Securing the Livelihoods of Informal Economy Workers in Times of Global Changes

Report

5 October 2026

3. Addressing the challenge of ageing with informality

Copy link to 3. Addressing the challenge of ageing with informality

Abstract

As the global population ages rapidly, the high incidence of informality in developing countries presents unique challenges for the livelihoods of the elderly. Informality leads to significant income risks during old age reflecting limited accumulation of retirement rights during working years and the subsequent lack of adequate retirement income. Using data from the Key Indicators of Informality based on Individuals and their Households (KIIbIH), which covers over 50 developing countries, this chapter begins by assessing the livelihoods of elderly individuals and their coping strategies in the face of substantial pension gaps. It then examines the current enrolment of workers in contributory pension schemes and their capacity to contribute, highlighting large coverage gaps among informal economy workers, even among relatively high-paid workers. Finally, the chapter reviews the strengths and weaknesses of various pension models and discusses the financing of pension systems in contexts of widespread informality. To conclude, the chapter outlines several policy options that governments should consider to secure the livelihoods of elderly individuals in such contexts.

In brief

Copy link to In brief

Rapid ageing and persistent informality create unique policy challenges to secure the livelihoods of the elderly of today and in years to come

Rapid ageing is one of the most pressing socio-economic megatrends. Globally, the share of the population aged above 65 years is expected to rise from 10% in 2022 to 16% in 2050, reaching 24% in 2100. In developing countries, ageing is taking place in contexts in which informality remains the dominant form of employment. This adds complexity to the quest to develop robust pension systems that can achieve decent coverage and offer adequate old-age benefits.

Income insecurity is widespread for the current elderly population

Across KIIbIH countries, 51% of elderly receive pension benefits – either contributory or non-contributory. However, the coverage varies drastically across countries, depending on their level of development, region and status in the demographic transition.
On average, urban men who belong to the richest households are more likely to receive contributory pension benefits, whereas women, people living in rural areas and the poor are more likely to be covered by non-contributory schemes.
In the absence of adequate retirement income, the elderly have adopted diverse coping strategies to secure their livelihoods, including delaying retirement for workers of the informal economy and relying on inter-generational support.

Few working-age workers participate in contributory schemes, putting their future old-age income at risk

Across KIIbIH countries, only 40% of employed individuals, and 8% of informal workers, contribute to some type of pension scheme.
Low enrolment affects not only low-paid informal economy workers with limited contributory capacities, but also medium- and high-paid informal workers and formal self-employed workers who could afford to contribute. This reflects a variety of barriers, ranging from low earnings and irregular income to lack of information, mistrust in the system and/or lack of relevant options, especially for non-employees.

Policies must focus on the triple challenge of coverage, adequacy and fiscal/financial sustainability

Policies should aim to coherently articulate non-contributory and contributory schemes, ensuring consistency and synergy to balance retirement income security with incentives for formalisation.
Non-contributory schemes are crucial to cover the elderly’s basic needs and shield them from poverty. They need to be complemented by contributory schemes that fit the needs of informal economy workers, with a view to gradually integrate both formal and informal workers in common schemes to avoid fragmentation and achieve meaningful scale.
Financing the expansion of pension systems is challenging but opportunities exist. Expanding non-contributory pensions – achievable at relatively low fiscal cost for basic anti-poverty benefits – offers significant social returns and large positive spillovers, but requires fiscal reforms to better mobilise domestic resources. In parallel, some scope exists to broaden the base of social security contributors and extend contributory pension coverage by reaching informal economy workers with the largest contributory capacity.

Introduction

Copy link to Introduction

Population ageing has become one of the most salient global trends of the 21st century. Between 2022 and 2050, the share of elderly people in the world – those aged more than 65 years – is expected to sharply increase from 10% to 16% (United Nations, 2023[1]). This trend is particularly evident and rapid in emerging markets and developing economies (EMDEs). As a result, many low-income and lower-middle income countries are expected to become “old” before they become “rich”.

Old age is fundamentally characterised by physical and cognitive decline, which progressively hampers individuals’ ability to work and earn an income. For more than 150 years, in a historical process that started in more advanced countries but progressively expanded to the entire world, governments have established pension systems as a means to better protect the elderly from the risk of income loss associated with old age. Although pension systems are complex and diverse policy instruments, most share the fundamental common characteristic of being structured as an insurance product whereby workers are expected to provide regular contributions in exchange for a future income during old age. Historically, the right to a pension has materialised through formal working arrangements and has been closely tied to the employer-employee relationship via co-payments of contributions.

To date, the design, establishment and development of pension systems have largely ignored the needs of informal economy workers, who dominate the labour force in developing countries (see Chapter 1). By and large, workers in the informal economy do not have secure access to contributory pension schemes. Rather, they have to rely on social pensions when these exist, or on ad hoc mechanisms such as private saving, intra-family transfers or delayed retirement.

In this context, navigating the challenges of ageing and informality is critical for governments in EMDEs. Informality undermines not only the current livelihoods of the elderly by limiting their pension entitlements, it also threatens the future well-being of today’s informal economy workers who are not contributing to any pension scheme and risk inadequate income once retired. Addressing this dual challenge will require policymakers to pursue a three-pronged approach of: i) improving income security for the current elderly who do not or inadequately benefit from contributory and/or non-contributory schemes; ii) enrolling informal economy workers in contributory pension schemes during their working life to protect their income during old age; and iii) ensuring the sustainable financing of pension systems.

Drawing upon KIIbIH data, this chapter investigates the multiple impacts ageing and informality have on the livelihoods of current and future elderly. The chapter starts by painting a global picture of the ageing transition, placing it in the context of the world’s large informality footprint. It then examines the livelihoods of current elderly people, analysing the main social protection gaps during old age and identifying the mechanisms the elderly adopt to cope with large deficits in pension systems. In turn, the chapter explores the livelihoods of the future elderly population, highlighting the links between informal employment and low participation in contributory pension schemes. Critically, the chapter reviews the strengths and weaknesses of diverse pension models in the context of high informality and provides some cost estimates of extending social pensions. Finally, the chapter identifies short- and long-term policy options to develop robust pension systems that can successfully blend non-contributory and contributory schemes, include informal economy workers, reconcile incentives for formal work and decent retirement income, and account for the diverse socio-economic, demographic and fiscal contexts of the countries covered by the KIIbIH.

Developing countries face the double challenge of ageing and informality

Copy link to Developing countries face the double challenge of ageing and informality

The entire world is ageing and developing countries are ageing faster

The world is ageing quickly, especially in EMDEs. The share of the global population aged above 65 years is expected to rise from 10% in 2022 to 16% in 2050 (United Nations, 2023[1]). Although the ageing trend started in developed countries, it is now affecting developing countries. By 2050, estimates predict that the majority of the population aged above 65 years will live in developing countries. As of 2025, most countries in Asia and in Latin America and the Caribbean are considered ageing societies,1 defined as having between 7% and 14% of the population aged above 65 years. In the next 30 to 40 years, most countries in these regions will be considered aged or superaged societies, meaning that the share of the population aged above 65 years will range from 15% to 20%, or will be superior to 21%. In Africa, the demographic transition is projected to take place later. This reflects a young population at present, with a median age of 19.3 years in 2025, and a total fertility rate of 3.95 live births per woman, which remains substantially above the replacement fertility rate (around 2.1 children per woman) and will fall below it only by the end of the 21st century (United Nations, 2024[2]). Despite this demographic resilience, nearly all African countries are expected to become ageing societies by 2075 (Figure 3.1).

Figure 3.1. The share of elderly is rapidly increasing across the world
Copy link to Figure 3.1. The share of elderly is rapidly increasing across the world

Share of individuals aged above 65 years as a proportion of total population, 2000-2100

A key feature of this societal ageing trend is that developing countries are ageing much more rapidly than developed countries. In most developed countries, it took between 40 and 120 years for the proportion of the population aged above 65 years to rise from 7% to 14% – that is, to transition from an ageing society to an aged one. In turn, it took (or will take) between 20 and 50 years for the share to increase from 14% to 21%, that is to transition further to a superaged society. By contrast, most developing countries will experience the initial doubling of the share of older persons (from 7% to 14%) in only 15 to 35 years and will experience the increase to 21% in just 10 to 30 years (United Nations, 2023[1]).

The KIIbIH dataset includes a heterogeneous set of countries at different stages of the demographic transition and ageing process. Using a demographic typology adapted from that of Ahmed et al. (2016[3]), KIIbIH countries can be classified in four broad categories depending on where they stand relative to the demographic dividend (i.e. the automatic increase in the share of working-age individuals relative to the share of dependents induced by the decline of fertility rates): pre-dividend, early-dividend, late-dividend and post-dividend countries (Figure 3.2 and Annex 3.A for methodological details on the demographic typology). As regards old-age income security, this demographic typology is crucial to identify key policy challenges and constraints. It offers insights on whether countries are still able to rely on a young and growing workforce that could ease the establishment or consolidation of contributory schemes. It also provides insights on the respective share of the elderly population to be covered and the associated magnitude of the fiscal cost of non-contributory pensions.

Figure 3.2. Countries covered by the KIIbIH dataset are at different stages of the demographic transition and ageing process
Copy link to Figure 3.2. Countries covered by the KIIbIH dataset are at different stages of the demographic transition and ageing process

Demographic typology of countries

Current evidence shows that between 2025 and 2100, ageing will be particularly severe in late- and post-dividend countries. In these countries by 2050, on average, the share of the population aged above 65 years will reach 19% and 30% respectively, increasing to 30% and 44% by 2100 (Figure 3.3). In post-dividend countries, the share of individuals aged above 80 years will also drastically increase, attaining, on average, 18% by 2075 and 26% by 2100. Over the period to 2100, the share of the working-age population will shrink to 59% of the total population across all KIIbIH countries, and to only 47% in post-dividend countries.

Figure 3.3. Ageing will be particularly severe in late- and post-dividend countries
Copy link to Figure 3.3. Ageing will be particularly severe in late- and post-dividend countries

Distribution of age groups over time (2025-2100) and across the demographic dividend-based typology

Rapid ageing is taking place in a context of high and persistent informality

In developing countries, rapid ageing of the population is taking place while informal employment remains widespread – or even dominates (see Chapter 1). In fact, informality is more prevalent in countries that are still at the beginning of the demographic transition. In pre-dividend countries, the unweighted average informality rate is 94% compared to 77% across early-dividend countries, and 47% among late-dividend countries (Figure 3.4).

The high incidence of informality adds complexity to the ability of countries at different stages of their demographic transition to develop robust and sustainable pension systems that can achieve decent coverage and offer adequate old-age benefits for the elderly of today and the future. Where a large majority of workers operate in the informal economy and have little access to social insurance, retirement comes with no replacement income from contributory pension schemes. In such situations, the elderly have to rely primarily on private transfers, private savings and/or non-contributory pensions – when available. In EMDEs, contributory pension systems typically lack the three things that enable the wide coverage observed in advanced and OECD economies: a large formal labour base, delivery capacity and strong institutions. As such countries struggle to develop multi-pillar and integrated pension systems, the informality footprint poses unique challenges that must inform policy decisions and guide the design of pension systems that address the needs and expectations of informal workers.

Figure 3.4. Rapid ageing is taking place in a context of high and persistent informality
Copy link to Figure 3.4. Rapid ageing is taking place in a context of high and persistent informality

Share of workers aged above 15 years in informal employment, latest year available

In developing countries, the large informality footprint jeopardises the livelihoods of many elderly

Copy link to In developing countries, the large informality footprint jeopardises the livelihoods of many elderly

A large gap in effective pension coverage exists among the elderly

At the global level, pensions are the most prevalent form of social protection (ILO, 2024[5]). The existence of many different designs across the world reflects the fact that pension schemes are among the most diverse and complex social protection mechanisms. Depending on the policy objectives of each scheme, a wide range of parameters can be customised, adapted and combined in many distinct ways, as discussed later in this chapter.

Progress in pension coverage has been uneven across countries, demographic groups and world regions

Over the last decades, EMDEs have undertaken substantial efforts to expand pension coverage with mixed results (see Box 3.1 on the definition and scope of pension coverage). Across KIIbIH countries for which data are available, only 52% of the population aged above the respective legal age of retirement receive pension benefits – either contributory or non-contributory (Figure 3.5). In reality, the situation varies drastically across countries.

Figure 3.5. Progress towards expanding pension coverage has been more limited in countries at the early stages of their demographic transition
Copy link to Figure 3.5. Progress towards expanding pension coverage has been more limited in countries at the early stages of their demographic transition

Share of individuals aged above the legal age of retirement who receive contributory and non-contributory pension benefits by demographic typology of countries, latest year available

In countries in the early stages of their demographic transition, expansion of coverage of contributory pensions has not been fast enough to guarantee adequate income security in old age. KIIbIH estimates show average rates of pension coverage at 89% in post-dividend countries and 67% in late-dividend countries, compared with 45% among early-dividend countries and only 6% among pre-dividend countries (Figure 3.5).

Geographically, progress in expanding pension coverage has been concentrated in the Latin America and Caribbean region, as well as in a small number of countries from other regions of the world. In two-thirds of the Latin American and Caribbean countries included in the analysis (12 countries out of 19), more than half of the elderly population receives pension benefits; in seven of these countries, the pension coverage rate is higher than 80%.2 Likewise, certain countries in Asia (e.g. Armenia, Mongolia and Thailand) and Africa (e.g. Namibia and South Africa) have achieved high rates of pension coverage.

In many of these countries, development of large social pension schemes underpins the high rates of pension coverage. This is notably the case in early-dividend countries that have achieved high rates of coverage, such as Bolivia, the Maldives, Namibia and South Africa. In these countries, more than 75% of the population above the legal age of retirement is covered by non-contributory pension benefits, with only a tiny fraction receiving both contributory and non-contributory benefits.

Box 3.1. Definition and scope of pension coverage according to the KIIbIH

Copy link to Box 3.1. Definition and scope of pension coverage according to the KIIbIH

Data from the KIIbIH rely on the harmonisation of data collected through household surveys by national statistical offices. Data and rates may differ from other official sources such as the International Labour Organization (ILO), which usually rely on administrative records (e.g. the Social Security Inquiry questionnaire to monitor Target 1.3 of the Sustainable Development Goals [SDGs], i.e. the proportion of persons effectively covered by a social protection system, including social protection floors). Data may also differ from estimates based on labour force surveys as the KIIbIH uses household consumption or income surveys that contain a labour module that captures enough information to compute adequate informality estimates.

Indicators on access to social protection in general, and to pension coverage in particular, refer to effective coverage and not legal coverage. In line with the ILO’s approach, effective coverage of pension schemes can then be measured by two complementary indicators: i) people contributing to a pension scheme (contributors); and ii) people actually receiving benefits (beneficiaries). In the second section of this chapter, coverage refers to the share of elderly who receive benefits from pension schemes. In the third section, coverage (also noted as enrolment) refers to the share of individuals who contribute to a pension scheme.

Another potential source of discrepancy compared to alternate sources of data is the population base. Unless specified otherwise, the KIIbIH reports statistics for contributors’ coverage for individuals aged between 15 years and the statutory age of retirement; statistics for beneficiaries’ coverage apply to individuals aged above the statutory age of retirement.

Elderly from wealthier households enjoy a higher coverage of contributory schemes while elderly in poorer households tend to rely more on non-contributory schemes

Elderly from wealthier households tend to be better covered by pensions systems than elderly who are poorer. Across all demographic groups – except in post-dividend countries – as well as all regions, the share of individuals aged above the legal age of retirement who receive a pension is larger among wealthier quintiles and deciles of welfare than among the poorest quintiles and deciles (Figure 3.6). Overall, across all KIIbIH countries, the average coverage difference between elderly belonging to the first quintile and the fifth quintile is 12 percentage points. In late-dividend countries, this difference rises to 19 percentage points. Regionally, the largest difference is found in Latin America and the Caribbean, where the coverage gap between the first and the fifth quintile reaches 17 percentage points.

To a large extent, the higher pension coverage rate observed among elderly people who belong to the richest quintiles reflects participation in contributory schemes. According to KIIbIH data, in early-dividend countries, 28% of the elderly from the fifth quintile receive a contributory pension, compared to only 8% of the elderly belonging to the first quintile. In late- and post-dividend countries, the coverage differences for contributory pensions between old individuals belonging to the first and fifth quintiles is, respectively, 29 and 17 percentage points. Such large inter-quintiles discrepancies3 tend to reflect past labour history: current recipients of contributory pensions are more likely to have worked in formal and better paid jobs during their working life.

In contrast, poorer elderly are more likely to be covered by non-contributory schemes. Across demographic groups and regions, a larger proportion of elderly people belonging to the poorest quintiles receive non-contributory pensions than those belonging to the richest quintiles. This is notably the case in late-dividend and post-dividend countries as well as in Latin America and the Caribbean, where non-contributory schemes appear particularly progressive and redistributive. This likely underlines the successful development of non-contributory schemes targeting the poorest segments of the elderly population, notably those who are not eligible to contributory pensions.

Figure 3.6. Pension coverage tends to be larger for individuals in richer households
Copy link to Figure 3.6. Pension coverage tends to be larger for individuals in richer households

Share of individuals aged above the legal age of retirement who receive contributory and non-contributory pension benefits by quintiles of household welfare

Old-age poverty is disproportionately high in countries with limited pension schemes

In terms poverty, the elderly are not necessarily more affected than working-age individuals. Using both the poverty line of USD 3.00 in 2021 PPP (otherwise often referred to as the “extreme poverty line”) and the poverty line of USD 4.20 in 2021 PPP, the difference in average poverty rates of the elderly compared with working-age individuals is less than one percentage point (Figure 3.7).

However, large differences exist across countries, demographic groups and world regions. In many African countries, the elderly have much higher poverty rates than working-age individuals. Conversely, in most Latin American and Caribbean countries, the elderly are less likely to be poor than working-age individuals, pointing towards the success of dramatic expansion of non-contributory pensions since the early 2000s (Reyes Hartley and Abels, 2025[6]). Poverty is also disproportionately high among the elderly in pre- and early-dividend countries, where pensions systems tend to be less developed. In pre- and early-dividend countries such as Gambia, Ghana, Kenya, Namibia and Zambia, the poverty rate (at USD 4.20) for the elderly exceeds that for working-age individuals by more than 5 percentage points. From a policy point of view, introducing anti-poverty measures targeted at the elderly, such as social pensions, could have large welfare effects. This is notably the case in Indonesia where poverty rates of the elderly have been markedly higher than the rest of the population for decades and the existing non-contributory targeted pension scheme has very low coverage. Introducing a modest social pension could have substantial welfare consequences (Kudrna, Piggott and Poonpolku, 2024[7]; OECD, 2024[8]; Priebe, 2017[9]).

Figure 3.7. Old-age poverty is disproportionately high in countries with limited pension schemes
Copy link to Figure 3.7. Old-age poverty is disproportionately high in countries with limited pension schemes

Poverty gap between the elderly and working-age individuals at USD 3.00 and USD 4.20 per day (2021 PPP), latest year available

Substantial gender gaps in overall pension coverage prevail in most KIIbIH countries, largely reflecting gender gaps in contributory schemes

In most KIIbIH countries, pension coverage of elderly women is lower than for elderly men. On average, the gender gap in pension coverage stands at only four percentage points, with 50% of women aged above the legal age of retirement receiving a pension, compared with 54% of men (Figure 3.8). However, in 31 of 43 countries (or 72%) for which sex-disaggregated data on pension coverage are available, the proportion of men receiving pension benefits is higher than the share of women. In nine of these countries, the coverage difference between elderly men and elderly women is superior to 10 percentage points.

Gender gaps in overall pension coverage primarily reflect large differences in favour of men in terms of contributory pensions. In nearly all KIIbIH countries, the share of men aged above the legal age of retirement who benefit from a contributory pension is substantially higher than for women. On average, in pre-, early- and post-dividend countries, the gap is five percentage points in favour of men. It reaches nine percentage points in late-dividend countries. This is largely in line with findings at the OECD level, which show that women are underrepresented among earnings-related pension recipients (OECD, 2025[10]).

In contrast, women tend to be slightly better covered than men by non-contributory pension schemes. On average, across KIIbIH countries, the average share of women aged above the legal age of retirement receiving a non-contributory pension is 25%, compared with 22% for elderly men. In certain countries, this gender gap in favour of women is substantial. For instance, in Uruguay, 37% of elderly women receive non-contributory pension benefits, compared with 5% of men. In the Bahamas, Barbados, Cyprus and South Africa, the gap in favour of women exceeds 10 percentage points.

Gender differences in contributory and non-contributory pension coverage largely reflect gender dynamics in the labour market and prevailing social norms regarding the role of women. Globally, women remain significantly less likely than men to participate in the labour force because of deeply entrenched discriminatory social norms that create structural barriers, discourage women from entering the labour market and confine them to care and reproductive roles. In 2022, men’s participation rate was 25 percentage points higher than women’s (OECD, 2023[11]). Since most contributory pension schemes are tied to the employment status of individuals, these labour market disparities directly translate into gender gaps in contributory pension coverage. Conversely, more limited employment history often means that women are overrepresented among recipients of non-contributory pensions, notably in countries where eligibility is determined by means or pension testing (e.g. the Bahamas, Barbados, Costa Rica, South Africa, Thailand and Uruguay). Because contributory benefits tend to be much lower than non-contributory benefits, these gender asymmetries in coverage raise important questions regarding the adequacy of revenue received by elderly women compared with men.

Figure 3.8. Gender gaps in pension coverage largely reflect gender gaps in contributory schemes
Copy link to Figure 3.8. Gender gaps in pension coverage largely reflect gender gaps in contributory schemes

Share of individuals aged above the legal age of retirement who receive contributory and non-contributory pension benefits by gender (Panel A) and gender gaps in the coverage of pension schemes among individuals aged above the legal age of retirement (Panel B)

Elderly people in urban areas are better covered by contributory schemes while larger proportions of elderly people in rural areas receive non-contributory pensions

Overall pension coverage among the elderly is higher in urban areas than in rural ones. On average, across KIIbIH countries, pension coverage in urban areas is six percentage points higher than in rural ones, with 52% of urban elderly receiving a pension, compared with 46% of rural individuals (Figure 3.9). This gap in favour of urban individuals is found in nearly all countries for which data are available and is particularly marked in pre-dividend countries where it reaches, on average, 11 percentage points.

Figure 3.9. Overall pension coverage is higher in urban areas, stemming from large rural-urban gaps in contributory pension coverage
Copy link to Figure 3.9. Overall pension coverage is higher in urban areas, stemming from large rural-urban gaps in contributory pension coverage

Share of individuals aged above the legal age of retirement who receive contributory and non-contributory pension benefits by location (Panel A) and rural-urban gaps in the coverage of pension schemes among individuals aged above the legal age of retirement (Panel B)

This overall gap in favour of urban areas largely reflects a higher coverage of contributory pension schemes in urban areas. In nearly all KIIbIH countries, the share of urban individuals aged above the legal age of retirement who benefit from a contributory pension is substantially higher than for the rural elderly (Figure 3.9). Non-contributory pension schemes, in contrast, tend to benefit more rural than urban elderly, notably because many non-contributory schemes are progressive by design and target poorer individuals who tend to be disproportionately concentrated in rural areas.

Elderly people rely on a variety of coping strategies to protect their livelihoods

Most informal workers have to continue working past the official retirement age

For many workers in the informal economy, reaching the legal retirement age makes no or little difference. In the absence of a decent old-age pension, many informal workers above the official retirement age have to continue working to sustain their livelihoods. Notably, weaker occupational safety and health protections, and limited access to health coverage over the working life may result in many such workers entering old age in poorer physical condition. KIIbIH data show a natural decline in the overall employment ratio of the elderly (Figure 3.10, Panel A). However, this decline conceals two different dynamics for formal and informal workers. For formal workers (i.e. those most likely to have accumulated enough pension rights), employment ratios sharply drop after retirement age. The average employment-to-population ratio of formal workers goes from 23% for those aged between 10 and 5 years younger than the legal age of retirement, to 4% for those aged between 5 and 10 years older than the legal age of retirement, and to less than 3% beyond 10 years (Figure 3.10, Panel C). Conversely, although employment ratios also decline for informal workers, informal employment remains substantial in many countries until an advanced age. Across KIIbIH countries, the average employment-to-population ratio of informal workers remains at 41% for those aged between 5 and 10 years older than the legal age of retirement, and at 27% for those 15 to 20 years older than the legal age of retirement (Figure 3.10, Panel B).

The contrast in informal versus formal employment dynamics after retirement age is especially stark in late- and post-dividend countries. In these regions, employment rates show a sharp decline: on average, the gap between the employment rate of the working-age population and that of individuals aged between the official retirement age and 74 is larger than 20 percentage points. This decline largely stems from formal workers exiting the labour force. In contrast, pre- and early-dividend countries – where formal employment is less prevalent – show a more gradual drop in employment rates after retirement age. This is mainly due to the slower decrease in informal employment among older individuals.

Figure 3.10. Employment ratios remain high for informal workers past the legal retirement age
Copy link to Figure 3.10. Employment ratios remain high for informal workers past the legal retirement age

Total, formal and informal employment-to-population ratios by age brackets, from 10 years prior to the legal age of retirement to 20 years past the legal age of retirement, latest year available

After retirement age, vulnerable forms of employment increase

Past the age of retirement, the nature of employment changes and becomes more vulnerable. One characteristic is the higher prevalence of informal employment among old-age workers. The underlying mechanism is a combination of three factors: i) formal workers exiting the labour force upon reaching the age of retirement (thereby automatically reducing the relative share of formal workers and increasing the share of informal workers); ii) formal workers joining the informal economy to continue working when pension benefits are too low; and iii) informal workers continuing to work until an advanced age. In Indonesia, for instance, research indicates that workers rarely move between the formal and informal economies during their careers (Kudrna, Piggott and Poonpolku, 2024[7]). However, after reaching the age at which they can access their pension, a notable share of Indonesia’s formal economy workers shift into the informal economy. This trend is particularly evident in late- and post‑dividend countries. On average, in the former, informality rates for working-age workers and for those aged between the official age of retirement and 74 years increase from 45% to 68%. In the latter, it increases from 39% to 54% (Figure 3.11).

Figure 3.11. The relative importance of informal employment increases after the legal age of retirement
Copy link to Figure 3.11. The relative importance of informal employment increases after the legal age of retirement

Rate of informal employment among working-age workers and elderly workers, latest year available

A rise in own-account work and the decline in wage employment are other key characteristics of the change in the nature of employment after retirement age. The shift towards own-account work is observed for both formal and informal workers but is even more pronounced for the latter. Across KIIbIH countries, the average share of own-account workers before and after retirement age increases from 44% to 62% for informal workers, and from 10% to 22% for formal workers. Conversely, the average share of employees before and after retirement age drops from 32% to 16% for informal workers, and from 84% to 69% for formal workers (Figure 3.12).

Figure 3.12. Past the age of retirement, elderly workers are overrepresented in own-account work
Copy link to Figure 3.12. Past the age of retirement, elderly workers are overrepresented in own-account work

Employment status of informal (Panel A) and formal (Panel B) workers, prior and past the legal age of retirement

Statistically, this clustering of elderly workers into self-employment could stem from intertwined dynamics: an exit from the labour force combined with a shift towards self-employment. On one side, employees claiming pension benefits and exiting the labour force can increase the relative proportion of own-account workers among the elderly. In parallel, the elderly previously working as employees may shift towards own-account work, particularly when pension benefits are too low and they need to continue working to sustain their livelihoods. In some countries (e.g. the People’s Republic of China, hereafter “China”), these dynamics may also be shaped by policies: the pension system does not allow individuals to continue working as a formal employee once in receipt of a formal sector pension. Individuals who wish to keep working (whether by necessity or choice) are forced to switch to a different form of work. From a policy point of view, this situation underlines the important question of whether retirement age in contributory systems is mandatory (or not) and whether the rules may differ between civil service/public and private sector workers who operate in the formal sector.

Beyond the transition towards more vulnerable forms of employment, elderly workers also face additional vulnerabilities. These notably include decent work deficits such as occupational and health issues, lack of representativity, risk of harassment and absence of employment-based injury protection.

In the absence of adequate pension systems, inter-generational support remains the norm

In countries where pensions schemes have a limited reach, elderly people are more likely to continue living in multi-generational households. In the absence of adequate old-age income, living arrangements whereby the elderly cohabitate with members of younger generations who work and earn an income are crucial to guarantee the former’s access to basic necessities such as food and shelter. KIIbIH data show that in countries with a larger share of individuals aged above 65 years receiving a pension, a greater proportion of elderly people live in households with no workers and/or only old individuals or children (Figure 3.13, Panel A). This is notably the case in late- and post-dividend countries in which pension systems have larger coverage and the demographic transition is profoundly affecting social structures. For instance, in late-dividend countries, 30% of individuals aged above 65 years live in households composed only of other non-working elderly people or children. In post-dividend countries, this share jumps to 49%. Conversely, in countries where coverage of the pension system among elderly is low, most of them continue living together with other working individuals (Figure 3.13, Panel B). On average, across pre-dividend countries, only 4% of individuals aged above 65 years live in households composed only of other non-working elderly people or children. Earlier studies have also found that the level of income directly received by the elderly influences elderly co-residence rates – which increase significantly as pension income decreases (Evans and Palacios, 2015[12])

Figure 3.13. The lower the coverage of pension, the higher the share of elderly living with other working individuals
Copy link to Figure 3.13. The lower the coverage of pension, the higher the share of elderly living with other working individuals

Correlation between the pension coverage of individuals aged above 65 years and the share of elderly living alone or together with other non-working elderly or children (Panel A) and the share of elderly living together with workers (Panel B)

When elderly people live in households together with other workers, they tend to disproportionately live in households that are fully or partially informal. Across KIIbIH countries, on average, 72% of individuals aged above 65 years live together with other individuals who work; and 51% of the elderly live in households with other individuals who all work in informal employment. Conversely, only 12% of elderly people live in fully formal households (Figure 3.14). This is notably the case in pre-dividend and most early-dividend countries. In late- and post-dividend countries, where pension systems are more developed and the informality footprint is less pronounced, the distribution of elderly people is more spread across the different types of households, including fully informal households, fully formal ones and households in which only non-working elderly people live.

Figure 3.14. Elderly people disproportionately live in households exposed to informality
Copy link to Figure 3.14. Elderly people disproportionately live in households exposed to informality

Distribution of individuals aged above 65 years by types of households in which they live, latest year available

The propensity of elderly people to live in multi-generational households, many of which are primarily composed of informal workers, offers crucial insights on the potential role of social pension. Through different channels, old-age benefits can have positive effects that span beyond the well-being of the elderly alone and can accrue to other generations (O’Keefe and Rongen, 2025[13]). In settings where they live in multi-generational households, elderly people receiving social transfers (e.g. non-contributory pensions or cash transfers) can use such funds to: i) strengthen investment in human capital and education of other members of the household; ii) channel resources to the household’s productive investment; and iii) improve the ability to face shocks of other individuals living in the same household.

Beyond the potential intra-household spillovers of social assistance, elderly co-residence with other generations is a complex but crucial dimension of household livelihoods that can have both positive and negative implications. In the absence of other means, the elderly may provide informal childcare services as well as opportunities for other household members, notably women, to participate in income-generating activities. Conversely, in contexts where income-support for the elderly and old-age care options are limited, this clustering can add strain on the household resource, increase medical costs and attention, and generate more unpaid care needs, likely provided by women, which would hamper their participation in the labour market. For the younger generations of the household, such co-habitation may also limit their ability to save, either for productive investment and/or for retirement, perpetuating the retirement challenge across generations.

Beyond income security, the livelihood of elderly people is further threatened by a lack of access to health insurance and long-term care

Most elderly are not covered by health insurance, even though they constitute a segment of the population with the largest short- and long-term care needs. Across KIIbIH countries with available data, the average health insurance coverage rate of the elderly is 46%. Although in most countries the elderly tend to be better covered than the working-age population, absolute coverage remains very low, notably in pre- and early-dividend countries (Figure 3.15). Limited coverage among the elderly constitutes a crucial vulnerability as older persons have higher needs for acute and potentially high-cost care. Combined with lower income in old age, these factors can lead to catastrophic health spending events (ILO, 2024[5]).

Figure 3.15. Health insurance coverage remains limited in many countries
Copy link to Figure 3.15. Health insurance coverage remains limited in many countries

Health insurance coverage of individuals aged between 15 years and the legal age of retirement and individuals aged above the legal age of retirement, latest year available

Even when health insurance coverage is high, it may concern only basic health services and not cover the cost of expensive, long-term care needs. Depending on countries, coverage by health insurance spans very different realities. Coverage may be fully or partly subsidised by the state, as is the case in countries such as China, Thailand and Viet Nam, and to some extent Indonesia and India. But subsidised schemes for informal workers or vulnerable individuals may provide lower financial protection, for instance though higher co-payments or tighter benefit packages. As the population ages, demand for long-term care services will likely increase. In the absence of adequate entitlements and levels of service provision, this puts additional pressure on household resources. The financial implications could be particularly severe for households that are already poor and/or informal, notably by potentially crowding out resources from contributory schemes and retirement savings and, ultimately, perpetuating the cycle of informality.

Increased demand for long-term care is taking place in a global context of changing demographics and social norms that transform traditional structures of support. A combination of factors – including economic development, increased participation of women in the labour market, reduced family size and rapid ageing of the population – exert downward pressure on family-based informal care (OECD, 2024[14]). These transformations will have lasting impacts on the livelihoods of the elderly. They will also require additional financial resources at the individual and household level to cover the cost of medical and formal care services. Importantly, these trends may also open an opportunity to develop more tax-financed care services, including as a pathway to formalisation (OECD, 2024[14]).

Low enrolment in contributory pension schemes threatens the livelihoods of future retirees

Copy link to Low enrolment in contributory pension schemes threatens the livelihoods of future retirees

In a context of rapid ageing, enrolment of current workers in pension schemes is crucial to accumulate pension rights that can sustain their livelihoods during old age. The portraits of informality presented in Chapter 2 show that the informal economy encompasses workers with very different realities in terms of income, employment status and sectors of activity. Such diversity has impacts on the types of risks workers face and the barriers that exclude them from pension systems. It also renders unrealistic uniform and “silver-bullet” solutions. Attempts to extend pension coverage to informal workers need to take into account this diversity within the informal economy, especially when it comes to differences in terms of potential eligibility, capacity to build-up entitlement or the existence of a disguised employment relationship.

A large pension gap exists between formal and informal workers

Globally, based on administrative data, the ILO estimates that 35% of the working-age population and 59% of the labour force actively contributed to a pension scheme in 2023 (ILO, 2024[5]). These shares are even lower among KIIbIH countries, which mostly include EMDEs with less-developed pension systems and for which data are based on household surveys.4 On average, across KIIbIH countries, only 40% of employed individuals aged between 15 years and the legal age of retirement contribute to any type of pension scheme. Of these, the proportion of individuals that can expect to receive a pension upon retirement may be even lower because of low contribution density and failure to meet different eligibility criteria.

Contributory gaps vary widely across countries at different stages of their demographic transition. Pension enrolment remains extremely low in pre-dividend countries and across many early-dividend countries, whereas the proportion of contributors is much larger in many late- and post-dividend countries. In 12 countries – most of which are late- and post-dividend countries – of the 35 with available data, more than 50% of workers aged between 15 years and the legal age of retirement contribute to a pension scheme (Figure 3.16). In countries such as the Bahamas, Bulgaria, Chile, China,5 Costa Rica, Croatia, Cyprus, Malta, Romania and Uruguay, this share is superior to 70%.

Figure 3.16. In some countries, enrolment in contributory pension schemes is alarmingly low
Copy link to Figure 3.16. In some countries, enrolment in contributory pension schemes is alarmingly low

Share of workers aged between 15 years and the legal age of retirement who contribute to a pension scheme, latest year available

Generally, the overall low share of contributors to pension schemes stems from a huge coverage gap among informal workers, who constitute the majority of the workforce in many countries. Across KIIbIH countries, the average coverage gap (i.e. the proportion of workers covered by a contributory scheme to workers not covered by it) between formal and informal workers stands at 75 percentage points. The average proportion of informal workers contributing to a pension scheme is just 8% (Figure 3.17). In nearly all KIIbIH countries, less than 10% of informal workers contribute to a pension scheme. By contrast, in only 86 of 35 countries included in the KIIbIH, this proportion is superior to 10%, reaching 61% in China.7

Figure 3.17. Low enrolment rates in pension schemes stem from large coverage gaps among informal workers
Copy link to Figure 3.17. Low enrolment rates in pension schemes stem from large coverage gaps among informal workers

Share of formal and informal workers aged between 15 years and the legal age of retirement who contribute to a pension scheme, latest year available

Although many informal workers operate in the formal sector, most do not contribute to any pension schemes, which is largely due to their specific employment arrangements and the design of contributory systems. Across KIIbIH countries, on average, 83% of formally employed workers contribute to pension schemes but only 70% of workers from the formal sector – which can comprise workers both formally and informally employed – do so. The discrepancy largely stems from countries in which large shares of the formal sector workers operate under informal arrangements and are not covered by social protection, thereby not contributing to any pension scheme. Overall, across KIIbIH countries, the average share of informal workers who operate in the formal sector and pay pension contributions is only 4%, a share similar to that of informal workers of the informal sector (Figure 3.18). Apart from a few exceptions, such as Costa Rica, Mongolia and Namibia, enrolment of informal workers who work for formal companies is extremely low, sometimes even lower than the enrolment of informal workers operating in the informal sector.

From a policy point of view, these results underscore the fact that many countries could expand enrolment in contributory pension schemes at a relatively low cost by integrating informal workers who already operate in the formal sector and are, thus, at the margin of informality. However, the factors that underpin this failure to contribute (even where social insurance legislation mandates it for workers of the formal sector) may be complex to address. For example, this failure to contribute may result from a tacit and mutual agreement between workers and employers to trade higher wages for lower social protection entitlements, notably for mobile workers whose portability of pension rights across subnational jurisdictions may not be assured. Where pension contribution rates are high, the conversion process cannot be assumed to be costless in terms of wages or labour market competitiveness, which may deter actors from the formal sector to contribute. Finally, the rapid transformation of labour market dynamics and the emergence of new forms of employment may also play significant roles. Typically, platform or gig workers – who are neither properly formal nor fully informal – continue to often fall in a grey area for whom pension schemes (and more generally, social protection systems) have yet to adapt in most countries. In this regard, emerging approaches, such as the monotax regimes in many countries of Latin America and the Caribbean, look promising.

Figure 3.18. Most informal workers who operate in the formal sector do not contribute to pension schemes
Copy link to Figure 3.18. Most informal workers who operate in the formal sector do not contribute to pension schemes

Share of informal workers of the informal and formal sectors aged between 15 years and the legal age of retirement who contribute to a pension scheme, latest year available

Enrolment in contributory pension schemes mirrors the traditional design of pension schemes, which is grounded in the formal employer-employee relationship

Employment status is an important driver that helps differentiate workers regarding the type of pension scheme they need to cover their old-age risks, the reasons explaining their exclusion from pension schemes, and the channels through which they could access pension schemes. Evidence from KIIbIH shows that the share of workers contributing to pension schemes is primarily driven by large shares (92%) of formal employees contributing to pension schemes. In contrast, the proportion of workers who contribute to pension schemes remains substantially lower for formal employers (35%) and formal, own-account workers (21%). It is extremely limited (below 15%) for all informal workers, regardless of their employment status (Figure 3.19).

Figure 3.19. Contributory pension schemes largely exclude informal workers and non-wage formal workers
Copy link to Figure 3.19. Contributory pension schemes largely exclude informal workers and non-wage formal workers

Share of workers aged between 15 years and the legal age of retirement who contribute to a pension scheme by employment and informality status

The data largely reflect the traditional design of pension schemes, which is grounded in the formal employer-employee relationship. For self-employed workers (own-account workers and employers), as well as for unpaid family workers, a common challenge is their exclusion from statutory access to social protection that tends to primarily cover the needs of salaried employees. In particular, although the design of pension schemes may offer self-employed workers the opportunity to opt in on a voluntary basis, they cannot rely on co-payments and must often pay both the employer’s and worker’s contributions, which can prove too costly. For wage workers without a contract and for own-account workers who are in disguised employment relationships (such as dependant contractors), a fundamental issue is the recognition of the employment relationship and the possibility to benefit from an extension of social protection schemes financed through employer and worker contributions.

Irrespective of employment status, many informal economy workers may have difficulties fulfilling the eligibility conditions for receiving benefits from insurance-based schemes. Their income may be too low and/or irregular to comply with the contributions required. The priorities and needs of informal workers may also differ from those of formal workers. For example, the long contribution periods required to qualify for certain benefits – notably in the case of pensions – can be discouraging and may deter informal workers from participating in these schemes. In Paraguay, workers can retire with a full or partial pension only under one of three conditions: i) 60 years of age and a minimum of 25 years of contributions; ii) 55 years of age and a minimum of 30 years of contributions; or iii) 65 years of age and a minimum of 15 years of contributions. Workers who fail to meet one of these three conditions cannot claim a pension and effectively lose their contributions. Simulations estimate that most workers who fail to fulfil these conditions belong to the poorest two quintiles of the labour income distribution (Bai and Zelko, 2022[15]). In this sort of architecture, entering the pension system comes with a high risk of never seeing contributions materialise. This might discourage low-income and/or informal workers with less long-term predictability in terms of employment and income opportunities.

For a substantial share of informal workers, low-paid work is a binding constraint that limits their capacity to contribute to pension schemes

Evidence from the KIIbIH indicates that low-pay work affects disproportionately informal workers and negatively impacts their contributory capacity. Based on a three-tier classification of workers relative to median labour earnings,8 Figure 3.20 reveals that formal workers are predominantly clustered in the high and medium earnings’ categories whereas informal workers tend to be more present in the medium and low earnings’ categories (see also Chapter 2).

In general, this concentration of informal workers in the lower tier of the earnings’ distribution primarily reflects well-documented differences in skills and productivity (Ohnsorge and Yu, 2022[16]). It also underlines the limited contributory capacity of such low-paid workers, whose present income is not enough to secure their future needs. Across the 21 countries for which data are available (which do not include any pre-dividend countries), only 24% of low-paid informal workers participate in contributory schemes. Across the nine early-dividend countries with available data, this share drops to 1%.

However, the relationship between skills, informality and ability to contribute may differ widely across countries. In many countries where the informal economy is extremely large and informal employment accounts for most of the workforce, most informal workers operate in jobs that require medium skills (level 2 in the International Standard Classification of Occupations [ISCO-08] classification). This includes clerical support workers, service and sales workers, skilled agricultural, forestry and fishery workers, and craft and related trade workers, as well as plant and machine operators and assemblers. This is notably the case in countries such as Benin, Gambia, Lao PDR, Madagascar, Mali, Niger and Togo, where more than 90% of the labour force is informal and more than 90% of these informal workers operate in medium skilled jobs. In other countries (e.g. Argentina, the Bahamas, China, Colombia, Cyprus and the Maldives), informality rates are relatively lower. However, informal workers are disproportionately concentrated in high skilled occupations (levels 3 and 4 in the ISCO-08 classification) such as managers, professionals or technicians and associate professionals. In these countries, informal workers have a higher capacity to finance contributory schemes. In countries such as Costa Rica, Suriname and Uruguay, informality rates are relatively lower but informal workers are disproportionately concentrated in low-skilled and elementary occupations. In these countries, policy instruments – such as fixed nominal subsidies and matching contributions – may have the most impact in raising the contributions of low earners who remain outside the scope of contributory pension systems (OECD, 2024[17]).

Figure 3.20. Low-paid work disproportionately impacts informal economy workers
Copy link to Figure 3.20. Low-paid work disproportionately impacts informal economy workers

Distribution of informal (Panel A) and formal (Panel B) workers aged above 15 years by earnings categories, latest year available

Enrolment in contributory pension schemes remains limited for high-paid informal workers, pointing towards other factors that hinder their ability to contribute

Not all informal workers are low earners; in pre- and early-dividend countries where the informal economy is particularly large, some belong to the upper tier of the earnings distribution. Across KIIbIH countries, on average, 18% of informal workers belong to the high wage category. In certain countries (e.g. Honduras, Kenya, Nigeria and Peru), the share of informal workers clustered in the upper tier exceeds 25%.

Strikingly, most informal workers who belong to the upper and medium tiers of the earnings distribution do not contribute to any pension scheme. Overall, across KIIbIH countries, as high as 72% of high-paid informal workers do not contribute to any pension schemes, compared to 75% for medium-paid informal workers and 76% for low-paid informal workers. The results underline the difficulty of most existing pension schemes to include informal economy workers, even when they might be able to afford contributions. Conversely, most formal workers, including those who belong the lower tier of the earnings distribution, contribute to pension schemes. The average share of formal low-paid workers contributing to pension schemes stands at 67% (Figure 3.21).

Figure 3.21. Even among relatively well-paid informal workers, contributions to pension schemes remain limited
Copy link to Figure 3.21. Even among relatively well-paid informal workers, contributions to pension schemes remain limited

Share of formal and informal workers aged between 15 years and the legal age of retirement who contribute to pension schemes, by earnings categories

Pension scheme enrolment rates across KIIbIH countries show that only a few have been able to integrate informal workers into contributory pension schemes, and that these few contributors are mostly in the upper tier of the earnings distribution. Some countries, such as Costa Rica and Namibia, show a relative high enrolment of informal workers in contributory schemes, but only for the upper tier of the earning distribution, with respectively 50% and 35% of high-paid informal workers contributing. In China, around 60% of informal workers contribute to pensions schemes, regardless of their level of earnings (but data primarily reflect the unique architecture of the Chinese pension system). While most of China’s pension schemes are purely contributory, some important programmes in terms of size and coverage (e.g. the New Rural Pension Insurance scheme) consist of the payment of regular flat insurance premium that opens eligibility rights for a basic social pension, funded via general revenue. These schemes contain an element akin to a contribution (insurance premium), but are not traditional contributory schemes per se, whereby benefits are tied to employment and to the level and density of contributions. Rather, China’s schemes consist of the acquisition of future rights to benefit from a non-contributory pension. These recent schemes have been instrumental in drastically boosting the share of the informal workers covered by the pension system.

Despite the challenge of low contributory pension coverage, the opportunity exists in many countries to channel the earnings of high- and medium-level earners towards contributory schemes. In settings where pension systems are underdeveloped, have limited coverage, or are inaccessible to informal workers, evidence from various contexts indicates that excluded individuals and their households are nonetheless able to accumulate and maintain significant savings through alternative means (e.g. investing in property or participating in voluntary savings schemes) (Joubert and Kanth, 2022[18]; Guven, Jain and Joubert, 2021[19]). The remaining challenges centre around ways to include informal economy workers into general pension schemes, avoid fragmentation and achieve meaningful scale (Joubert et al., 2026[20]).

These findings highlight the difficulty many existing pension systems face, particularly in developing countries, in trying to address specific barriers that prevent different types of workers from enrolling in contributory pension schemes. These include not only the limited contributory capacity among low-paid informal workers, but also the way pension schemes are designed and their inability to serve workers outside the traditional formal employer-employee relationship. As such, a common objective must be to make pension schemes more inclusive of informal workers with limited and/or irregular earnings and more compatible with formal or informal workers who are not employees.

Many pension systems display institutional and structural gaps that prevent adequate coverage of informal workers

Copy link to Many pension systems display institutional and structural gaps that prevent adequate coverage of informal workers

The notion of old-age pension encompasses a wide range of instruments and mechanisms that share the common objective of providing individuals above a certain age with an adequate, affordable and sustainable income to sustain their livelihoods. From a conceptual point of view, pensions can be classified in two main, broad categories: contributory and non-contributory schemes. Contributory pension schemes, which primarily serve consumption-smoothing objectives, require contributors to make regular payments in return for the promise of a lump sum or a steady income upon retirement. By contrast, non-contributory pension schemes – commonly known as social pensions – are funded by general government revenues and are chiefly designed to alleviate poverty among individuals past the retirement age. Eligibility is often based on reaching a certain age (categorical) or meeting certain welfare conditions (means-tested), or a combination of both. Depending on how these conditions are structured and implemented, eligibility can range from being highly targeted to more inclusive and universal.

Beyond this broad categorisation, pension systems structures are extremely diverse across countries and design features can be combined in many different ways (Chomik, O’Keefe and Piggott, 2024[21]).

Contributory schemes are essential to provide workers with adequate benefits

The principles of insurance lie at the heart of the concept behind pensions and old-age income. As individual insurance instruments, contributory schemes play a crucial role in financing pension systems. By offering a reliable and stable financing mechanism, they can reduce the fiscal pressure on a government budget. Generally, contributory pension schemes offer a higher level of protection than non-contributory schemes. Nonetheless, many existing contributory schemes suffer from severe design flaws that limit their coverage, weaken the adequacy of their benefits and/or jeopardise their financial sustainability.

The low coverage of contributory schemes, notably among informal workers, appears to be equally linked to low capacity to contribute for some workers as to a design issue for many others who could contribute. As shown earlier, the large informality footprint in most EMDEs is a substantial factor explaining the low coverage of contributory schemes; in fact, it excludes most informal workers. Historically, these schemes have been developed within the framework of formal working arrangements and the employer-employee relationship. By design, they remain limited to formal employees with contributions typically shared between workers and employers via co-payments. At the same time, as traditional systems of elderly care provided through the family weaken and more women join the labour force, evidence shows that individuals have a capacity to save – often in illiquid form under the form of housing – and that many informal workers may have a strong interest in participating in pension schemes, provided that the right conditions exist (Giles, Joubert and Tanaka, 2025[22]).

Other barriers also play critical roles, such as workers’ myopia to long-term savings, lack of trust in pension systems and fear of changes in parameters. Alternative savings options with higher rates of return and greater accessibility in case of shocks may also be more attractive.

As countries look for ways to extend contributory pension schemes to informal workers, governments must take into consideration some key design and policy questions. The first question is the type of system that is best suited to the needs of informal workers – between so-called defined-contributions (DC) or defined-benefits (DB) schemes. The second question is the interactions among various contributory schemes targeting different types of workers and whether these schemes should be operated separately or in an integrated fashion. The third question revolves around the key specific features that should be integrated to make contributory schemes attractive for informal workers.

Defined-benefits and defined-contribution schemes carry clear policy and fiscal trade-offs that need to be carefully assessed

All existing contributory schemes fall across the spectrum between defined contributions (DC) or defined benefits (DB), with variations, including hybrid structures, combining elements of both such as notionally defined contribution (NDC)9 schemes. All types of schemes carry clear trade-offs. In DC schemes, which rely on individual pension accounts, pension benefits are a product of the amount of money accumulated by a given contributor, including the returns on investment minus the administrative commission (OECD, 2022[23]). In DB schemes, pension benefits are guaranteed at a certain level in exchange for meeting minimum contribution requirements. DB benefits, therefore, are not a product of the amount of money accumulated but are set according to the number of years of contribution and constitute one of the key internal parameters of the scheme.

The nature and holder – i.e. the individual or the state – of risk are another crucial difference between DC and DB schemes. For DC schemes, in the absence of a minimum guarantee from the state, the main potential drawback is the low adequacy of benefits and the fact that workers bear all the risks. For DB schemes, mutualisation of risks and solidarity principles are embedded; the main drawback is a fiscal risk that usually sits with the sponsor of the scheme – often the state.

Worldwide, primarily for historical reasons, DB schemes constitute the most frequent type of pension schemes implemented. In 2023, the ILO estimated that 84% of countries rely on at least one DB scheme and 64% of countries rely exclusively on DB schemes (ILO, 2024[5]). These schemes currently remain the main entry point to receive pension support. Historically, most were initially developed for the public workforce. As such, they were underpinned by strong financial guarantees from the state.

While these schemes share the same objective of guaranteeing an adequate level of pension to future retirees, they have different long-term consequences in terms of coverage, level of benefits, and financial and fiscal sustainability.

DC schemes are financially sustainable by design, but levels of benefits may not be adequate

In DC schemes, financial sustainability is guaranteed by design (in the absence of generous minimum benefit guarantees). Because the level of benefits is a product only of the contributions accumulated by an individual and the returns on such accumulations, pure DC schemes do not rely on fiscal resources. They offer an interesting option to governments that face tight budgetary constraints and make it possible to mobilise resources to finance social protection systems. Moreover, as individual, accounts-based mechanisms, DC schemes tend to offer better transparency of information: any worker can know – at any given point – how much resources they have accumulated. Individual pension accounts establish a transparent link between contributions and subsequent retirement benefits (Joubert, 2014[24]). In particular, and contrary to DB schemes, DC schemes are immune from political considerations regarding the level of benefits distributed to pensioners. In this regard, DC systems (when no minimum benefit guarantee is provided) tend to act as automatic stabilisers, as no fixed promise is made about the level of pension until the point of retirement (i.e. longer life expectancies automatically mean the flow of benefits will be lower or last less long) (Chomik, O’Keefe and Piggott, 2024[21]). Political interference can still materialise from factors such as politically directed investment or through nationalisation of the assets of private pension funds (Price, 2018[25]).

The main challenge of DC schemes lies in their ability to deliver meaningful and adequate income during old age, notably because of low or incomplete contribution density, generous early withdrawal rules, low investment returns and lack of intergenerational solidarity across workers. The level of resources accumulated may not be sufficient to guarantee an adequate level of pension benefits for workers that have low and/or irregular contributions. For informal workers with low levels of contributions, this can become a major barrier to securing old-age income. Also, the fact that informal workers do not benefit from employer co-payments means that the burden of contributing falls entirely on their own shoulders. Even when DC accumulations are more substantial, lump sum withdrawal rules can compromise adequacy, particularly in contexts where annuity markets10 are absent or underdeveloped, and fail to ensure an income stream across retirement (Rusconi, 2008[26]). In contrast to many DB schemes, when the capital in a DC scheme is converted into an annuity or withdrawn in phases, benefits are rarely indexed and thus provide more limited protection against future inflation.

As their design relies on investment returns, DC schemes are also severely exposed to market-based risks. In 2008, for instance, the global financial crisis had dramatic impacts on DC pension plans: in countries with mandatory DC systems, pensions funds experienced investment losses as high as 20-25% (Antolín et al., 2009[27]). Often, these risks are mitigated via life-long investment strategies whereby, as individuals get closer to retirement, the portfolio investment progressively shifts towards low-risk and liquid assets (such as sovereign bonds). In some countries, in the absence of competition and adequate regulation, investment fees have sharply reduced investment returns over time.

DB schemes provide stronger guarantees of old-age income but can face large risks of financial unsustainability

DB systems provide clarity and predictability in terms of future pensions. Although DB systems can exhibit weaker contribution-benefit links because of embedded redistribution mechanisms, by definition, benefits are guaranteed and known in advance by any contributing individual. Apart from the contributions, the financial burden therefore weighs on the pension provider, usually the state. From the workers’ perspective, DB systems translate into stronger guarantees of future revenue. Conversely, DC schemes come with greater uncertainties about the future level of pension and require a basic understanding of capital accumulation and compounded interest rates. In India, for instance, the Atal Pension Yojana (APY) scheme, is a government-backed pension scheme for unorganised sector workers. Subscribers are requested to indicate the level of guaranteed pension they would like to obtain upon retirement, and the corresponding contribution is calculated accordingly. In contexts where financial literacy is low, DB schemes may constitute a more attractive and successful option to enrol workers (as compared with DC schemes).

Ensuring the financial sustainability of DB schemes, however, may be more complex. Evolving contexts and design assumptions, notably when underlying demographic trends start shifting rapidly, often require undertaking parametric reforms. These reforms notably include options such as: raising the statutory age of retirement; extending the minimum contribution period; increasing the level of contributions from workers; shifting indexation rules for benefits in payment (e.g. from wage to price indexation); reducing the accrual rate on contributions; and/or lowering pension benefits. All these types of reforms can prove very costly politically and can lead to inaction and fiscal degradation over the long term.

A small contributory base and a contraction in the number of contributors, together with generous guaranteed benefits, can rapidly jeopardise the sustainability of the DB scheme and lead to severe financial liabilities. Across EMDEs, this vicious circle has materialised for many DB schemes for public servants (and, to some extent, for private sector workers). A review of civil service pension schemes in six major economies – Brazil, Chile, China, India, Malawi and South Africa – showed that many have been structured as completely separate from the general scheme, often with lower funding ratios, higher liabilities per member than other arrangements, and no accumulation of reserves (OECD, 2020[28]). In many cases, because of their political sensitivity, imbalances of public DB schemes have been increasingly covered by resources stemming from general revenue instead of standard contributions (OECD, 2020[28]).

In NDC schemes, which are akin to DB schemes in that they operate as pay-as-you-go mechanisms, the benefit level is adjusted according to life expectancy at retirement of each cohort. This ensures fiscal sustainability but may compromise adequacy.

Overall, DB schemes have largely failed to integrate informal workers in a meaningful and scaled way. In most cases, to qualify for a DB pension, a worker is required to make substantial contributions over an extended period of time and for a minimum number of years. These contribution requirements do not necessarily suit the profile of informal workers, who often have unpredictable and irregular income, or show a tendency towards partial withdrawal. Strict vesting rules11 may also penalise certain individuals, such as women who may take career breaks to raise children or for those who switch between formal and informal sectors over their careers.

Raising the economic value of contributions via subsidies or matching mechanisms can help attract more workers

Increasing the economic value of contributions via financial incentives is crucial to make contributory pension schemes attractive. Historically, tax incentives have dominated, providing favourable tax treatment to retirement savings as compared to other types of savings. However, tax incentives tend to favour the richest who are subjected to income tax while yielding only limited benefits for the poorest segments of the population (Tanzi and Zee, 2001[29]).

More recently, new forms of financial incentives have emerged that are weakly linked to the tax system. Notably, these include matching contributions, by which the government matches a worker’s payment into a pension scheme – typically up to a nominal ceiling – and nominal subsidies, by which the government directly deposits funds into the pension accounts of eligible individuals. These mechanisms can be applied in both DB and DC schemes. Importantly, such incentives are financed through general revenues – such as value-added tax (VAT) – that are collected from all taxpayers. As a result, they may subsidise workers who are not among the poorest, thereby creating implicit transfers from lower-income individuals to relatively better-off segments of the population. In reality, it could be argued that, by covering their deficits with general revenues, many governments have long subsidised unfunded and financially imbalanced DB schemes.

The tools described above could substantially encourage enrolment in pension schemes, but the evidence remains mixed due to their relative novelty and the lack of long-term evaluations of their effectiveness. Evidence from OECD countries underlines that non-tax incentives, such as fixed nominal subsidies, are better tools to encourage retirement savings among low-income earners, who are less sensitive to tax incentives (OECD, 2018[30]). In EMDEs, evidence on the impact of such tools is more limited, although recent research suggests that they could have real impacts, notably for workers with low income and limited contributory capacities and, more generally, for salaried workers (Giles, Joubert and Tanaka, 2025[22]). Important uncertainties remain regarding the optimal design of these mechanisms, notably because of the multiplicity of systems and designs existing across the world (Joubert et al., 2026[20]).

Fragmentation of pension systems is a challenge in many countries

How to articulate diverse contributory pension schemes that target different types of workers remains an important challenge for many governments throughout the world. Pension schemes have often been fragmented between public and private formal employees. In addition, many countries have established new, dedicated and separate schemes to enrol informal workers (OECD, 2024[31]). These schemes tend to differ greatly from pension schemes offered to formal employees in terms of design features, level of contribution and level of benefits. Institutional fragmentation of the pension system can result in the concentration of high-risk workers in atypical forms of employment into specific schemes, increase administrative inefficiencies and hinder portability of pension rights (OECD, 2025[32]; IMF, ILO and the World Bank, 2024[33]). Instead, policies should aim to pool as many people as possible in the same scheme, even if access can be differentiated to maximise enrolment, for instance through different registration procedures, methods to collect contributions and so forth.

Increasingly, pension systems seek to integrate innovative design features to encourage the participation of informal workers

In recent years, certain countries have taken more innovative approaches to expand the coverage of contributory schemes to uncovered segments of the population and, notably, informal workers. These approaches seek to build on existing structures and mechanisms to bolster enrolment, while also developing new features designed to meet the expectations of informal workers and raise the attractiveness of pension schemes.

To overcome the hurdles of administrative procedures and lack of trust, some contributory schemes have leveraged local aggregators and networks (Reyes Hartley and Abels, 2025[6]). In Rwanda, co-operatives play a central role in enrolling members into Ejo Heza – the country’s flagship voluntary DC scheme, primarily targeted at informal workers. In 2023, estimates indicated that over 60% of Ejo Heza members joined through cooperatives (Guven and Jain, 2023[34]). In Kenya, the Haba Haba scheme – a voluntary pension scheme targeting workers in informal employment – specifically targets organised groups in peri-urban areas, such as transport service providers and market vendors (World Bank, 2024[35]). Overall, aggregators and local networks can boost enrolment through peer effects. They can also improve outreach by facilitating default contributions, collecting payments (notably in remote and rural areas), enhancing trust, and providing information (Giles, Joubert and Tanaka, 2025[22]). Such mechanisms can also help reduce the cost of recruiting informal economy workers into pension schemes (Joubert et al., 2026[20]). Despite these successes in attracting and registering participants, contribution levels remain too low to ensure adequate retirement benefits. Many members express a stronger preference for short-term emergency savings, suggesting that these schemes often reach individuals with incomes too low to sustain long-term savings. This highlights the need for better targeting and for aligning product design and benefit packages with the actual needs of participants.

To better meet the needs and expectations of informal workers, some countries have started to bundle pension benefits with other types of benefits. These bundles may include short-term benefits (e.g. microloans, maternity coverage, health insurance, sickness and unemployment benefits) and/or long-term benefits (e.g. funeral expenses, life insurance, and survivor or disability pension) (Reyes Hartley and Abels, 2025[6]; Joubert et al., 2026[20]). Bundling offers two main advantages. First, it provides informal workers with a risk coverage similar to that of formal workers. Second, it encourages informal workers to participate in pension schemes by offering them products that may be more immediately appealing.

Evidence across the world shows that schemes that have successfully enrolled a substantial share of informal workers often hinge on mechanisms that allow for flexible contributions. While KIIbIH data show that a non-negligible share of informal workers tends to have some contributory capacities, their income is often volatile and irregular. This means they cannot comply with mandatory contribution rules that characterise most pension schemes – and ultimately results in their exclusion from pension schemes. To address this barrier, many pension schemes designed for informal workers have introduced mechanisms that allow flexible and irregular contributions. Often, this flexibility takes the form of minimum annual mandatory contributions instead of monthly fixed contributions (as required in traditional pension schemes) (Joubert et al., 2026[20]). Flexibility in pension contributions can be enhanced through increased use of digital tools that allow informal workers to make small contributions at any time – or so-called “consumption-based” pensions (Rupper Bulmer, Winkler and Mote, 2017[36]).

Pensions schemes for informal workers also increasingly include options for early withdrawals but these features carry significant risks. For low-income, informal workers with limited savings and no social protection, the possibility to tap into pensions savings – after a vesting period or for a portion of contributions – can constitute an attractive safety net in emergencies. Provident funds often formalise this through separate accounts: one for retirement income and another for short-term contingencies, which may cover specified needs (e.g. health shocks, housing, education) and can be used more flexibly. A major challenge is that early withdrawals can substantially undermine investment returns. Schemes that allow for short-term access must maintain liquidity, which can weaken overall portfolio performance (Joubert et al., 2026[20]). Early withdrawals can also compromise future retirement adequacy. In Peru, authorised early withdrawals from pension funds between 2020 and 2024 led to disinvestment equivalent to nearly 11% of GDP (Reyes Hartley and Abels, 2025[6]). In Chile, exceptional withdrawals authorised during the COVID-19 pandemic resulted in 37% of social security affiliates fully depleting their contributory pension balances by 2021 (Inzunza and Madeira, 2025[37]).

Non-contributory schemes are crucial to protect the livelihoods of the most vulnerable

Non-contributory pension schemes, used alongside contributory schemes, are crucial to pension systems. This is particularly true where a large share of the working-age population is inactive or lacks sufficient contributory capacity to secure an adequate stream of income during retirement. In many countries, wide pension coverage gaps stem from a substantial low-paid, informal workforce and a large segment of the population that is inactive, especially among women who may be incited to stay at home to care for the household and its members. To address these challenges, numerous countries have established non-contributory pension schemes, often referred to as social pensions. By design, entitlement to these tax-financed schemes is not tied to employment or payroll contribution records. This allows such schemes to bridge large gaps in contributory pension beneficiaries. Since the early 2000s, the expansion of non-contributory pensions has been particularly pronounced across the world, notably in countries in Latin America and the Caribbean and in Asia and the Pacific (ILO, 2024[5]). This growth has provided workers of the informal economy with a minimum income in old age and has notably supported the most vulnerable segments of society, including the poorest households and women.

Non-contributory pension benefits are often insufficient to ensure older persons can maintain a decent standard of living

The effectiveness of non-contributory schemes as an anti-poverty measure and to provide the elderly with a basic minimum income varies widely across countries. It depends largely on factors such as the benefit levels and eligibility conditions, both of which are closely intertwined with fiscal sustainability. Since non-contributory schemes are financed via governments’ general revenues, any increase in benefit levels or in the number of beneficiaries can significantly raise government spending and negatively impact fiscal balance. Article 67(c) of the ILO Convention No. 102 of 1952 stipulates that non-contributory pensions, including those subject to means-testing, should be sufficient to maintain beneficiaries and their families in health and decency (ILO, 1952[38]). In practice, however, many countries prioritise other fiscal objectives resulting in low coverage of social pensions and/or inadequate benefit levels (United Nations, 2018[39]).

In many countries, non-contributory pension benefits are too low to adequately sustain the livelihood of the elderly, even when conceived as a basic and anti-poverty measure. In 2023, a review of 100 schemes from 79 countries estimated that the average amount of benefits offered by non-contributory pension schemes accounted for 28% of the minimum wage, and the minimum amount of benefits offered represented only 20% (ILO, 2024[5]). In many countries, the level of non-contributory benefits stood well below the national poverty line. In Bangladesh, India and Sri Lanka, the minimum level of tax-financed, old-age pension benefits for a single person accounted for less than 15% of the national poverty line.

Legal age of eligibility to claim non-contributory benefits are sometimes too low in relation to changing demographics

Countries have adopted diverse approaches to determine eligibility for non-contributory pensions. These schemes may form part of broader social assistance programmes (e.g. cash transfers) targeted at poor households, thereby including older persons who fall within this category. Alternatively, they can be established as stand-alone programmes, consisting of dedicated non-contributory pension schemes that focus explicitly on older individuals, regardless of their economic status (Reyes Hartley and Abels, 2025[6]).

Irrespective of the form under which non-contributory old-age income is provided, eligibility criteria remain a key parameter of both coverage and benefit adequacy. Universal non-contributory pensions – for which anyone above the eligibility age qualifies – tend to naturally achieve the broadest coverage. However, depending on the size of the eligible population, this approach can entail substantial fiscal costs, which may become unsustainable in the context of population ageing coupled with low capacity to collect tax and generate sufficient public revenue. In this regard, the eligibility age is crucial to balancing two objectives: providing a minimum basic income to older individuals who can no longer work and maintaining fiscal sustainability.

Comparisons between objective measures of life expectancy and the legal age for claiming non-contributory benefits indicate that eligibility ages are relatively low in several countries. In nearly all KIIbIH countries, individuals who reach the eligibility age for a social pension still have more than 10 years of healthy life remaining, based on healthy life expectancy at age 60. In 8 countries, this period exceeds 15 years, reaching up to 20 years in Malawi (Figure 3.22).

Figure 3.22. The legal age of pension eligibility is relatively low compared to life expectancy
Copy link to Figure 3.22. The legal age of pension eligibility is relatively low compared to life expectancy

Difference, in years, between total and healthy life expectancy at age 60 and men’s legal age of eligibility to claim non-contributory benefits

A relatively low eligibility age has become a growing challenge for many countries that struggle to ensure the financial sustainability of non-contributory pension schemes. Rather than raising the eligibility age – which is often politically sensitive – governments have frequently opted to keep benefit levels low. In some cases, this has been exacerbated by de-indexation, such that the real value of benefits has eroded over time due to inflation. This approach disproportionately affects informal workers, who often continue working beyond the legal eligibility age and, once they stop working, must rely solely on these modest non-contributory benefits.

Reforming the eligibility age is, therefore, critical not only for improving the financial sustainability of non-contributory schemes but also for ensuring adequate and equitable support in old age. To be effective and fair, the retirement age should be informed by key indicators such as the median age at which informal workers exit the labour force. Combined with efforts to maintain or increase benefit levels, adjusting the eligibility age offers a more sustainable and equitable path forward for informal workers.

Non-contributory schemes often integrate additional targeting and eligibility criteria that substantially restrict the number of beneficiaries and increase administrative costs

To contain fiscal cost and channel resources towards those most in need, many countries have introduced targeting mechanisms and eligibility criteria other than age, effectively restricting non-contributory benefits based on observable measures of welfare. For various reasons, including fiscal constraints and political economy considerations, many social pensions are means-tested or proxy-means-tested (PMT) and, in some cases, community-based-targeted (CBT). Under these approaches, eligibility is determined by an individual’s or household’s income, assets or any other proxy for welfare. Some countries combine means tests with pensions tests, granting non-contributory benefits only to those who do not receive contributory benefits. In Viet Nam, individuals aged between 60 and 79 years are subject to strict means-testing, which results in only about 2% of that cohort receiving benefits. Conversely, those over the age of 80 years are pension-tested, resulting in about three-quarters of that cohort receiving benefits (Chomik, O’Keefe and Piggott, 2024[21]).

Recent evidence suggests that targeting – notably means-testing – has yielded mixed to negative results. Widespread informality in many countries often renders the pre-retirement incomes of informal workers as largely unobservable. This limits the ability of governments to effectively restrict social pensions to the poorest (Joubert et al., 2026[20]). Even when eligibility to non-contributory benefits relies on proxies that are theoretically easier to observe (e.g. demographics, education, housing type, durable goods or means of production), exclusion errors of intended recipients remain substantial. A 2020 review of 38 poverty-targeted social assistance programmes across 25 low- and middle-income countries found exclusion errors ranging from 44% to 97%. In other words, between 44% and 97% of intended beneficiaries were excluded from these programmes (Kidd and Athias, 2020[44]).

Automatic adjustment mechanisms have been crucial to maintain the adequacy of benefits over the long term

Country experiences show that benefits indexation to price inflation and/or wage growth is critical for maintaining the adequacy of benefits. Without indexation, benefit levels can erode substantially over time, bringing their real value below the poverty line and undermining the social objective of non-contributory schemes. However, indexation can substantially increase nominal costs, notably in high inflation contexts, and can pose technical difficulties in calibrating and benchmarking adjustment mechanisms (Gentilini et al., 2021[45]). These potential long-term cost increases primarily depend on the objective, size and design of the scheme and, in particular, on its adjustment mechanism. Broadly, indexation policies hinge on three main parameters: i) the method of indexation (whether adjustments are automatic, ad hoc, or absent); ii) the underlying benchmark or indicator to determine the level of adjustment (e.g. price inflation, wage growth or a combination of both); and iii) the frequency and timing of adjustments.

In 2021, a World Bank review of 56 social pension programmes across developed and developing countries found that 38% (i.e. 21 programmes) had automatic indexation, 55% relied on ad hoc adjustments, and 5% had no mechanism. An additional 2% lacked sufficient data to be assessed (Gentilini et al., 2021[45]). Most automatic adjustments were anchored against inflation, with only a small fraction tied to wages or a combination of both. By contrast, ad hoc adjustments were largely discretionary and not informed by any clear underlying indicator, depending rather on factors such as budget availability or political considerations (e.g. benefit increases before elections).

In response to rapidly evolving demographic conditions, countries are increasingly introducing automatic adjustment mechanisms (AAMs) in their pension systems. AAMs are predefined rules that automatically adjust pension parameters (e.g. eligibility age or benefit levels) based on the evolution of demographic, economic or financial indicators. These mechanisms can apply to both contributory and non-contributory schemes. They can also take multiple forms, including: linking eligibility age to life expectancy; indexing benefits to life expectancy, demographic ratios, wage bill or GDP; and ensuring short- and long-term financial balance of schemes through combined changes in level of benefits, contribution rates or pension points (Chomik, O’Keefe and Piggott, 2024[21]). While AAMs are not immune to political interference, and their parameters can be changed, diluted or cancelled, they can help insulate pension systems and protect financial stability from the impacts of a changing and uncertain environment (de Tavernier and Boulhol, 2021[46]).

The fiscal implications of automatic adjustment mechanisms vary, depending on the parameters used

From a policy perspective, indexing non-contributory benefits is essential and widely considered best practice. However, adjustment parameters must be carefully tailored and calibrated to each country’s fiscal and economic context. Recent high-inflation episodes have renewed attention on indexation rules. The two main indicators for indexation – prices and wages – carry trade-offs with different fiscal implications. Price indexation, the most common approach, tends to reduce the relative value of pension benefits over time, particularly during periods of real-wage growth that are fuelled by productivity gains. Conversely, in high-inflation episodes and when real wages stagnate or decrease, price indexation can become more favourable for pensioners but sharply deteriorate the situation of public finances or pension providers, as seen recently in OECD countries (OECD, 2023[47]). Wage indexation preserves benefits relative to average wages throughout the retirement period, reducing future poverty risks and maintaining retirees’ living standards. However, this method requires reliable wage data that are often less readily available in EMDEs, making price-based adjustments more practical despite their limitations.

Successful pension systems coherently articulate non-contributory and contributory schemes

In context of widespread informality, a well-functioning pension system requires coherent integration of contributory and non-contributory components. Countries that rely heavily on a single, dominant non-contributory scheme often face sustainability challenges, while those that depend exclusively on contributory schemes fail to provide adequate coverage across the entire population (Price, 2019[48]).

Ideally, non-contributory and contributory schemes should form complementary elements within a multi-pillar system that pools risks across the population and reconciles key policy objectives, notably solidarity and economic efficiency. In recent years, several countries have pursued this approach, with some evident success. Landmark reforms establishing efficient multi-pillar systems include those in China and Chile (OECD, 2020[28]). In Chile, universal non-contributory benefits begin to decrease once a pensioner’s contributory benefits exceed a certain threshold and are gradually reduced as those benefits rise (Inzunza and Madeira, 2025[37]; Ugarte and Vergara, 2022[49]; Joubert, 2014[24]).

Financing the expansion of pension systems is challenging but significant opportunities exist

Copy link to Financing the expansion of pension systems is challenging but significant opportunities exist

In many developing countries, low pension coverage is closely tied to limited financing for social protection, often reflecting low public revenue. According to ILO estimates, in 2023, high-income countries allocated 16.2% of GDP to social protection (excluding healthcare), compared with 8.5% in upper-middle income countries, 4.2% in lower-middle income countries, and less than 1% in low-income countries (ILO, 2024[5]). Tax revenues as a share of GDP also remain limited in many developing countries, standing at 15% to 20% of GDP – roughly half the level observed in OECD countries (34% on average in 2022) (OECD, 2025[50]). These figures underscore the challenge of financing the expansion of pension systems. But they also reveal strong potential for governments to mobilise additional domestic resources, through both general taxation (i.e. extension of non-contributory pensions) and social security contributions (i.e. extension of contributory schemes).

Tax reforms can generate more public revenue to finance the expansion of non-contributory pension

The establishment and expansion of multi-pillar pension systems, with a basic non-contributory component, hinges on a given country’s ability to mobilise sufficient public revenue. Reprioritising government expenditure and improving spending efficiency, while important, will not be sufficient on their own. Importantly, in such contexts, most developing countries have significant untapped tax potential. Although potential gains differ greatly across regions and countries, the tax revenue frontier (i.e. the tax revenue attainable given a country’s structural characteristics) is 5% of GDP higher than current revenue for low-income countries and about 9% of GDP for lower-middle-income countries (Brys et al., 2025[51]; OECD, 2024[31]). For most developing countries, raising the tax-to-GDP ratio by five percentage points over a decade is ambitious but considered feasible (IMF, OECD, United Nations and the World Bank, 2016[52]). Achieving such an increase also depends on tax buoyancy, i.e. the extent to which tax revenues rise as the economy grows. In practice, sustained gains in the tax-to-GDP ratio require a buoyant tax system, in which existing taxes naturally generate more revenue as incomes and consumption expand. This, in turn, depends on reforms that broaden the tax base, reduce exemptions, improve compliance and strengthen tax administration.

To generate the necessary resources, governments can rely on a range of fiscal instruments. These include broadening the personal income tax base, introducing simplified and compulsory tax regimes (including presumptive contributions for microenterprises and their workers), and implementing new health and environmental taxes on products (such as tobacco, alcohol, sugar-sweetened beverages and carbon emissions) (OECD, 2025[32]; Coxhead and Grainger, 2018[53]). Reassessing and reducing inefficient corporate tax incentives, as well as redesigning or eliminating detrimental subsidies (e.g. on fuel) can also free up fiscal space for financing non-contributory pensions. In recent years, some countries (e.g. Mongolia or Zambia) have also earmarked revenues from natural resources to support social pensions (UNICEF, 2025[54]; ILO, 2016[55]). Other countries have sought to maximise indirect tax revenues (e.g. VAT or sales taxes) by reducing exemptions on non-essential goods and channelling these funds towards social protection programmes.

Any strategy to increase tax revenue must be carefully designed and evaluated at the country level to assess its revenue generation potential against its broader effects on social, economic and environmental dynamics. The design and implementation of reforms should reflect each country’s unique structural characteristics and remain adaptable to evolving circumstances. When evaluating potential tax measures, countries should systematically consider whether there is scope to achieve four things (one, a combination or all): i) improve compliance; ii) limit international leakages; iii) broaden the tax base and minimise domestic leakages; or iv) raise rates or introduce new taxes (OECD, 2024[31]). Most importantly, it is essential to assess the impacts of tax reforms on economic development and investment, ensuring that fiscal policy supports inclusive and sustainable growth. In this regard, taxes that are generally less harmful to economic growth (e.g. property and indirect taxes) may constitute better options (IMF, OECD, United Nations and the World Bank, 2016[52]).

Universal social pensions could be established at a relatively low fiscal cost and would generate large positive spillover effects on other groups of the population

The principle of universal social protection is rooted in social justice and human solidarity. Both the ILO “Social Protection Floors Recommendation” (Recommendation No. 202) and the ILO “Resolution concerning the second recurrent discussion on social protection” emphasise that access to universal social protection is fundamental for achieving social justice, decent work, and inclusive and sustainable development (ILO, 2021[56]; ILO, 2012[57]). The approach recognises social security as a human right and seeks to promote health and dignity. In this context, non-contributory pension schemes play a vital role in closing social protection gaps for older people.

Beyond this developmental and rights-based perspective, recent evidence suggests that establishing universal non-contributory pensions (allocated exclusively on the basis of age) would have limited fiscal cost – provided the eligibility age is not set too low and benefits remain reasonable, consistent with anti-poverty objectives. Recent simulations in Indonesia, which is characterised by large informality (80% of employment) and rapid ageing, show that two combined reforms – an increase in the access age for the formal pension and the introduction of a social pension – could yield large welfare gains for the entire society and better shield most of the population against old-age adverse uncertainties while keeping the fiscal cost of the contributory pension system under control (Kudrna, Piggott and Poonpolku, 2024[7]).

New estimates from KIIbIH data confirm that universal social pensions can be fiscally affordable. Based on country-level data for the latest year available, Figure 3.23 presents cost simulations for universal non-contributory pensions under varying assumptions regarding access age and benefit levels (see Annex 3.B for more details on the methodology and results for all scenarios). Across most KIIbIH countries, the cost of providing a universal basic pension equal to 50% or 75% of the median per-capita income12 remains within a range of 0.5% to 4% of GDP, depending on the access age. More specifically:

Providing a benefit equal to 50% of median per-capita income to all individuals above the current legal retirement age would cost less than 1.6% of GDP in half of the KIIbIH countries, and less than 2.2% in three-quarters of them (Scenario 1).
Providing a similar level of benefit (50% of median per-capita income) to individuals aged 60 and over (Scenario 2), 65 and over (Scenario 3) and 70 and over (Scenario 4) would cost, respectively, less than 1.7%, 1.2%, and 0.7% of GDP in half the KIIbIH countries, and less than 2.4%, 1.7% and 1% in three-quarters of them.
By setting the access age to life expectancy at age 60 minus 10 years (Scenario 5), the cost of providing a benefit equal to 50% the median per-capita income is less than 1.1% of GDP in three-quarters of the countries.
By tying the access age to life expectancy at age 60 minus 5 years (Scenario 6), the cost of providing a benefit equal to 50% the median per-capita income would fall below 0.7% of GDP in three-quarters of the KIIbIH countries. Increasing the level of benefits to 75% of the median per-capita income maintains the cost below 1.1% of GDP in three-quarters of the countries.

Figure 3.23. Establishing universal non-contributory pensions could be done at a relatively low fiscal cost
Copy link to Figure 3.23. Establishing universal non-contributory pensions could be done at a relatively low fiscal cost

Estimated cost of universal social pension as a share of GDP across KIIbIH countries

In addition to the potential to deliver on policy objectives at relatively limited fiscal cost, establishing universal, non-contributory pensions could also produce significant indirect, positive effects on children and poor households, specifically through intra-household transfers. This is notably true in countries where the elderly tend to live in intragenerational households. For instance, under a scenario in which universal benefits are distributed to all individuals aged above 65 years (Scenario 3), on average across all KIIbIH countries, the median share of beneficiaries would be 7.2% of the population. Meanwhile, positive impacts would reach a median share of 13.8% children and 23.3% of poor households (Table 3.1).

	Median share (%) impacted:					Median cost (% of GDP) at:
	Individuals	Households	Poor individuals	Poor households	Children	50% of median per-capita income	75% of median per-capita income	Median per-capita income
Scenario 1: Benefits claimed at current legal age of retirement
KIIbIH	10.2	30.1	6.9	31.7	18.6	1.6	2.3	3.1
Pre-dividend	5.7	24.4	5.4	25.7	19.2	1.2	1.8	2.4
Early-dividend	8.4	29.3	6.9	32.5	20.7	1.4	2.1	2.7
Late-dividend	14.7	32.1	8.8	30.6	15.7	1.9	2.9	3.8
Post-dividend	22.4	39.7	25.0	46.4	17.0	2.3	3.5	4.6
Scenario 3: Benefits claimed at age 65 years
KIIbIH	7.2	23.2	5.0	23.3	13.8	1.2	1.8	2.3
Pre-dividend	3.7	15.9	3.5	18.1	12.8	0.8	1.2	1.6
Early-dividend	5.4	22.1	4.8	23.0	13.8	0.9	1.3	1.8
Late-dividend	10.5	24.5	7.1	21.2	12.0	1.4	2.1	2.8
Post-dividend	17.2	36.0	22.0	39.9	14.8	1.9	2.8	3.7
Scenario 5: Benefits claimed 10 years prior to life expectancy age
KIIbIH	5.5	17.5	4.1	17.4	10.6	0.8	1.2	1.6
Pre-dividend	3.1	13.7	3.2	15.3	10.9	0.6	1.0	1.3
Early-dividend	4.5	17.9	4.1	18.4	11.4	0.7	1.0	1.4
Late-dividend	6.7	16.6	4.9	14.9	8.7	1.0	1.5	1.9
Post-dividend	11.0	22.1	13.6	26.5	10.4	1.2	1.8	2.3

Note: For each selected scenario, the table presents coverage and cost estimates. In Scenario 1, the access age is set at the current legal age of retirement. In Scenario 3, the access age is 65 years. In Scenario 5, the access age is set 10 years prior to life expectancy at age 60. Depending on the country, median per-capita income is computed as the household total income or total consumption divided by the number of household members. For all countries, GDP data are based on the latest year available in the KIIbIH.

Source: Authors’ simulations based on OECD (2026[4]), IMF (2025[58]) and United Nations (2021[43]).

Statlink

In many countries, opportunities exist to collect more social security contributions to support expansion of contributory schemes

For many countries, increasing contributory revenues remains challenging – but not impossible. Since the early 2000s, revenues from social insurance contributions have remained stable or even grown in numerous contexts, serving as a major source of financing for social protection systems. Globally, social security contributions accounted for 18.7% of total tax revenue and 5.3% of GDP in 2023 (OECD, 2025[50]).

Income-based social security contributions should remain the foundation of contributory pension schemes, as this model offers several advantages. First, it relies on the administrative and operational simplicity of collecting payroll taxes. Second, it directly links benefits to workers’ earnings and contributions, thereby incentivising formal employment. This model also enhances transparency, as individuals can clearly see the relationship between their contributions and future benefits. Additionally, it allows to better control programme costs and provides financial stability through earmarked funding (Calligaro and Cetrangolo, 2023[59]).

Data from the KIIbIH further highlight the potential to mobilise greater social security contributions, particularly by tapping into the substantial share of informal workers with the capacity to contribute. Across KIIbIH countries, high-income individuals account for 16% of informal workers who do not contribute to any pension scheme (Figure 3.24). With appropriate mechanisms in place, these workers could represent a valuable source of additional financing. As a large proportion of non-contributing informal workers remain in the medium- and low-income brackets, governments will still need to provide subsidies and matching contributions as part of a broader strategy to increase the coverage of contributory pension schemes among low-paid workers and ensure the financial sustainability of such schemes.

Figure 3.24. Many informal workers who do not contribute may have sufficient contributory capacity
Copy link to Figure 3.24. Many informal workers who do not contribute may have sufficient contributory capacity

Distribution by earnings categories of all, formal and informal workers aged between 15 years and the legal age of retirement who do not contribute to pension schemes

Introducing presumptive tax regimes (PTR), also known as monotax systems, could boost revenue collection for social security contributions. In recent years, these regimes have become increasingly popular, notably in Latin America and the Caribbean, with examples including the monotributo systems in Argentina and Uruguay. PTRs allow micro and small enterprises, as well as self-employed individuals, to pay a single, consolidated tax that covers multiple obligations – including social security contributions. This approach reduces administrative burdens for tax authorities and lowers compliance costs for taxpayers. The revenue generated is allocated, by the administering authority, among relevant institutions, such as tax administrations and social security funds (OECD, 2024[60]). Importantly, successful implementation of this approach requires careful calibration of the eligibility criteria and the tax structure in order to balance i) incentives to join the PTR and formalise, ii) the integrity of the standard tax regime, and iii) equity among taxpayers (OECD, 2025[61]; Mas-Montserrat, Colin and Brys, 2024[62]; Mas-Montserrat et al., 2023[63]). This approach would also entail making it mandatory for independent workers to contribute to social security.

Conclusion and policy discussion

Copy link to Conclusion and policy discussion

The demographic transition towards ageing and aged societies is accelerating across all regions and countries of the world. Africa is still at an earlier stage of this transition, but ageing pressures are expected to intensify over the coming decades. Globally, the pace of the transition, coupled with the fact that it is advancing in contexts in which labour informality continues to dominate, may threaten the livelihoods of millions of people and trigger crucial policy challenges. Urgent action and forward-looking planning are needed, even in countries that still have a youth bulge. While most countries use both contributory and non-contributory pension schemes to support the income needs of elderly people, pension coverage gaps remain enormous in many. In most countries, limited fiscal space challenges the ability of non-contributory schemes to ensure adequate livelihoods for the elderly, reflecting both low levels of domestic resource mobilisation and the growing number of older individuals requiring protection. The coverage and adequacy of contributory schemes are, in parallel, constrained by the prevalence of informal employment, which often limits the contributory base and the level of contributions.

Looking ahead, policymakers must develop strategies that address the triple challenge of coverage, benefit adequacy and fiscal sustainability. Effective approaches should aim to provide the largest possible share of the population (coverage) with pension benefits that are high enough (adequacy), without jeopardising the fiscal balance or the internal financial equilibrium of pension schemes (sustainability) (Rupper Bulmer, Winkler and Mote, 2017[36]; United Nations, 2018[39]). In addition to these core principles, considerations of fairness, equity and administrative feasibility are essential for building broad support and ensuring successful implementation. Policymakers should also consider the increasing role of technology and digital public infrastructure to advance financial inclusion of informal workers, and to better target and deliver social benefits. Finally, any national strategy should be informed by a thorough understanding of the policy space available, which varies across countries according to the prevalence of informal employment, the demographic situation and the level of tax revenues.

Identify how informality, demographic dynamics and fiscal revenues shape the policy space

To be successful, policy strategies for pensions must account for diverse local realities across three key dimensions: demographic transition, informality prevalence and ability to increase the fiscal space. Figure 3.25 provides simple correlations of the informal employment rate with the projected share of individuals aged above 65 years by 2050 (Panel A) and with the tax-to-GDP ratio (Panel B). From this figure, it is possible to extract useful patterns and to develop a typology of seven groups of countries according to country-level priorities and available policy options.

Group 1: Countries with low informality but at a later stage of the ageing process and with already high tax-to-GDP ratios (quadrants A1 and B1). This group includes a mix of European and Latin American countries such as Brazil, Bulgaria, Costa Rica, Croatia, Malta, Romania and Uruguay, as well as Armenia and the Maldives. Nearly all are in the late- or post-dividend phase. While all have made substantial progress towards formalisation and domestic resource mobilisation, they face imminent demographic pressures. Such countries should prioritise leveraging their formal workforce to quickly consolidate the financial sustainability of contributory pension schemes. This would allow them to address the needs of a rapidly growing elderly population by reprioritising expenditures and channelling existing resources towards pension schemes.
Group 2: Countries with low informality but at a later stage of the ageing process and with relatively low tax-to-GDP ratios (quadrants A1 and B3). These countries typically have tax-to-GDP ratios close to 20% (examples include the Bahamas and Chile). Their situation is similar to Group 1 countries, except that their current fiscal space is much more limited. The main objective should be to increase markedly their fiscal revenues to enhance non-contributory pension schemes.
Group 3: Countries with low informality, at an early stage of the ageing process and with already high tax-to-GDP ratios (quadrants A3 and B1). These countries are in a favourable position to develop robust pension systems (examples include Mongolia and South Africa). With more time left before demographic pressures intensify, they can rely on a relatively large formal workforce to strengthen and expand contributory pension schemes. As the scope of these countries to increase substantially tax revenues may be constrained, they should carefully manage their fiscal space and reprioritise expenditures according to the evolving needs of their population.
Group 4: Countries with high informality and at a later stage of the ageing process but with relatively low tax-to-GDP ratios (quadrants A2 and B4). These countries (which include the Dominican Republic, Indonesia, Mexico, Peru, Thailand and Viet Nam) face an important policy challenge. They have little time before becoming “aged” or “superaged” societies, while widespread informality limits the coverage of contributory pension schemes. These countries should pursue formalisation efforts and expand tax revenues (through a mix of contributory and non-contributory pensions) to meet the needs of a growing elderly population. Importantly, high levels of informality may not necessarily constitute a barrier to mobilise more fiscal resources in these countries. Indeed, several countries with similar or higher levels of informality still manage to collect revenues exceeding 20% of the GDP (Figure 3.25, Panel B).
Group 5: Countries with high informality, at a later stage of the ageing process and with already high tax-to-GDP ratios (quadrants A2 and B2). These countries face the greatest policy challenge (examples include Argentina, Barbados, China, Colombia, El Salvador and Jamaica). In context of substantial informality, they must address urgently the needs of their rapidly ageing societies – but with probably little room left to leverage additional fiscal resources. Public policies should address this by focusing on public expenditure reforms and formalisation efforts.
Group 6: Countries with high informality, at an early stage of the ageing process and with relatively low tax-to-GDP ratios (quadrants A4 and B4). Some 20 countries13 covered by the KIIbIH (all located in Africa except for Cambodia, Guatemala, Lao PDR and Paraguay) have time to prepare. Still, they must prioritise formalisation and tax collection. Policy efforts should focus on structuring the pension system, which implies both building adequate non-contributory schemes financed through tax reforms and enrolling informal workers in mandatory contributory schemes through regulatory reforms and incentives.
Group 7: Countries with high informality, at an early stage of the ageing process and with already high tax-to-GDP ratios (quadrants A4 and B2). These countries face a situation similar to Group 6, although they have already managed to increase their fiscal space. For such countries (examples include Bolivia, Honduras and Nicaragua), the time left until they become aged societies should be used to boost formalisation efforts and maximise public spending efficiency.

Figure 3.25. Cross-tabulating informality with ageing and fiscal space can help determine country policy priorities and available options to reform pension systems
Copy link to Figure 3.25. Cross-tabulating informality with ageing and fiscal space can help determine country policy priorities and available options to reform pension systems

Correlations of the informal employment rate with the projected share of individuals aged above 65 years by 2050 (Panel A) and with the tax-to-GDP ratio (Panel B)

Build multi-pillar pension systems that combine non-contributory and contributory schemes

As part of their strategies to achieve the triple goal of coverage, benefit adequacy and fiscal sustainability, and within the policy space available to them, countries should adopt multi-pillar pension systems that integrate and articulate contributory and non-contributory schemes (Figure 3.26). Country evidence shows that strong and adequate pensions systems – i.e. those that protect the income of elderly people and shield them from the risk of old-age poverty – all rely on different tiers that mix non-contributory and contributory mechanisms (OECD, 2023[47]). The design of multi-pillar systems should also incentivise formal employment, encourage the participation of more women in the labour force and their formalisation, boost productivity, and enhance the collection of contributions and taxes.

Figure 3.26. Stylised structure of multi-pillar pension systems
Copy link to Figure 3.26. Stylised structure of multi-pillar pension systems

Non-contributory pension schemes (Pillar 0) should protect from poverty the most vulnerable segments of the population. This typically includes individuals who are not and may never be covered by contributory pension schemes: informal workers with low contributory capacities, rural women, inactive individuals, part-time workers, etc. The level of benefits should be sufficient to operate as a relevant anti-poverty measure and benefits should be universal (to the degree that the fiscal space allows), avoiding means tests. Policymakers should consider using pension tests with a relatively low “taper” rate – i.e. the rate at which non-contributory pension benefits are gradually reduced as contributory pension benefits increase.

Mandatory, contributory pension schemes should continue to constitute the cornerstone of pension systems. Pillar 1 should aim to guarantee decent livelihoods by replacing a substantial portion of lifetime pre-retirement income. Pillar 1 can integrate inter- and intra-generational solidarity mechanisms, especially when structured as DB plans. Pillar 2 should be designed as a complementary and mandatory contributory scheme that smooths consumption during old age. Ideally, Pillars 1 and 2 should combine DB and DC schemes to diversify the risks associated with these pension plans. For instance, if Pillar 1 is structured as a DB or NDC scheme, Pillar 2 should be structured as a DC scheme.

Beyond these three main pillars, other resources can also help support and preserve levels of consumption among the elderly. These include, for instance, additional voluntary pension plans that could allow workers with sufficient contributory capacity to build larger reserves. Importantly, these other forms of old-age income support should complement strong and coherent mandatory pillars, and should not be used as a reason to delay the development of systems that are available to all workers.

Overall, policymakers must ensure citizens’ trust in pension systems and guarantee long-term financial sustainability. For both DB and DC schemes, policymakers should establish and enforce clear rules to adjust the main parameters of the system: level of benefits, contribution requirements and statutory age of retirement. Credible and sustainable DC schemes also require transparent investment governance as well as capital market regulations (OECD, 2022[23]; OECD, 2016[65]).

Adapt pensions systems to the realities of the informal economy

Expanding the coverage of contributory pensions schemes to informal workers requires approaches that leverage and combine multiple policy tools. This includes: i) offering subsidies and matching contributions for low-paid workers to make participation more attractive; ii) authorising flexible and irregular contributions with a monthly or annual minimum; iii) integrating schemes to avoid fragmentation and inefficiency; and iv) engaging aggregators and networks (e.g. cooperatives, online work platforms, mobile money operators, or local post offices) to facilitate enrolment, information dissemination and contribution collection. Additionally, bundling pension products with complementary benefits (such as short-term insurance or credit) can enhance their appeal. Policymakers should also assess the impact of and remove regulatory barriers that prevent individuals from working beyond the statutory retirement age – which allows people to continue building pension density – or while receiving retirement benefits (partial retirement). Finally, it is important to revise or eliminate rules that exclude workers with incomplete contribution histories, ensuring they have the option to retire with a partial or reduced pension.

Overall, the selection and combination of policy tools should be guided by each country’s specific context. Factors such as the extent of informality and the profile of non-contributing workers – notably their contributory capacity – are crucial to identify the most effective policy mix.

Ensure the adequacy of benefits – including over time

Ensuring that non-contributory pension benefits remain adequate over time is crucial for these schemes to fulfil their anti-poverty purpose. Policymakers should incorporate automatic and predictable adjustment mechanisms – such as indexing benefits to price inflation, wage growth or a combination of both – into the design of non-contributory pension systems. These adjustments help maintain the real value of benefits and protect recipients from the effects of rising costs of living.

Contributory pension schemes, especially those structured as DB plans, can also face benefit erosion over time. In these cases, policymakers should implement comparable indexation mechanisms to ensure that benefit levels are preserved in the long term.

The eligibility age for non-contributory pensions often has limited relevance for informal workers, who typically continue working well beyond this age to support themselves in the absence of contributory benefits. For a given level of fiscal resources allocated to non-contributory schemes, setting a low eligibility age can result in a large proportion of the population being covered, including informal workers in employment, but with benefits too modest to adequately support informal workers once they are no longer able to work. To address this, policymakers should consider adjusting the eligibility age based on objective indicators such as life expectancy. Aligning the eligibility age closer to the point at which informal workers are likely to exit the labour force would reduce the number of beneficiaries but also provide higher benefit amounts to those most in need. For contributory schemes, policymakers should ensure that the statutory age of retirement aligns with the internal financial sustainability of schemes.

Guarantee the financial and fiscal sustainability of the pension system via parametric reforms and domestic resource mobilisation

To finance the expansion of universal non-contributory pension schemes (Pillar 0), policymakers need to mobilise additional domestic resources and maximise spending efficiency. Several fiscal reforms can support these dual goals, including: eliminating deductions on personal income tax and lowering excessively high exemption thresholds; reassess corporate tax incentives to ensure they effectively promote investment and formal employment; introducing new excise taxes (e.g. health taxes on tobacco, alcohol and sugar-sweetened beverages); reducing fossil fuel subsidies; and introducing a personalised VAT14 to further broaden the revenue base.

For contributory pension schemes, promoting the formalisation of workers and businesses should remain the primary objective – especially for self-employed workers, who are often excluded by design. However, acknowledging that no single solution will achieve full formalisation, policymakers should also consider alternative and innovative methods to increase contribution collection. One promising approach is the implementation of PTRs (or monotax systems), which consolidate multiple tax obligations – including social security contributions – into a single payment. To be effective, PTRs must be carefully designed, with eligibility criteria and tax structures calibrated to balance incentives for formalisation, maintain the integrity of the standard tax regime and ensure equity among taxpayers. Additionally, policymakers can leverage the growing digitalisation of payments to create voluntary micro-contribution channels at points of sale – the so-called consumption-based pensions. These innovative mechanisms are particularly important as the gig economy expands.

Annex 3.A. Demographic typology

Copy link to Annex 3.A. Demographic typology

The demographic typology used in the chapter is adapted from the global typology developed in Ahmed et al. (2016[3]). It classifies countries into four categories based on demographic characteristics and future development potential. Two main parameters are used to classify countries:

The projected growth of the share of the working-age population over the period 2024-39. In the original typology of Ahmed et al. (2016[3]), the period of reference is 2015-30.
The total fertility rate in 1994 (i.e. 30 years15 before the reference year of 2024, reflecting the approximate length of time that defines a “generation”) and in 2024. In the original typology of Ahmed et al. (2016[3]), these reference years are 1985 and 2015, respectively.

Annex Figure 3.A.1 describes the use of these two parameters to classify countries into the four demographic groups labelled pre-dividend, early-dividend, late-dividend and post-dividend countries.

Annex Figure 3.A.1. Parameters and methodology used to build the demographic typology of countries
Copy link to Annex Figure 3.A.1. Parameters and methodology used to build the demographic typology of countries

Based on this updated typology, KIIbIH countries were classified into four groups. This classification can help identify policy priorities and opportunities for countries at different stages of the demographic transition and its associated dividend.

Pre-dividend countries: In a first group of countries, all located in Africa, the potential window for the first demographic dividend16 – i.e. the automatic increase in the share of working-age individuals relative to the share of dependents induced by the decline of fertility rates – will occur in the future (Annex Table 3.A.1). In these pre-dividend countries, the proportion of the working-age population is expected to grow until 2039. However, current fertility rates – at four or more births per woman – indicate that these countries are still in the midst of the demographic transition and have not yet experienced the decline in child dependency that typically accompanies the first demographic dividend.
Early-dividend countries: In a second group of countries, spanning Africa, Asia and Latin America and the Caribbean, the potential window for the first demographic dividend is ongoing or occurred recently. In early-dividend countries, working-age population shares are expected to grow between 2024 and 2039. However, current fertility rates – below four births per woman – imply that these countries have been progressing through the demographic transition model and will experience rapid reductions in the population share of its youth.
Late-dividend countries: In a third group of countries, potential for the first demographic dividend is currently passing. In these late-dividend countries, working-age population shares are expected to contract between 2024 and 2039. With fertility rates from 1994 (or 30 years ago, reflecting the approximate length of time that defines a “generation”) being at or above replacement levels (2.1 births per woman), it implies that these countries are entering the final phase of the demographic transition and are still able to reap the first demographic dividend.
Post-dividend countries: In the fourth group of countries, potential for the first demographic dividend has already faded. In these post-dividend countries, working-age population shares are expected to contract between 2024 and 2039. Fertility rates in 1994 were already below replacement levels, indicating that these countries are furthest along in their demographic transition and the window to exploit the first demographic dividend has closed. Most of these countries are already or close to becoming superaged societies, with those aged above 65 years accounting for at least 21% of the population.

ISO	Country	Demographic group	Region	Income level	Growth of working-age over 2024-39 (% points)	Total fertility rate in 1994	Total fertility rate in 2024
ALB	Albania	Late-dividend	Europe	Upper-middle	-3.6	2.8	1.3
ARG	Argentina	Early-dividend	Americas	Upper-middle	2.1	2.9	1.5
ARM	Armenia	Post-dividend	Asia	Upper-middle	-2.1	2.1	1.7
BEN	Benin	Pre-dividend	Africa	Lower-middle	4.6	6.4	4.5
BFA	Burkina Faso	Pre-dividend	Africa	Low	7.0	6.8	4.1
BGR	Bulgaria	Post-dividend	Europe	High	-2.7	1.4	1.7
BHS	Bahamas	Late-dividend	Americas	High	-3.7	2.5	1.4
BOL	Bolivia	Early-dividend	Americas	Lower-middle	2.3	4.5	2.5
BRA	Brazil	Late-dividend	Americas	Upper-middle	-2.5	2.6	1.6
BRB	Barbados	Post-dividend	Americas	High	-4.5	1.7	1.7
CHL	Chile	Late-dividend	Americas	High	-2.6	2.4	1.1
CHN	China (People’s Republic of)	Post-dividend	Asia	Upper-middle	-4.7	1.6	1.0
CMR	Cameroon	Pre-dividend	Africa	Lower-middle	4.5	6.0	4.3
COL	Colombia	Late-dividend	Americas	Upper-middle	-2.7	2.9	1.6
CRI	Costa Rica	Late-dividend	Americas	Upper-middle	-2.2	2.9	1.3
CYP	Cyprus	Late-dividend	Asia	High	-2.8	2.2	1.4
DOM	Dominican Republic	Early-dividend	Americas	Upper-middle	0.8	3.2	2.2
ETH	Ethiopia	Early-dividend	Africa	Low	3.9	7.1	3.9
GHA	Ghana	Early-dividend	Africa	Lower-middle	3.2	5.3	3.3
GMB	Gambia	Early-dividend	Africa	Low	6.3	6.0	3.9
GTM	Guatemala	Early-dividend	Americas	Upper-middle	4.8	5.2	2.3
HND	Honduras	Early-dividend	Americas	Lower-middle	2.7	4.9	2.5
HRV	Croatia	Post-dividend	Europe	High	-3.2	1.5	1.5
IDN	Indonesia	Late-dividend	Asia	Lower-middle	-0.6	2.9	2.1
IND	India	Early-dividend	Asia	Lower-middle	0.7	3.7	2.0
JAM	Jamaica	Late-dividend	Americas	Upper-middle	-1.7	2.9	1.4
KEN	Kenya	Early-dividend	Africa	Lower-middle	3.6	5.4	3.2
KHM	Cambodia	Early-dividend	Asia	Lower-middle	2.3	5.3	2.5
LAO	Lao People’s Democratic Republic	Early-dividend	Asia	Lower-middle	3.3	5.6	2.4
LBR	Liberia	Early-dividend	Africa	Low	5.0	6.2	3.9
MDG	Madagascar	Early-dividend	Africa	Low	3.6	6.0	3.9
MDV	Maldives	Early-dividend	Asia	Upper-middle	0.3	4.7	1.6
MEX	Mexico	Late-dividend	Americas	Upper-middle	0.0	3.1	1.9
MLI	Mali	Pre-dividend	Africa	Low	4.8	7.1	5.5
MLT	Malta	Post-dividend	Europe	High	-1.6	1.8	1.1
MMR	Myanmar	Early-dividend	Asia	Lower-middle	0.0	3.2	2.1
MNG	Mongolia	Early-dividend	Asia	Lower-middle	4.8	2.8	2.6
MWI	Malawi	Early-dividend	Africa	Low	5.5	6.5	3.6
NAM	Namibia	Early-dividend	Africa	Upper-middle	5.4	4.7	3.2
NER	Niger	Pre-dividend	Africa	Low	5.9	7.8	5.9
NGA	Nigeria	Pre-dividend	Africa	Lower-middle	5.6	6.3	4.4
NIC	Nicaragua	Early-dividend	Americas	Lower-middle	2.2	3.9	2.2
PER	Peru	Late-dividend	Americas	Upper-middle	-0.2	3.4	2.0
PRY	Paraguay	Early-dividend	Americas	Upper-middle	1.8	4.2	2.4
ROU	Romania	Post-dividend	Europe	High	-3.0	1.4	1.7
RWA	Rwanda	Early-dividend	Africa	Low	4.3	6.4	3.6
SEN	Senegal	Early-dividend	Africa	Lower-middle	3.8	5.9	3.8
SLE	Sierra Leone	Early-dividend	Africa	Low	5.1	6.6	3.7
SLV	El Salvador	Early-dividend	Americas	Upper-middle	1.6	3.6	1.8
SUR	Suriname	Late-dividend	Americas	Upper-middle	-1.1	2.9	2.2
TGO	Togo	Pre-dividend	Africa	Low	3.3	5.7	4.1
THA	Thailand	Post-dividend	Asia	Upper-middle	-6.8	1.9	1.2
TZA	Tanzania	Pre-dividend	Africa	Lower-middle	4.0	5.9	4.5
UGA	Uganda	Pre-dividend	Africa	Low	6.4	6.9	4.2
URY	Uruguay	Late-dividend	Americas	High	-0.1	2.4	1.4
VNM	Viet Nam	Late-dividend	Asia	Lower-middle	-0.8	2.9	1.9
ZAF	South Africa	Early-dividend	Africa	Upper-middle	0.6	3.3	2.2
ZMB	Zambia	Pre-dividend	Africa	Lower-middle	4.4	6.3	4.0

Note: Income levels refer to the income classification developed by the World Bank. Regional classification of countries follows the geographical grouping of the United Nations’ M49 standard.

Source: Adapted from Ahmed et al. (2016[3]).

Statlink

Annex 3.B. Estimating the cost of non-contributory pensions

Copy link to Annex 3.B. Estimating the cost of non-contributory pensions

Estimates of the fiscal cost of an age-based universal benefit, or universal non-contributory pension, without any means-test nor pensions-test, rely on two main parameters: the access age (i.e. the age at which an individual becomes eligible to receive non-contributory benefits) and the level of benefits.

Annex Figure 3.B.1. Scenarios and parameters to estimate the cost of universal non-contributory pensions
Copy link to Annex Figure 3.B.1. Scenarios and parameters to estimate the cost of universal non-contributory pensions

Estimates were produced for six different access age scenarios (Annex Figure 3.B.1).

In Scenario 1 (or the baseline scenario), the access age is set at the current legal age of retirement. Depending on the country, this age can differ for men and women.
In Scenarios 2, 3 and 4, the access age is defined as an absolute age limit.
In Scenarios 5 and 6, the access age is defined relative to life expectancy at age 60. Some studies argue in favour of using alternative measures such as healthy life expectancy. However, these measures can be complex to compute and often rely on a set of underlying indicators the quality of which can be questioned, notably in countries with limited statistical capacities. In this regard, OECD (2023[47]) notes that “available indicators of healthy life expectancy are not suited to determine how retirement ages should evolve. [...] Using the WHO measure for automatic links in pension policies would thus entail a complex procedure requiring more data and entailing a higher risk of errors than using remaining life expectancy, for little gain.” For this reason, the simulation exercise favours the use of life expectancy over healthy life expectancy.

For each access age scenario, the level of benefit was simulated at 50%, 75% and 100% of the median per-capita income, computed as the household total income divided by the number of household members. For countries in which welfare data stem from consumption or expenditure data, these were used as proxy for income and computed as the household total consumption or expenditure divided by the number of household members.

The combination of age and benefit cases results in a total of 18 distinct scenarios (Annex Figure 3.B.1).

Cost estimates, in monetary values, were then expressed as a share of GDP, using GDP data from the IMF’s 2025 World Economic Outlook (IMF, 2025[58]). Annex Table 3.B.1 presents the country-level cost estimates expressed as a share of GDP obtained for the different scenarios.

Access age		Current legal age of retirement			60 years			65 years			70 years			Life expectancy at age 60, minus 10 years			Life expectancy at age 60, minus 5 years
Level of benefits (% of median per-capita income)		50%	75%	100%	50%	75%	100%	50%	75%	100%	50%	75%	100%	50%	75%	100%	50%	75%	100%
ISO	Country	1.1	1.2	1.3	2.1	2.2	2.3	3.1	3.2	3.3	4.1	4.2	4.3	5.1	5.2	5.3	6.1	6.2	6.3
ALB	Albania	1.5	2.3	3.1	1.8	2.7	3.7	1.3	1.9	2.5	0.8	1.3	1.7	0.7	1.1	1.5	0.4	0.6	0.8
ARG	Argentina	0.9	1.3	1.8	1.0	1.5	2.1	0.8	1.1	1.5	0.5	0.8	1.0	0.5	0.8	1.0	0.3	0.5	0.6
ARM	Armenia	2.2	3.3	4.4	2.7	4.0	5.4	1.9	2.8	3.8	1.2	1.8	2.3	1.4	2.1	2.8	0.8	1.3	1.7
BEN	Benin	1.1	1.7	2.3	1.1	1.7	2.3	0.7	1.0	1.4	0.4	0.7	0.9	0.6	0.9	1.1	0.3	0.5	0.7
BFA	Burkina Faso	1.6	2.4	3.1	1.6	2.4	3.1	1.0	1.5	2.1	0.7	1.0	1.4	0.9	1.4	1.9	0.6	0.9	1.2
BGR	Bulgaria	2.1	3.2	4.2	2.4	3.6	4.8	1.9	2.8	3.7	1.3	1.9	2.6	1.6	2.4	3.3	1.1	1.6	2.2
BHS	Bahamas	1.2	1.9	2.5	1.8	2.7	3.6	1.2	1.9	2.5	0.8	1.2	1.6	0.9	1.3	1.8	0.5	0.8	1.1
BOL	Bolivia	3.5	5.3	7.0	3.1	4.7	6.2	2.1	3.2	4.2	1.4	2.0	2.7	2.3	3.4	4.6	1.5	2.2	3.0
BRA	Brazil	1.7	2.5	3.4	2.1	3.2	4.3	1.5	2.2	3.0	1.0	1.5	1.9	1.0	1.5	1.9	0.6	0.9	1.2
BRB	Barbados	1.6	2.3	3.1	2.3	3.5	4.7	1.7	2.6	3.4	1.1	1.7	2.2	0.9	1.4	1.9	0.6	0.9	1.2
CHL	Chile	2.7	4.1	5.5	3.2	4.7	6.3	2.3	3.4	4.5	1.5	2.2	3.0	1.1	1.7	2.2	0.7	1.0	1.3
CHN	China	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
CMR	Cameroon	1.3	1.9	2.6	1.3	1.9	2.6	0.9	1.3	1.7	0.6	0.8	1.1	0.7	1.1	1.5	0.4	0.6	0.8
COL	Colombia	1.7	2.6	3.4	1.6	2.4	3.3	1.1	1.7	2.3	0.7	1.1	1.5	0.7	1.1	1.5	0.4	0.7	0.9
CRI	Costa Rica	3.1	4.6	6.1	3.2	4.9	6.5	2.2	3.4	4.5	1.5	2.2	2.9	1.0	1.5	1.9	0.5	0.8	1.1
CYP	Cyprus	2.9	4.3	5.7	3.9	5.8	7.8	2.9	4.3	5.7	2.0	3.0	4.0	1.5	2.3	3.1	0.8	1.2	1.6
DOM	Dominican Republic	1.4	2.1	2.7	1.4	2.1	2.7	0.9	1.4	1.9	0.6	1.0	1.3	0.6	1.0	1.3	0.4	0.6	0.9
ETH	Ethiopia	1.0	1.5	2.0	1.0	1.5	2.0	0.6	1.0	1.3	0.4	0.6	0.9	0.5	0.7	1.0	0.3	0.4	0.5
GHA	Ghana	1.2	1.8	2.4	1.2	1.8	2.4	0.8	1.2	1.7	0.6	0.9	1.2	0.7	1.0	1.4	0.5	0.7	0.9
GMB	Gambia	1.4	2.2	2.9	1.4	2.2	2.9	0.9	1.4	1.9	0.6	0.8	1.1	0.7	1.0	1.4	0.4	0.6	0.8
GTM	Guatemala	1.4	2.0	2.7	1.4	2.0	2.7	0.9	1.4	1.9	0.6	0.9	1.2	0.8	1.2	1.6	0.5	0.8	1.0
HND	Honduras	2.0	3.0	4.0	2.0	3.0	4.0	1.4	2.1	2.7	0.9	1.4	1.8	1.2	1.7	2.3	0.8	1.1	1.5
HRV	Croatia	3.6	5.3	7.1	4.4	6.7	8.9	3.3	5.0	6.7	2.3	3.4	4.5	2.0	3.1	4.1	1.2	1.8	2.5
IDN	Indonesia	0.4	0.6	0.8	0.3	0.5	0.6	0.2	0.3	0.4	0.1	0.2	0.2	0.2	0.3	0.3	0.1	0.2	0.2
IND	India	1.2	1.8	2.4	1.1	1.7	2.2	0.7	1.1	1.5	0.4	0.7	0.9	0.6	0.9	1.2	0.3	0.5	0.7
JAM	Jamaica	2.2	3.2	4.3	3.1	4.6	6.2	2.2	3.2	4.3	1.5	2.3	3.1	1.9	2.8	3.8	1.3	1.9	2.6
KEN	Kenya	0.8	1.2	1.6	0.8	1.2	1.6	0.6	0.8	1.1	0.4	0.5	0.7	0.6	0.8	1.1	0.4	0.5	0.7
KHM	Cambodia	2.6	3.9	5.2	2.6	3.9	5.2	1.7	2.6	3.4	1.0	1.5	2.1	1.1	1.7	2.3	0.6	1.0	1.3
LAO	Lao PDR	2.2	3.4	4.5	1.8	2.7	3.6	1.2	1.8	2.3	0.8	1.1	1.5	1.0	1.5	2.0	0.6	0.9	1.2
MDG	Madagascar	0.5	0.7	1.0	0.5	0.7	1.0	0.3	0.4	0.6	0.2	0.3	0.4	0.2	0.3	0.5	0.1	0.2	0.3
MDV	Maldives	0.8	1.2	1.5	1.2	1.8	2.4	0.8	1.2	1.5	0.5	0.7	0.9	0.4	0.6	0.8	0.3	0.4	0.5
MEX	Mexico	1.1	1.7	2.2	1.6	2.4	3.2	1.1	1.7	2.2	0.7	1.1	1.4	0.8	1.3	1.7	0.5	0.8	1.1
MLI	Mali	2.0	3.0	4.0	2.0	3.0	4.0	1.3	2.0	2.6	0.8	1.2	1.6	1.2	1.8	2.4	0.7	1.1	1.4
MLT	Malta	2.4	3.7	4.9	3.4	5.1	6.8	2.4	3.7	4.9	1.5	2.3	3.1	0.7	1.1	1.4	0.0	0.0	0.0
MMR	Myanmar	2.1	3.1	4.1	2.1	3.1	4.1	1.3	2.0	2.7	0.9	1.3	1.8	1.1	1.7	2.2	0.7	1.1	1.4
MNG	Mongolia	1.5	2.3	3.1	1.2	1.8	2.4	0.7	1.1	1.4	0.4	0.6	0.8	0.6	0.8	1.1	0.3	0.5	0.7
MWI	Malawi	1.9	2.8	3.7	1.0	1.6	2.1	0.8	1.2	1.6	0.5	0.8	1.1	0.7	1.0	1.4	0.4	0.6	0.8
NAM	Namibia	0.6	1.0	1.3	0.6	1.0	1.3	0.5	0.7	0.9	0.3	0.5	0.6	0.5	0.7	0.9	0.3	0.5	0.6
NER	Niger	1.4	2.1	2.8	1.4	2.1	2.8	0.9	1.3	1.7	0.5	0.8	1.1	0.7	1.1	1.4	0.4	0.6	0.8
NGA	Nigeria	0.3	0.5	0.7	0.3	0.5	0.7	0.2	0.3	0.5	0.1	0.2	0.3	0.2	0.3	0.4	0.1	0.2	0.2
NIC	Nicaragua	2.0	3.0	4.0	2.0	3.0	4.0	1.4	2.1	2.8	1.0	1.5	1.9	1.1	1.7	2.3	0.7	1.1	1.4
PER	Peru	1.2	1.8	2.3	1.7	2.6	3.4	1.2	1.8	2.3	0.7	1.1	1.5	0.8	1.2	1.6	0.5	0.7	1.0
PRY	Paraguay	2.0	3.0	3.9	2.0	3.0	3.9	1.3	2.0	2.7	0.8	1.3	1.7	1.2	1.8	2.4	0.8	1.1	1.5
ROU	Romania	2.6	3.9	5.2	3.1	4.6	6.1	2.4	3.6	4.8	1.6	2.5	3.3	1.9	2.9	3.9	1.2	1.8	2.4
RWA	Rwanda	1.0	1.5	2.0	1.0	1.5	2.0	0.7	1.0	1.3	0.4	0.6	0.8	0.5	0.8	1.1	0.3	0.5	0.7
SEN	Senegal	1.9	2.8	3.7	1.9	2.8	3.7	1.2	1.8	2.4	0.8	1.1	1.5	1.0	1.5	2.0	0.6	0.9	1.1
SLE	Sierra Leone	1.3	2.0	2.6	1.3	2.0	2.6	0.9	1.3	1.8	0.6	0.9	1.1	0.8	1.1	1.5	0.5	0.7	0.9
SLV	El Salvador	3.4	5.0	6.7	2.8	4.3	5.7	2.0	3.1	4.1	1.4	2.1	2.9	1.5	2.3	3.0	1.0	1.5	2.0
SUR	Suriname	2.1	3.2	4.2	2.1	3.2	4.2	1.3	1.9	2.6	0.8	1.2	1.6	1.1	1.6	2.1	0.6	1.0	1.3
TGO	Togo	1.9	2.8	3.8	1.9	2.8	3.8	1.2	1.8	2.4	0.8	1.2	1.6	1.1	1.6	2.1	0.7	1.0	1.4
THA	Thailand	2.5	3.7	4.9	2.5	3.7	4.9	1.7	2.5	3.3	1.0	1.6	2.1	0.7	1.1	1.4	0.4	0.6	0.7
TZA	Tanzania	0.6	0.9	1.1	0.6	0.9	1.1	0.4	0.6	0.8	0.3	0.4	0.5	0.3	0.5	0.7	0.2	0.3	0.4
UGA	Uganda	0.6	0.9	1.2	0.6	0.9	1.2	0.4	0.6	0.8	0.2	0.4	0.5	0.3	0.4	0.5	0.2	0.3	0.3
URY	Uruguay	4.2	6.3	8.3	4.2	6.3	8.3	3.1	4.6	6.1	2.1	3.2	4.2	2.0	2.9	3.9	1.2	1.8	2.4
VNM	Viet Nam	4.0	6.0	8.0	3.2	4.8	6.4	2.1	3.2	4.3	1.4	2.1	2.9	1.2	1.8	2.5	0.8	1.2	1.6
ZAF	South Africa	0.6	0.9	1.2	0.6	0.9	1.2	0.4	0.6	0.8	0.2	0.4	0.5	0.4	0.6	0.8	0.2	0.4	0.5
ZMB	Zambia	0.5	0.7	1.0	0.3	0.5	0.7	0.2	0.4	0.5	0.1	0.2	0.3	0.2	0.4	0.5	0.1	0.2	0.3

Source: Authors’ simulations based on OECD (2026[4]), IMF (2025[58]) and United Nations (2021[43]).

Statlink

References

[3] Ahmed, S. et al. (2016), “Demographic change and development: A global typology”, Policy Research Working Paper, No. 7893, World Bank, Washington DC, http://documents.worldbank.org/curated/en/867951479745020851.

[27] Antolín, P. et al. (2009), “Investment Regulations and Defined Contribution Pensions”, OECD Working Papers on Insurance and Private Pensions, No. 37, OECD Publishing, Paris, https://doi.org/10.1787/222771401034.

[15] Bai, H. and B. Zelko (2022), “La densidad de cotizaciones al IPS en Paraguay”, Informes Técnicos OIT Cono Sur, No. 32, International Labour Organization, Santiago, https://www.ilo.org/sites/default/files/wcmsp5/groups/public/@americas/@ro-lima/@sro-santiago/documents/publication/wcms_857549.pdf.

[70] Barreix, A. et al. (2022), “Revisiting Personalized VAT: A Tool for Fiscal Consolidation with Equity”, IDB Discussion Papers, No. IDB-DP-939, Inter-American Development Bank, https://doi.org/10.18235/0004147.

[51] Brys, B. et al. (2025), “Mobilising domestic resources in low- and middle-income countries: An analytical framework”, OECD Taxation Working Papers, No. 74, OECD Publishing, Paris, https://doi.org/10.1787/be3ffbbb-en.

[59] Calligaro, F. and O. Cetrangolo (2023), “Financing Universal Social Protection: The Relevance and Labour Market Impacts of Social Security Contributions”, WIEGO Working Paper, No. 47, Women in Informal Employment: Globalizing and Organizing (WIEGO) and International Labour Organization (ILO), https://www.wiego.org/wp-content/uploads/2023/03/financing-universal-social-protection_0.pdf.

[21] Chomik, R., P. O’Keefe and J. Piggott (2024), “Pensions in Aging Asia and the Pacific: Policy Insights and Priorities”, ADB Economics Working Paper Series, No. 746, Asian Development Bank, Manila, Philippines, https://doi.org/10.22617/wps240491-2.

[53] Coxhead, I. and C. Grainger (2018), “Fossil Fuel Subsidy Reform in the Developing World: Who Wins, Who Loses, and Why?”, Asian Development Review, Vol. 35/2, pp. 180-203, https://doi.org/10.1162/adev_a_00119.

[46] de Tavernier, W. and H. Boulhol (2021), “Automatic adjustment mechanisms in pension systems”, in Pensions at a Glance 2021: OECD and G20 Indicators, OECD Publishing, Paris, https://doi.org/10.1787/d9c5d58d-en.

[12] Evans, B. and R. Palacios (2015), “An Examination of Elderly Co-residence in the Developing World”, Social Protection & Labor Policy Note, No. 17, World Bank, Washington DC, https://documents1.worldbank.org/curated/en/600291467992481736/pdf/96557-BRI-P153286-PUBLIC-Box391458B-ADD-SERIES-SPL-Policy-Note-17.pdf.

[45] Gentilini, U. et al. (2021), Keep the Pace: How Inflation Erodes Cash Transfers and What to Do About it, World Bank, Washington DC, https://hdl.handle.net/10986/42485.

[22] Giles, J., C. Joubert and T. Tanaka (2025), “Participation in Pension Programs in Low- and Middle-Income Countries”, Policy Research Working Paper, No. 11105, World Bank, Washington DC, http://documents.worldbank.org/curated/en/099305404242537228.

[34] Guven, M. and H. Jain (2023), The Promise of Ejo Heza: A Brighter Future for All Rwandans, World Bank, Washington DC, https://documents.banquemondiale.org/fr/publication/documents-reports/documentdetail/099033123085039385.

[19] Guven, M., H. Jain and C. Joubert (2021), Social Protection for the Informal Economy: Operational Lessons for Developing Countries in Africa and Beyond, World Bank, Washington DC, http://hdl.handle.net/10986/36584.

[40] IHME (2024), Global Burden of Disease 2021: Findings from the GBD 2021 Study, Institute for Health Metrics and Evaluation (IHME), Seattle, https://www.healthdata.org/sites/default/files/2024-05/GBD_2021_Booklet_FINAL_2024.05.16.pdf.

[42] IHME (2022), Global Burden of Disease Study 2021 (GBD 2021), Institute for Health Metrics and Evaluation (IHME), Seattle, https://vizhub.healthdata.org/gbd-results/.

[64] ILO (2025), “Informal employment rate by sex (%) - Annual”, ILOSTAT (database), https://rshiny.ilo.org/dataexplorer08/?lang=en&segment=indicator&id=SDG_0111_SEX_AGE_RT_A (accessed on 1 December 2025).

[5] ILO (2024), World Social Protection Report 2024-26, International Labour Organization, Geneva, https://doi.org/10.54394/zmdk5543.

[56] ILO (2021), “Resolution concerning the second recurrent discussion on social protection (social security)”, 109th Session of the International Labour Conference (Geneva, 7-18 June 2021), No. ILC.109/Resolution III, International Labour Organization, Geneva, https://www.ilo.org/sites/default/files/2025-11/ILC109_Resolution_3.pdf.

[55] ILO (2016), “Financing social protection through taxation of natural resources: Mongolia”, Country Note Series, International Labour Organization, Geneva, https://www.social-protection.org/gimi/Media.action;jsessionid=_x90ezZkS24KGq74RvhsUBn0_SR41THePOEdEpN57a1zrthiJfu5!1358910551?id=15577.

[57] ILO (2012), “Social Protection Floors Recommendation, 2012 (No. 202)”, 101st Session of the International Labour Conference (30 May-15 June 2012), International Labour Organization, Geneva, https://normlex.ilo.org/dyn/nrmlx_en/f?p=NORMLEXPUB:12100:0::NO::P12100_INSTRUMENT_ID:3065524.

[38] ILO (1952), “Social Security (Minimum Standards) Convention, 1952 (No. 102)”, 35th Session of the International Labour Conference (4-28 June 1952), International Labour Organization, Geneva, https://normlex.ilo.org/dyn/nrmlx_en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C102.

[58] IMF (2025), “WEO Data: April 2025 Edition”, World Economic Outlook Database (database), https://www.imf.org/en/Publications/WEO/weo-database/2025/april.

[33] IMF, ILO and the World Bank (2024), “Toward an Inclusive, Equitable, and Sustainable National Pension System in Iraq”, IMF Analytical Note, No. 2024/001, International Monetary Fund, Washington DC, https://www.imf.org/en/Publications/analytical-notes/Issues/2024/04/30/Towards-an-Inclusive-Equitable-and-Sustainable-National-Pension-System-in-Iraq-544930.

[52] IMF, OECD, United Nations and the World Bank (2016), Enhancing the Effectiveness of External Support in Building Tax Capacity in Developing Countries: Prepared for Submission to G20 Finance Ministers, International Monetary Fund, Washington DC, https://www.imf.org/external/np/pp/eng/2016/072016.pdf.

[37] Inzunza, A. and C. Madeira (2025), “The Chilean pension withdrawals and the 2025 reform: Fiscal and retirement consequences”, Journal of Pension Economics and Finance, Vol. 24/4, pp. 516-542, https://doi.org/10.1017/s1474747225000058.

[24] Joubert, C. (2014), “Pension Design with a Large Informal Labour Market: Evidence from Chile”, IZA Discussion Paper, No. 8211, Institute for the Study of Labor (IZA), Bonn, https://docs.iza.org/dp8211.pdf.

[18] Joubert, C. and P. Kanth (2022), “Life Cycle Savings in a High-Informality Setting: Evidence from Pakistan”, Policy Research Working Paper, No. 10121, World Bank, Washington DC, http://hdl.handle.net/10986/37666.

[20] Joubert, C. et al. (2026), “Inclusive Pensions for an Aging World: Evidence and Strategies for Engaging Informal Workers”, Working Paper, Consultative Group to Assist the Poor (CGAP), Washington DC, https://www.cgap.org/research/inclusive-pensions-for-aging-world-evidence-and-strategies-for-engaging-informal-workers.

[44] Kidd, S. and D. Athias (2020), “Hit and Miss: An assessment of targeting effectiveness in social protection with additional analysis”, Working Paper, Development Pathways and Act Church of Sweden, Orpington and Uppsala, https://www.developmentpathways.co.uk/wp-content/uploads/2019/03/Hit-and-miss-long-report-.pdf.

[7] Kudrna, G., J. Piggott and P. Poonpolku (2024), “Sustainable and Equitable Pension Reform for Emerging Economies: An Application to Indonesia”, Working Paper Series, No. 2024/02, ARC Centre of Excellence in Population Ageing Research (CEPAR), Sydney, https://www.cepar.edu.au/publications/working-papers/sustainable-and-equitable-pension-reform-emerging-economies-application-indonesia.

[68] Lee, R. and A. Mason (2006), “What Is the Demographic Dividend?”, Finance and Development, Vol. 43/3, https://www.imf.org/external/pubs/ft/fandd/2006/09/basics.htm.

[62] Mas-Montserrat, M., C. Colin and B. Brys (2024), “The design of presumptive tax regimes in selected countries”, OECD Taxation Working Papers, No. 69, OECD Publishing, Paris, https://doi.org/10.1787/58b6103c-en.

[63] Mas-Montserrat, M. et al. (2023), “The design of presumptive tax regimes”, OECD Taxation Working Papers, No. 59, OECD Publishing, Paris, https://doi.org/10.1787/141239bb-en.

[69] Mason, A. (2005), “Demographic transition and demographic dividends in developed and developing countries”, in United Nations Expert Group Meeting on Social and Economic Implications of Changing Population Age Structures, United Nations, Mexico City, https://www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/unpd_egm_200508_09_full_report.pdf.

[4] OECD (2026), “Key Indicators of Informality based on Individuals and their Households (KIIbIH)”, OECD International Development Statistics (database), https://data-explorer.oecd.org/ (accessed on 1 May 2026).

[32] OECD (2025), Financement de la protection sociale au Sénégal : Le rôle de la fiscalité générale, des cotisations sociales et de la contribution du secteur informel, OECD Publishing, Paris, https://doi.org/10.1787/00d67508-fr.

[61] OECD (2025), Financing Social Protection through General Tax Revenues, Social Security Contributions and Formalisation in Thailand, OECD Publishing, Paris, https://doi.org/10.1787/b5cc1a43-en.

[50] OECD (2025), “Global Revenue Statistics - Comparative tax revenues”, Global Revenue Statistics Database (database), https://data-explorer.oecd.org/s/3mw.

[10] OECD (2025), Pensions at a Glance 2025: OECD and G20 Indicators, OECD Publishing, Paris, https://doi.org/10.1787/e40274c1-en.

[60] OECD (2024), Breaking the Vicious Circles of Informal Employment and Low-Paying Work, OECD Publishing, Paris, https://doi.org/10.1787/f95c5a74-en.

[31] OECD (2024), “Extending social protection to informal workers”, in Breaking the Vicious Circles of Informal Employment and Low-Paying Work, OECD Publishing, Paris, https://doi.org/10.1787/f95c5a74-en.

[17] OECD (2024), OECD Pensions Outlook 2024: Improving Asset-backed Pensions for Better Retirement Outcomes and More Resilient Pension Systems, OECD Publishing, Paris, https://doi.org/10.1787/51510909-en.

[8] OECD (2024), Pensions at a Glance Asia/Pacific 2024, OECD Publishing, Paris, https://doi.org/10.1787/d4146d12-en.

[14] OECD (2024), SIGI 2024 Regional Report for Southeast Asia: Time to Care, Social Institutions and Gender Index, OECD Publishing, Paris, https://doi.org/10.1787/7fc15e1c-en.

[47] OECD (2023), Pensions at a Glance 2023: OECD and G20 Indicators, OECD Publishing, Paris, https://doi.org/10.1787/678055dd-en.

[11] OECD (2023), SIGI 2023 Global Report: Gender Equality in Times of Crisis, Social Institutions and Gender Index, OECD Publishing, Paris, https://doi.org/10.1787/4607b7c7-en.

[23] OECD (2022), “Recommendation of the Council for the Good Design of Defined Contribution Pension Plans”, OECD Legal Instruments, No. OECD/LEGAL/0467, OECD Publishing, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0467.

[28] OECD (2020), “Civil service pension reform in developing countries: Experiences and lessons”, OECD Development Policy Papers, No. 33, OECD Publishing, Paris, https://doi.org/10.1787/f872f328-en.

[67] OECD (2019), “Architecture of national pension systems”, in Pensions at a Glance 2019: OECD and G20 Indicators, OECD Publishing, Paris, https://doi.org/10.1787/66b65776-en.

[30] OECD (2018), Financial Incentives and Retirement Savings, OECD Publishing, Paris, https://doi.org/10.1787/9789264306929-en.

[65] OECD (2016), “Recommendation of the Council on Core Principles of Private Pension Regulation”, OECD Legal Instruments, No. OECD/LEGAL/0429, OECD Publishing, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0429.

[66] OECD/WHO (2022), “Ageing”, in Health at a Glance: Asia/Pacific 2022: Measuring Progress Towards Universal Health Coverage, OECD Publishing, Paris, https://doi.org/10.1787/f7ebe745-en.

[16] Ohnsorge, F. and S. Yu (eds.) (2022), The Long Shadow of Informality: Challenges and Policies, World Bank, Washington DC, https://doi.org/10.1596/978-1-4648-1753-3.

[13] O’Keefe, P. and G. Rongen (2025), “Should Malaysia expand its social pension? Global evidence, design issues and options”, Social Protection & Labor Discussion Paper, No. 2605, World Bank, Washington DC, https://hdl.handle.net/10986/43850.

[48] Price, W. (2019), “Developing Coherent Pension Systems: Design Issues for Private Pension Supplements to NDC Schemes”, Social Protection and Jobs Discussion Paper, No. 1929, World Bank, Washington DC, http://documents.worldbank.org/curated/en/148951556875885451.

[25] Price, W. (2018), “Designing Pension Systems with Coherent Funded Private Pillars Including Issues for Notional Defined Contribution Schemes”, Policy Research Working Paper, No. 8420, World Bank, Washington DC, http://hdl.handle.net/10986/29757.

[9] Priebe, J. (2017), “Old‐age Poverty in Indonesia: Measurement Issues and Living Arrangements”, Development and Change, Vol. 48/6, pp. 1362-1385, https://doi.org/10.1111/dech.12340.

[6] Reyes Hartley, G. and M. Abels (2025), “Riding the Demographic Wave Pensions and Retirement Income in an Aging World: The State of Social Protection Report 2025 Background Paper #5”, Social Protection Discussion Paper, No. 2512, World Bank, Washington DC, http://hdl.handle.net/10986/43042.

[36] Rupper Bulmer, E., H. Winkler and H. Mote (2017), “Expanding Social Insurance Coverage to Informal Workers”, Jobs Working Paper, No. 6, World Bank, Washington DC, http://hdl.handle.net/10986/27932.

[26] Rusconi, R. (2008), “National Annuity Markets: Features and Implications”, OECD Working Papers on Insurance and Private Pensions, No. 24, OECD Publishing, Paris, https://doi.org/10.1787/240211858078.

[29] Tanzi, V. and H. Zee (2001), “Tax Policy for Developing Countries”, Economic Issues, Vol. 27, https://doi.org/10.5089/9781589060203.051.

[49] Ugarte, G. and R. Vergara (2022), “PGU and Replacement Rates in the Chilean Pension System”, Economía y Políticas Públicas, No. 633, Centro de Estudios Públicos (CEP), Chile, https://static.cepchile.cl/uploads/cepchile/2022/12/pder633_ugarte_vergara.pdf.

[54] UNICEF (2025), Pathways to the Sustainable Financing of Social Protection in Eastern and Southern Africa, United Nations Children’s Fund (UNICEF), Nairobi, Kenya, https://www.unicef.org/esa/media/16931/file/UNICEF-Pathways-Sustainable-Financing-Social-Protection-ESA-2025.pdf.

[2] United Nations (2024), World Population Prospects 2024, United Nations, Department of Economic and Social Affairs, Population Division, https://population.un.org/wpp/.

[1] United Nations (2023), World Social Report 2023: Leaving No One Behind in an Ageing World, United Nations, Department of Economic and Social Affairs, New York, https://doi.org/10.18356/9789210019682.

[43] United Nations (2021), “Life expectancy E(x) - complete”, United Nations Data Portal, Population Division (database), https://population.un.org/dataportal/.

[39] United Nations (2018), Report on the World Social Situation 2018: Promoting Inclusion Through Social Protection, United Nations, Department of Economic and Social Affairs, New York, https://doi.org/10.18356/5ef37a49-en.

[41] Vollset, S. et al. (2024), “Burden of disease scenarios for 204 countries and territories, 2022–2050: a forecasting analysis for the Global Burden of Disease Study 2021”, The Lancet, Vol. 403/10440, pp. 2204-2256, https://doi.org/10.1016/s0140-6736(24)00685-8.

[35] World Bank (2024), Making Pension Savings Easy and Efficient for Informal Sector Workers - Learning from Kenya’s Haba Haba Pilot, World Bank, Washington DC, http://documents.worldbank.org/curated/en/099051524110034843.

Notes

Copy link to Notes

← 1. “Ageing society” refers to countries in which the share of individuals aged above 65 years accounts for 7% to 14% of the total population. “Aged society” refers to countries in which this group accounts for 15% to 20% of the total population. “Superaged society” refers to countries in which this group accounts for more than 21% of the total population (OECD/WHO, 2022[66]).

← 2. Argentina, the Bahamas, Bolivia, Brazil, Chile, Mexico and Suriname.

← 3. For both overall and contributory pension coverage gaps, the same patterns are found when using deciles of household welfare instead of quintiles. Larger overall pension coverage and contributory pension coverage are found among elderly people who belong to the richest deciles of the household welfare distribution.

← 4. KIIbIH data rely on the harmonisation of data collected by national statistical offices through household surveys. Data and rates may differ from other official sources such as the ILO, which usually rely on administrative records (e.g. the Social Security Inquiry questionnaire to monitor SDG Target 1.3, i.e. the proportion of persons effectively covered by a social protection system, including social protection floors). See also Box 3.1.

← 5. In China, the share of current contributors is computed by aggregating contributions to various pension schemes of diverse natures. While most are purely contributory schemes, some important schemes in terms of size and coverage (e.g. the New Rural Pension Insurance scheme) consist of a membership fee that opens eligibility rights for a basic social pension funded via general revenue. Despite having an element akin to a contribution (fee), this is not a traditional contributory scheme whereby benefits are tied to employment as well as the level and density of contributions. Rather it consists of the acquisition of future rights to benefit from a non-contributory pension.

← 6. Bulgaria, China, Costa Rica, Croatia, Cyprus, Malta, Mongolia and Peru.

← 7. For China, the large share of informal workers contributing to pension schemes stems from the fact that the country integrates informal workers who contribute to schemes whereby the worker pays a membership fee that opens eligibility rights for a basic social pension, funded via general revenue. This is not a traditional contributory scheme; rather, it consists of the acquisition of future rights to benefit from a non-contributory pension.

← 8. Earnings categories classify workers based on their labour earnings into three categories, defined relative to median earnings: low wage quality groups individuals ranging from the bottom of the earnings distribution to 50% of median earnings; medium wage quality groups individuals ranging from 50% to 150% of median earnings; and high wage quality groups individuals above 150% of median earnings.

← 9. Notionally defined contribution (NDC) schemes are “pay-as-you-go” schemes with individual accounts that apply a notional rate of return to contributions made (mimicking DC plans). The accounts are “notional” in that the balances exist only on the books of the managing institution. At retirement, the accumulated notional capital is converted into a monthly pension using a formula based on life expectancy and other actuarial factors (OECD, 2019[67]).

← 10. Annuity markets are financial vehicles that convert accumulated retirement savings into an income stream appropriate to protect against these risks. For instance, it can consist in annuities that are purchased at retirement, converting a lump sum into a lifetime income stream (Rusconi, 2008[26]).

← 11. Vesting rules are the conditions that determine when pension rights become permanent for a contributor and can no longer be forfeited. They specify how long an individual must participate in the scheme before becoming entitled to employer-financed pension benefits. Two common types of vesting are: cliff vesting, where full entitlement is acquired at a specific point in time, or gradual (graded) vesting, where entitlement builds up progressively over several years of participation.

← 12. Depending on data availability, median per-capita consumption was used to proxy income in certain countries.

← 13. Burkina Faso, Cambodia, Cameroon, Ghana, Guatemala, Kenya, Lao PDR, Madagascar, Malawi, Mali, Namibia, Niger, Nigeria, Paraguay, Rwanda, Senegal, Sierra Leone, Togo, Uganda and Zambia.

← 14. A personalised VAT is a proposed mechanism designed to address the so called “impossible trinity of consumption taxes”. It includes a broad base, a single rate and targeted relief for those most in need. In practice, this approach combines three elements: i) universal application, implying a general tax base; ii) a single VAT rate achieved by eliminating exemptions and reduced rates for specific goods; and iii) targeted transfers to the poorest population deciles, in the form of self-financed tax refunds, to offset the resulting increase in tax burden. See Barreix et al. (2022[70]) for further details.

← 15. Thirty years describes the approximate length of time that defines a “generation” – from the birth of a parent to the birth of their child – although the exact span may vary by country and across time.

← 16. The first demographic dividend refers to the dividend mechanically induced by a decline in fertility that leads to a substantial and sustained, but ultimately transitory, increase in the share of working-age individuals relative to dependents, and a growth of output per capita. Although more complex to estimate, the second demographic dividend refers to the potential increase in saving rates and accumulation of human and physical capital as the population ages and life expectancy increases, which can further stimulate output per capita on a permanent or self‐sustaining basis (Lee and Mason, 2006[68]; Mason, 2005[69]).

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Countries

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Publications

Publications

Featured publications

Data

Data

Featured data

News & events

News & events

Featured events

About OECD

About

Engage with us

Work with us

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment