practices with the goal of ensuring continuing service quality. The overall outlook appears to be uncertain, and it may already be too late to implement some of the potentially most useful measures. A case is made for the International Social Security Association (ISSA) to use its networks and platforms for information dissemination to play a useful role in continuing to call attention to this problem and the need to act soon in response to it. Of particular pertinence are McKinnon's questions about political will and politically driven decision-making.

Political will is also important when addressing the many different challenges of meeting the social security needs of migrants. The manner in which the subject of migrants and social security is normally framed has shifted. The debate about migrants has moved beyond the narrow and decidedly self-interested national question of what migrants can do for the betterment of social security systems in societies that are ageing to one that is more balanced and which actively includes the policy objective of defining what social security systems can do to offer better social protection for the increasing number of international migrants. Previously, and most commonly in Europe, there was a suggestion that increases in the flow of international immigrants of working age would smooth the age distribution of ageing societies, increase labour supply and strengthen the finances of social security systems. However, “in a context of widespread institutional and social constraints on immigration, the magnitude of the increases needed to smooth the age distribution is inordinately large and not, as a practical matter, likely to be a decisive response to population ageing” (Bloom, Canning and Fink, 2010, p. 322). In large part, it is on this basis that policy arguments in more-aged societies are emphasizing the necessity of retaining a greater number of older workers in the workforce (where workers are physically capable of so doing), and doing so for longer.

In the global South in particular, many migrants are chronically poor and frequently have no rights to social security in either the “sending” or “receiving” country. In this regard Rachel Sabates-Wheeler and Johannes Koettl offer a disquieting picture of the plight of migrants in relation to their ability to benefit from social protection programmes in their host country. They describe the limited extent to which migrants are protected by bilateral or international agreements that allow them to benefit from some social security arrangements. Not only with pensions, but also in the case of other programmes meant to help the poor, migrants are often deemed ineligible. Particularly in the case of South-South migrants, not only do contributory social security programmes (e.g. social insurance schemes and national provident funds) often cover a small minority of workers, but earned rights are frequently not portable across borders or between different types of schemes. The consequent inability of migrants to benefit fully (or, often, at all) from the programmes that are in place in the country of origin and in the host country detracts significantly from their well-being and their security. Certain subsets of migrants, including women, are particularly disadvantaged. The authors point out that in many low-income countries, the absence of basic coverage for the majority of citizens and weak institutional capacity to operate social programmes is, in the first instance, a more fundamental problem than the lack of portability of rights to benefits. They conclude that South-South migration needs to be understood as being significantly different from North-North migration, where the social protection issues are much more tractable. But for all migrants, countries can act to ensure certain benefits and rights.

Social security is not only a matter of access to legally-defined benefits and rights, it is about shared values and goals and how these influence the choices societies make in allocating their finite resources to different population groups and cohorts. For all countries considering the nature of what is commonly referred to as the intergenerational bargain, the strong likelihood of revisiting the “share of the pie” accorded to each generation implies that a trade-off must be made, with relative winners and losers. The intergenerational bargain is characterized by flows that predominantly come from governments (e.g. for social security, education, and other public transfers), markets (which facilitate the accumulation of assets) and families (which provide economic support to children in all societies and to the elderly in many). It is against this background that Andrew Mason, Ronald Lee, and Sang-Hyop Lee explore how population ageing affects intergenerational flows (which are affected by demographic shifts, differences in labour income patterns and their determinants, spending on health, and spending on human capital) and the contrasting institutional approaches to such flows around the world. The direction of these flows used to be almost entirely toward children, but the worldwide trend toward greater longevity, the development of financial markets, and longer periods of retirement have led to a redirection of intergenerational flows toward the elderly. The authors build on data from National Transfer Accounts of an array of countries that differ from one another economically and demographically and that have chosen different policies regarding intergenerational transfers. They present a set of charts that show the different patterns of production and consumption over the lifecycle and the manner in which these patterns differ across countries. The methodology used involves a careful examination of old-age and total dependency ratios, and leads to a set of alternative projections of these ratios based on different lifecycle scenarios. The authors suggest that new labour and economic policies that take into account demographic realities will be useful in countering concerns about intergenerational equity and a possible slowing of economic growth.

Drawing together many of the important threads of discussion running through the articles that comprise the major part of this special issue, the contribution from Sarah Harper provides a forward-looking concluding assessment of the capacity of social security and health care institutions, and of individuals, to adapt to the non-demographic as well as demographic challenges of an ageing world. Her policy conclusions make the case for a holistic public policy approach to the social, economic, labour market and health care challenges that are expected to accompany population ageing. Henceforth, just as public social security and health care institutions should be more appropriately integrated with mechanisms of private and family transfers, so too there will there be a heightened need for all individuals to be aided by public policy interventions, as and when these are deemed appropriate, that support access to employment opportunities, promote active ageing and life-long learning and which encourage preventive approaches to life-style choices and the use of health care. Harper argues that the building blocks for delivering financial security and health and care policies to address population ageing are already in place in many countries. To an important degree, these policies are framed by what Harper identifies as national collective goals. The implication, however, is that the manner in which these building blocks are stacked in the future, ostensibly to address the many, broad challenges of population ageing, may continue to be framed by collective goals defined by a previous era. In spite of this, the adaptation that Harper argues as essential includes the likely shouldering of a greater share of income risk by individuals, which necessarily will be accompanied by the development of alternative and/or complementary institutional mechanisms. However, such change is not expected to herald an overall reduction in the role of the State in ensuring the healthy functioning of national social protection systems.

That the role of the State is expected to remain integral to social security in the future is a key policy message to emerge from this special issue. In the immediate term, however, in order to better respond to the current and emerging substantive policy challenges that confront social security, more and better data and analysis, as well as more institutional collaboration, is sorely required. For this reason, a second broad objective of this special issue is to support the objective of improved data collection and analysis and the fostering of institutional collaboration. Many of the contributions draw attention to the importance of enhancing capacities to develop comparative cross-national data, not least to support increasingly systematic and coordinated efforts to first identify and then inform policy-actors about common challenges and to distil out pertinent policy lessons. Standing out in this regard is the final authored contribution to this special issue by Jinkook Lee, which focuses specifically on the development of data sets on pensions and health care provision.

Lee's detailed description of surveys of the elderly around the world provides a very efficient path for grasping their scope and the differences among them, and is illustrative of the use to which coordinated collection of cross-country data is being put. The importance of this set of efforts stems from the fact that carefully harmonized data — whose collection depends on extensive communication among the different teams carrying out surveys — makes possible in-depth, cross-country, comparative research about the social, economic, and health status of the elderly, and the determinants of that status. The inclusion of extensive, well-harmonized questions on health in these surveys underlines both the importance and the feasibility of using cross-country data to approach the public policy challenges of ageing societies from a holistic perspective. The data analysis emerging from such efforts will be available to inform the design of programmes that seek to ensure the well-being of the elderly in a highly varied array of situations. Lee makes clear that the effort to achieve cross-country consistency in the data will be very useful to decision-makers.

As a further contribution towards improving data sharing, this special issue also presents a small statistical appendix of global, regional and national demographic and social security indicators to accompany the papers. The purpose of the appendix is twofold. First, the data are meant to complement and support the authored articles. Second, it provides a stand-alone reference source of key data, as well as of principal data sources, that will permit readers of the Review to further develop their own study of the numerous cross-cutting policy issues presented. Taken together, the development, improvement, and sharing of data is expected to contribute positively to improving the transparency of all social security programmes among all stakeholders. This, in turn, should contribute further to building trust in institutions, including public and private social security bodies.

Looking forward to the coming decades, all efforts to create better performing and increasingly integrated national systems of social protection imply a requirement for stronger governance and continuing heavy State involvement. This is likely to be especially so as regards decisions about the allocation of public finance, in particular to those deemed most in need. Although the full extent of all aspects of the demographic challenge for social security remain to be fully defined and understood (a task to which this set of papers seek to contribute), it is clear that the sustainable delivery to individuals, households and families of adequate social protection will require political will, evidence-based policy-making and, no less important, appropriately designed, equipped and professionally-staffed social security administrations.


  1. 1. Support for David Bloom's work on this special issue was provided by the Program on the Global Demography of Aging at Harvard University, funded by Award Number P30AG024409 from the National Institute on Aging. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging or the National Institutes of Health. The editors acknowledge the valuable comments and assistance of Elizabeth Cafiero and Larry Rosenberg in the preparation of this special issue, and Frédérique Bocquet, Florence Bonnet, Barbara Kritzer and Jenna-Dawn Shervill for support in preparing the statistical appendix.

  2. 2. The United Nations Population Division collects and synthesizes the data shown in these graphs. The figures through 2005 are based on actual data. The UN bases its population projections on a model that takes into account fertility, mortality and migration. Fertility projections are derived from the current fertility rate, recent changes in that rate in a given country, the historical pace of fertility decline in a wide range of countries, and a rate to which all countries are expected to eventually converge. All projections starting in 2010 used in this article are based on the UN's medium-variant fertility scenario. Mortality projections are derived from a model that assumes that life expectancy will increase less rapidly in countries where it is already high. Recent country- and sex-specific trends are taken into account, as are the effects of HIV/AIDS. Migration trends are assumed to follow past patterns, as potentially modified by current immigration policies. The UN's methodology is described in detail in “Assumptions Underlying the Results of the 2008 Revision of World Population Prospects”, <>.

  3. 3. The distinction here follows the practice of the United Nations Population Division.

  4. 4. This section borrows heavily from Bloom, Canning and Fink (2010).

  5. 5. Amongst other issues, the perspectives of the two international organizations may be differentiated on the grounds of the relative importance accorded to collective risk-pooling vis-à-vis individual responsibility; the relative importance of ensuring, at least, minimum “standards” of retirement income for all throughout the duration of retirement; the respective and relative roles of public agencies and private institutions responsible for pension administration and fund management; and the relative mix of general tax revenue, pay-as-you-go, and fully-prefunded financing mechanisms.

  6. 6. See Bloom et al. in this issue for an outline of the evolution of the World Bank's multi-pillared pension system model.

  7. 7. PAYG system reform options include raising retirement ages; equalizing the retirement ages of men and women; increasing the number of years of contributions required to qualify for a full pension; removing incentives to earlier retirement; linking individual benefits more closely to the value of individual contributions paid; and reducing the expected rate that future benefits will increase, by switching to indexation based on prices rather than wage growth, or by using composite indexes using measures of both.

  8. 8. For many low- and middle-income countries in particular, the preoccupying concern is not one of how to increase the employment levels of older workers; rather, it is how to reduce high levels of youth unemployment and underemployment.

  9. 9. For an assessment of the experience of Chile, see Berstein, Larraín and Pino (2005). For an assessment of the experience of Kazakhstan, see Becker et al. (2009).