OLD-AGE INCOME SECURITY IN ASEAN MEMBER STATESPOLICY TRENDS, CHALLENGES AND OPPORTUNITIES215OLD-AGE INCOME SECURITY IN ASEAN MEMBER STATES POLICY TRENDS, CHALLENGES AND OPPORTUNITIESiThe Association of Southeast Asian Nations (ASEAN) was established on 8 August 1967. The Member States are Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Philippines, Singapore, Thailand and Viet Nam. The ASEAN Secretariat is based in Jakarta, Indonesia. For inquiries, contact: The ASEAN Secretariat Community Relations Division (CRD) 70A Jalan Sisingamangaraja Jakarta 12110, Indonesia Phone: (62 21) 724-3372, 726-2991 Fax: (62 21) 739-8234, 724-3504 E-mail: public@ASEAN.org Catalogue-in-Publication Data Old-age Income Security in ASEAN Member States - Policy Trends, Challenges and Opportunities Jakarta, ASEAN Secretariat, June 2020 331.252095 1. ASEAN – Pensions – Policy 2. Retirement – Pension FundsISBN 978-602-5798-75-7ASEAN: A Community of Opportunities for All ASEAN Secretariat owns the copyright of the original English version of this publication. The text of this publication may be freely quoted or reprinted, provided proper acknowledgement is given and a copy containing the reprinted material is sent to the Community Relations Division (CRD) of the ASEAN Secretariat, Jakarta. General information on ASEAN appears online at the ASEAN Website: www.ASEAN.org Copyright Association of Southeast Asian Nations (ASEAN) 2020. All rights reserved. iiFOREWORD BY MINISTER OF MINISTRY OF LABOUR, INVALIDS AND SOCIAL AFFAIRS OF VIET NAMAs forecasted, ASEAN will enter the stage of aging population in 2030 with about 72 million elderly people and as estimated, this number will reach 124 million by 2050. In response to the impacts of aging society, together with the aim of expressing strong commitment in social protection, in recent years ASEAN Member States (AMS) have adopted a number of important Declarations such as: Kuala Lumpur Declaration on Aging: Empowering Older Persons in ASEAN; ASEAN Declaration on Strengthening Social Protection and Action Plan to implement this Declaration. I highly appreciate the efforts of relevant sectoral bodies in promoting the implementation of the initiatives to realize the ASEAN Leaders’ Declarations. Among these efforts, the report “Old-age income security in ASEAN Member States - Policy trends, challenges and opportunities” is indeed a remarkable achievement of the labor sector. We all acknowledge that in the transition to an aging population in each country, should the country not be well-prepared and synchronized, the impacts on the social welfare and social protection for the people in general and the elderly in particular will be tremendous. The fact has shown that the health and the working ability to work of the elderly are declining while the issues related to social protection, especially old-age pension for this particular group have not been up to expectation. Over the past few decades, some AMS have introduced old-age pension regimes to ensure livelihoods for the elderly and provide income security for some vulnerable groups. Besides the aforementioned achievements, ASEAN is still facing many difficulties and challenges, with regard to resources and experience, as well as strategies to create an effective pension system. The access to pension is solely for a limited number of subjects, and currently many older people have to rely on family support. Furthermore, ensuring the financial sustainability of the pension system is also a challenge for the social security system of each country. The report “Oldage income security in ASEAN Member States - Policy trends, challenges and opportunities” provides an overview of the status of the social security system in each member country, especially for the elderly, the challenges and policy gaps in such system, and on the ground of which, proposes recommendations on the policy. I hope with these recommendations, the Ministers, policy makers and stakeholders in ASEAN could issue policies according to the each socio-economic condition of each nation, the management mechanisms that proactively address the issues related to the pension system in a long term. On this occasion, I would like to express my appreciation for the cooperation and participation of the AMS, the efficient coordination of ASEAN Secretariat and technical support from International Labor Organization (ILO) to be with the Ministry of Labor, Invalids and Social Affairs of Viet Nam in developing this report. I strongly hope that ASEAN Senior Labor Officials will continue their efforts to implement the recommendations of the report towards promoting social protection system, building an inclusive and sharing society, empowering labor workers in general and older persons in particular in the region.DAO NGOC DUNG Minister Ministry of Labour, Invalids and Social Affairs of Viet NamiiiFOREWORD BY THE SECRETARY- GENERAL OF ASEANWe are witnessing an unprecedented demographic transition , with the global proportion of older people increasing dramatically. In the ASEAN region, around 64 million older persons in 2017 are projected to rise 2.6 times by 2050, reaching 168 million.i Similarly, the old-age dependency rate in the region is also expected to jump from 8.8 in 2015 to 23 .9 in 2050.ii In light of this, countries with ageing societies need to raise the responsiveness of their social protection, which includes pension schemes . Reviewing and strengthening policies related to income security of older persons is therefore critical in achieving ASEAN ‘s broader goal of enhancing the well-being of all people . Therefore, I am pleased to present this Study Report on Old-Age Income Security in ASEAN Member States - Policy Trends, Challenges and Opportunities , which is relevant and timely as ASEAN puts a premium on significantly expanding social security systems and ensuring benefit adequacy that meet international benchmarks. I commend the ASEAN Senior Labour Officials Meeting (SLOM) for leading this initiative under the coordination of the Ministry of Labour, Invalid and Social Affairs ofVietNam in cooperation with the ASEAN Secretariat and International Labour Organization (ILO), and supported by the ASEAN Development Fund. I hope that the study’s analysis and recommendations would contribute to the development of effective and inclusive old-age pension systems within ASEAN Member States . I encourage relevant stakeholders to consider the study’s proposed mechanisms in building an inclusive ASEAN community that promotes high quality of life and equitable access to opportunities for all.i United Nations, Department on Economic and Social Affairs, Population Division, World Population Prospects: The 2017 revision ii United Nations, Department on Economic and Social Affairs, Population Division, World Population Prospects: The 2017 revisionivFOREWORD BY ASSISTANT DIRECTOR- GENERAL AND REGIONAL DIRECTOR OF ILO REGIONAL OFFICE FOR ASIA AND THE PACIFICThere is growing attention to the significance of ageing populations in ASEAN Member States. Recent national and regional public debates have discussed the possible impacts of this structural change and the measures that countries can and should put in place.It is undeniable that the population in the ASEAN Community as a whole is ageing. For some countries, such as Singapore, Thailand and VietNam, the pace is extremely fast. But even countries with a large youth population, like Cambodia and the Lao People’s Democratic Republic, are rapidly ageing.These developments are expected to put families and individuals under extraordinary pressure, making ageing a challenge to societies as a whole. It will further be a challenge to public finances. It is clear that the demographic transition towards an ageing society has deep social, economic and political implications. Understandably, policy makers are concerned about public health costs and the fiscal wellbeing of pension systems. What’s more, if not properly addressed, ageing might also negatively affect a country’s overall productivity.In the midst of this demographic trend, citizens in ASEAN Member States are increasingly calling on their governments to assume greater responsibility in providing old-age income security. Individuals are also more open to the idea of contributing to a public pension system.This being said, the reality is that old-age pensions in many ASEAN countries are still only accessible to a minority of people. Among ASEAN Member States, the proportion of older persons receiving pensions varies significantly from country to country. Moreover, the gender gap in old-age pension coverage highlights further inequalities within countries.In this particular context, policy makers face opposing pressures. One relates to the urgency to expand coverage and increase benefits, while the other has to do with tackling long-term financial sustainability.Guaranteeing old-age income security takes on additional importance when viewed in the light of the 2030 Agenda for Sustainable Development and its 17 Sustainable Development Goals (SDGs). In particular, SDG target 1.3 calls for national social protection systems for all, including social protectionfloors.To guarantee that no older person is left behind, policy-makers and decision-makers should consider comprehensive and integrated public pension systems based on the principle of universality. Indeed, the development of an inclusive old-age pension system can address several issues and have a multiplier effect beyond ending poverty in all its forms everywhere (SDG 1). It can contribute considerably to supporting gender equality and the empowerment of women (SDG 5) and helping to reduce inequality within and among countries (SDG 10).With this in mind, the ASEAN Secretariat and VietNam’s Ministry of Labour, Invalids and Social Affairs contacted the ILO to collaborate on preparing this report. The ILO welcomes this initiative, especially since the findings of the report will feed into national and regional debates to help ASEAN Member States make evidence-based decisions on old-age pension policy.The report points to the importance and urgency of implementing measures to reduce the old-age pension coverage gap, increase benefits and improve the financial sustainability of pension systems. As a whole, the report can contribute to more comprehensive national approaches to effective policy making for old-age income security.Finally, the recent COVID-19 pandemic has reinforced the importance of social protection as a mechanism to deal with the economic and public health crisis. The pandemic has shown that countries equipped with inclusive social protection schemes, including pensions, are better positioned to deal with the impacts of the crisis, particularly on the most vulnerable, including older people.CHIHOKO ASADA-MIYAKAWA Assistant Director-General and Regional Director ILO Regional Office for Asia and the PacificvACKNOWLEDGEMENTSThis Study Report on Old-Age Income Security in ASEAN Member States: Policy trends, Challenges and Opportunities was planned under the Senior Labour Officials Meeting’s Working Group on Progressive Labour Practices to Enhance the Competitiveness of ASEAN (SLOM-WG) Work Plan 2016-2020 under the leadership of the Ministry of Labour, Invalids and Social Affairs (MOLISA) of Viet Nam with the support of the ASEAN Secretariat, ASEANJapan Fund for Building Social Safety Nets and International Labour Organization (ILO).The study report was produced through a series of ASEAN workshops and consultations with ASEAN Member States for validation including through the convening of the ILO-ASEAN Seminar on the Future of Social Protection in ASEAN on 4-5 July 2019 in Bangkok, Thailand. Participating were tripartite representatives and social security agencies from ASEAN Member States, Ministry of Health, Labour and Welfare of Japan, ASEAN Secretariat, ILO, ASEAN Trade Union Council, and ASEAN Confederation of Employers. Subsequently, the Study Report was finalised and ad-referendum endorsed by SLOMWG on 9 March 2020. The study report also took into account the outcomes of the Workshop on Sustaining Financing Mechanism for Social Insurance including Social Pension which was held on 21-22 September 2017 in Ho Chi Minh, Viet Nam with the support of the ASEAN-Japan Fund for Building Social Safety Nets.MOLISA would like to express our sincerest gratitude to the following individuals for their technical expertise and knowledge in making this study worthwhile.To the focal points, all of whom we can not acknowledge individually, of labour ministries of ASEAN Member States for your valuable information and insights;To the ASEAN Secretariat, particularly Ms. Mega Irena, Ms. Pitchanuch Supavanich (former Senior Officer), Ms. Madyah Rahmi Lukri, and Ms. Sarah Choirinnisa for your valuable inputs and support throughout the process;To the study co-authors, Mr. Wolfgang Scholz (deceased), Research Associate, University Bonn-RheinSieg and Mr. Nuno Cunha, Senior Social Protection Specialist, ILO Decent Work Team Bangkok;To Ms. Loveleen Dee (ILO Lao PDR); Ms. Sara Elder (Regional Economic and Social Analysis Unit of ILO Regional Office for Asia and Pacific); Mr. Markus Ruck (Decent Work Team for Asia and Pacific); Ms. Christina Behrendt (SOCPRO Geneva); Mr. Fabio Duran Valverde (SOCPRO Geneva); Mr. Andres Acuna (SOCPRO Geneva); Ms. Quynh Anh Nguyen (SOCPRO Geneva); Mr. Charles Knox-Vydmanov (Independent Consultant); Mr. Florian Juergens (Help Age International); Ms. Vanessa Steinmayer (UN ESCAP); Ms. Kalaivani Karunanethy (Independent Consultant); Mr. Andi Sanusi (Independent Expert); Ms. Laura Schmitz (Independent Consultant); and Ms. Silas Theile (Independent Consultant);Lastly, MOLISA would also like to express our sympathies with the untimely passing of Mr. Wolfang Scholz whose contribution to the study has been significant.In memory of Wolfgang, who is a professor, a unique economist, a public policy expert, a historian, a man of the world. It was a unique privilege and honour to collaborate with you in this Report. Thanks for all you taught me. I will always take those lessons with me. ~ Nuno Cunha (Co-Author)viTABLE OF CONTENTSFOREWORD BY MINISTER OF MOLISA VIET NAMIIIFOREWORD BY SECRETARY-GENERAL OF ASEANIVFOREWORD BY ASSISTANT DIRECTOR-GENERAL AND REGIONAL DIRECTOR OF ILOREGIONAL OFFICE FOR ASIA AND THE PACIFICVACKNOWLEDGEMENTSVIBOXES, FIGURES AND TABLESIX1. INTRODUCTION12. PENSIONS: OBJECTIVES, TRADITIONS AND SOCIOECONOMIC IMPLICATIONS72.1 Pension objectives and traditions..........................................................................................................................................82.2 Contemporary pension systems...........................................................................................................................................12A generic overview and socioeconomic implications...............................................................................................12Old-age pension financing options and macroeconomic considerations.......................................................18Old-age pensions and the informal sector.....................................................................................................................263. OLD-AGE PENSION PROBLEMS AND POLICY TRENDS IN THE ASEAN REGION353.1 Overview of challenges affecting ASEAN countries....................................................................................................363.2 Old-age pension systems under construction in ASEAN countries: Achievements, deficiencies and incompatibilities......................................................................................................44Provident funds...........................................................................................................................................................................51Asia: Aged before rich? The issue of pension affordability.......................................................................................553.3 Social protection and pension trends in the ASEAN region.....................................................................................574. PENSION POLICY RECOMMENDATIONS FOR ASEAN MEMBER STATES714.1 General considerations on pension policy-making for ASEAN...............................................................................72vii4.2 Recommended priority areas per country and steps of coverage and adequacy reform..........................784.3 Widening the fiscal space for social security, and the demands for fiscal space of other programmes.......... 99Pension cost estimation........................................................................................................................................................1005. CONCLUSIONS AND RECOMMENDATIONS105Conclusions........................................................................................................................................................................................105Recommendations..........................................................................................................................................................................107REFERENCES110ANNEX I STATISTICAL TABLES122ANNEX II COUNTRY BRIEFS125COUNTRY BRIEF: BRUNEI DARUSSALAM................................................................................................................................125COUNTRY BRIEF: CAMBODIA......................................................................................................................................................133COUNTRY BRIEF: INDONESIA......................................................................................................................................................141COUNTRY BRIEF: LAO PEOPLE’S DEMOCRATIC REPUBLIC..............................................................................................149COUNTRY BRIEF: MALAYSIA.........................................................................................................................................................157COUNTRY BRIEF: MYANMAR........................................................................................................................................................165COUNTRY BRIEF: PHILIPPINES.....................................................................................................................................................174COUNTRY BRIEF: SINGAPORE......................................................................................................................................................185COUNTRY BRIEF: THAILAND........................................................................................................................................................196COUNTRY BRIEF: VIET NAM..........................................................................................................................................................204viiiBOXES, FIGURES AND TABLESBOXES Box 1. Old-Age Allowance in Thailand.............................................................................................................. 29 Box 2. Lack of enforcement and compliance as a major barrier to expanding coverage..............33 Box 3. Statistical development in ASEAN Member States and social security policies...................41 Box 4. Pension coverage........................................................................................................................................46 Box 5. Central Provident Fund of Singapore...................................................................................................55 Box 6. ASEAN Declaration on Strengthening Social Protection: A milestone....................................58 Box 7. Brunei Darussalam’s old-age flat-rate demogrant: An ASEAN experience with universal old-age social pensions............................................................................................................60 Box 8. Singapore’s Silver Support Scheme: Tax-financed and means-tested old-age income supplement................................................................................................................................................................ 61 Box 9. Low protection against longevity risks in the Malaysian Employee’s Provident Fund.......64 Box 10. Addressing the foremost challenges in the Cambodian old-age income security agenda: Cost projections for providing a universal floor of old-age protection................................................80 Box 11. Fiscal implications for the adoption of an old-age demogrant in Indonesia......................82 Box 12. Policy dialogue with the Lao People’s Democratic Republic: Two scenarios of cost projections for an old-age demogrant and other recommendations..................................................84 Box 13. Myanmar: Recommendations for increasing old-age pension coverage............................87 Box 14. Social pensions: Estimated costs of increasing old-age coverage in the Philippines......88 Box 15. Expanding coverage to self-employed workers in the Philippines........................................90 Box 16. Improving financial sustainablity in Thailand’s old-age pension system............................. 93 Box 17. Multitier old-age pension system in Viet Nam...............................................................................95 Box 18. Simulated impacts and costs of reforming the social pension in Viet Nam........................ 98FIGURES Figure 1. A: ILO multi-pillar pension model....................................................................................................12 Figure 2. Typical developments of alternative types of contribution rates......................................... 23 Figure 3. Conceptual framework: Informal employment..........................................................................27 Figure 4. Composition of informal employment (including agriculture), by employment status, latest available year (%)..........................................................................................................................................37ixFigure 5. Population aged 60 years or older, by region, 1980–2050 (thousands).............................38 Figure 6. Percentage of population older than the statutory pensionable age in Asia and the Pacific receiving an old-age pension, latest available year.............................................44 Figure 7. Overview of old-age pension schemes in ASEAN Member States, by type of scheme, 2015 or latest available year..........................................................................................48 Figure 8. GDP growth rates in ASEAN Member States, 2008–17 (%) ....................................................58 Figure 9. Social protection expenditure, including health care, in ASEAN Member States, latest available year (% of GDP)......................................................................................................................................62 Figure 10. Distribution of Cambodian workers contributing to the National Social Security Fund, 2015............................................................................................................................................................................... 80 Figure 11. Cost of a monthly universal pension in Cambodia, 2018–2117 (% of GDP)...................81 Figure 12. Costs of providing a universal old-age pension in Indonesia to persons aged 60 or 65 or older, 2014–62 (% of GDP).................................................................................................................................................83 Figure 13. Cost projections for two scenarios of old-age universal social pensions in the Lao People’s Democratic Republic, 2017–30 (% of GDP)............84 Figure 14. Projection of costs of a pension-tested demogrant in Myanmar for persons aged 60 or older and 65 or older, 2017–97 (as % of GDP)..................................................87 Figure 15. Scenarios of costs projections of the social pension in the Philippines, 2017–28 (% of GDP).............................................................................................................................................................................. 89 Figure 16. Tentative cost estimates of a non-contributory pension in Singapore............................ 91 Figure 17. Projection of Thailand’s Social Security Office reserves levels, 2014–54 (baht).......... ..94 Figure 18. Reduction in poverty rates at the national level and among social pension recipients...98TABLES Table 1. Types of contributions according to financing method and their funding level.............. 22 Table 2. Informality concepts...............................................................................................................................27 Table 3. Share of workers in vulnerable employment in the ASEAN Member States......................36 Table 4. Informal employment in ASEAN and subregions as a share of total employment, by sex, 2016............................................................................................................................................................................... 37 Table 5. Number and percentage of persons aged 60 or older in ASEAN Member States, 2017 and 2050............................................................................................................................................................................... 39 Table 6. Fertility rates in ASEAN Member States since the 1950s...........................................................40 xTable 7. Gini coefficients in ASEAN Member States and selected countries, latest available year .42 Table 8. Population below the national poverty line in ASEAN Member States, 2006–15 (% of total population).................................................................................................................................................................. 43 Table 9. Active contributors to an old-age pension scheme in the ASEAN Member States (as share of working-age population and labour force), by sex, latest available year.....................47 Table 10. Old-age pensions beneficiaries in ASEAN Member States, latest available year (as % of reference population)............................................................................................................................49 Table 11. Public social protection expenditure, by branch in the ASEAN Member States, latest available year (% of GDP)..........................................................................................................................53 Table 12. General social and economic aspects of ASEAN Member States.........................................59 Table 13. Singaporean labour force, by citizenship status, 2010–17 (thousands)............................92 Table 14. Proposal for a multi-tier old-age pension system in Viet Nam.............................................. 96 Table 15. Estimates for a comprehensive pension package under full coverage of older persons in 2050 (as % of GDP), and current fiscal space allocated on old-age pensions in ASEAN Member States (latest available year).............................................................................................................................. 101 Table 16. Estimated cost of the first full year of implementation of universal tax-financed old-age pension in ASEAN Member States (% of GDP)............................................................................................ 102 Table A1. Scope of social security legislation in the ASEAN Member States................................... 122 Table A2. Features of old-age pensions programmes and legal coverage....................................... 123 Table A3. Overview of pension-related features in ASEAN.................................................................... 124 Table CB1. Key socioeconomic and demographic datafor Brunei Darussalam.............................. 127 Table CB2. Labour market and pension indicatorsfor Brunei Darussalam, latest available year.128 Table CB3. Pension scheme parameters for Brunei Darussalam.......................................................... 131 Table CB1. Socioeconomic and demographic datafor Cambodia....................................................... 135 Table CB2. Labour market and pension indicatorsfor Cambodia, latest available year............... 136 Table CB3. Pension scheme parameters for Cambodia........................................................................... 139 Table CB1. Socioeconomic and demographic data for Indonesia....................................................... 143 Table CB2. Labour market and pension indicatorsfor Indonesia, latest available year............... 144 Table CB3. Pension scheme parameters for Indonesia............................................................................ 147 Table CB1. Socio-economic and demographic dataforthe Lao People’s Democratic Republic.151xiTable CB2. Labour market and pension indicatorsforthe Lao People’s Democratic Republic, latest available year.............................................................................................................................................................................152 Table CB3. Pension schemes parameters forthe Lao People’s Democratic Republic................... 155 Table CB1. Socioeconomic and demographic data for Malaysia......................................................... 159 Table CB2. Labour market and pension indicatorsfor Malaysia, latest available year.................. 160 Table CB3. Pension schemes parameters for Malaysia............................................................................ 163 Table CB1. Socioeconomic and demographic data for Myanmar....................................................... 167 Table CB2. Labour market and pension indicatorsfor Myanmar, latest available year................. 168 Table CB3. Pension scheme parameters for Myanmar............................................................................. 171 Table CB1. Socioeconomic and demographic data forthe Philippines............................................. 177 Table CB2. Labour market and pension indicatorsforthe Philippines, latest available year...... 178 Table CB3. Pension scheme parameters for the Philippines.................................................................. 182 Table CB1. Socioeconomic and demographic data for Singapore...................................................... 188 Table CB2. Labour market and pension indicatorsfor Singapore, latest available year............... 189 Table CB3. Pension scheme parameters for Singapore........................................................................... 193 Table CB1. Socioeconomic and demographic data for Thailand......................................................... 198 Table CB2. Labour market and pension indicatorsfor Thailand, latest available year.................. 199 Table CB3. Pension schemes parameters for Thailand............................................................................. 202 Table CB1. Socioeconomic and demographic data for Viet Nam........................................................ 206 Table CB2. Labour market and pension indicatorsfor Viet Nam, latest available year................. 207 Table CB3. Pension scheme parameters for Viet Nam.............................................................................. 211xii1. INTRODUCTION1One of humanity’s major achievements is the general rise in life expectancy, which, together with other societal changes,has generated a global ageing-population trend. The world is experiencing an unprecedented demographic transformation and, by 2050, the number of persons aged 60 years or older will increase from 600 million now to almost 2 billion, and the proportion of them will double, from 12.7percent to 21.3 percent (table 5).This positive development, summarized and discussed under the overarching term “ageing”, is no longer a phenomenon exclusive to developed countries. The population in Asia as a whole is ageing, with old-age dependency ratios (the ratio of people aged 65 or older to those aged 15–64) projected to increase. A demographic transition consisting of “fewer babies and longer lives” took around a century to transpire in Europe and North America, whereas in Asia, it has occurred or is currently occurring in different countries in a single generation, or less than 30 years (OECD, 2013). For some countries, the pace is extremely fast, as in China, Thailand and Viet Nam. Even countries with a large youth population, like Cambodia and the Lao People’s Democratic Republic, will have aged by the end of the century. Around one third of their populations will be older than 60 by 2100, with this share of older persons growing five times in Cambodia and six times in the Lao People’s Democratic Republic (DESA Population Division, 2017b).1 This scenario has public health cost implications for the fiscal balance of pension systems and, possibly (if not properly addressed), on countries’ productivity levels.Contrary to what has taken place in other regions, such as Europe and North America, many Asian countries, including Member States of the Association of Southeast Asian Nations (ASEAN), might become old before becoming rich, a configuration that will mean additional challenges. Also, due to their longer life expectancies, women will form the majority of older persons, which will decrease the sex ratio – the proportion of men in relation to women.In the past, caring for older relatives rested upon the family’s many children or intrinsically upon the support of co-resident family members. That also is changing. Half a century ago, a woman in Cambodia or the Philippines would, on average, give birth to nearly seven children. Now the number is down to three. Fertility rates will continue to decrease (table 6). A young couple just starting their working career in China will possibly have four older parents as dependants (Zhang and Goza, 2006). Circumstances such as these put families under extraordinary pressure, making ageing a challenge to societies as a whole as well as for public finance. The demographic transition towards an ageing society in ASEAN Member States has deep social, economic and political implications.Despite their importance, old-age pensions are only accessible for a minority of people in many countries. While effective old-age pension coverage is 68 percent globally, it is only 55.2 percent in Asia and the Pacific (ILO, 2017d). In South-Eastern Asia, the percentage is even smaller, at only 31.5 percent of the population older than the statutory pensionable age receiving an old-age pension (ILO, 2017d). Among the ASEAN Member States, the proportion of older persons receiving a pension varies significantly from country to country, ranging from 81.7 percent in Brunei Darussalam and 83 percent in Thailand to roughly 40 percent in the Philippines and Viet Nam and substantially less in1 In 2015, the share of persons aged 60 or older in Cambodia and the Lao People’s Democratic Republic was 6.8 percent and 6.1 percent, respectively. By 2100, these shares are projected to be 32.4 percent and 33.9 percent (DESA Population Division, 2017b).2Cambodia, Indonesia and the Lao People’s Democratic Republic. The gender gap in old-age pension coverage in the Philippines, where 53.2 percent of older men and only 29 percent of older women are covered, highlights further inequalities within countries in the receipt of old-age pensions.In many cases, including in countries with extensive population coverage, old-age pensions in payment do not allow recipients to lead independent lives but serve only as subsidies to the cost of living that is assumed to be covered by other forms of income, namely intra-family transfers, casual or part-time work, personal savings, community support and social assistance. The main reasons for old-age pensions to assume such an “accessory” role include the continued high informality of labour markets and the generally low government investment and engagement in redistributive public provisions.Many older persons rely on family support only where traditional family support structures are still intact. According to HelpAge International (2013), children are still the major source of income for 37 percent of older persons in Thailand and 32 percent in Viet Nam (Knodel et al., 2015). Income from work still represents the major source of income for 45 percent of older persons in the Philippines, 33 percent in Thailand and 29 percent in Viet Nam (Jackson and Peter, 2015). This illustrates a challenging situation wherein, despite the changing demographic and social patterns, including urbanization, pensions still have a limited role in guaranteeing income security in old age for many ASEAN countries.While this is the context in the ASEAN region, globally, many high-income countries undertaking fiscal consolidation are reforming their pension system to achieve cost reductions, including by raising the retirement age, reducing benefits and increasing contribution rates. These adjustments aim to reduce government (tax) transfers and labour costs and are aligned with reduced state responsibility for guaranteeing income security in old age as they shift large parts of the economic risks associated with pension provision onto individuals. Not only do these adjustments undermine the importance of adequacy of pension benefit levels but they also reduce the role of public pension in preventing poverty in old age. In this regard, national pension systems must be designed such that they provide for adequate benefits, guaranteed at least at the minimum level stipulated in the International Labour Organization (ILO) Conventions related to old-age pensions: the Social Security (Minimum Standards) Convention, 1952 (No. 102) and the Invalidity, Old-Age and Survivors’Benefits Convention, 1967 (No. 128).2These measures have influenced the pension debate everywhere in the world. In the wider context of overall sustainability and stability of public and private finance, the old-age pension debate in ASEAN countries brings together conflicting elements of fiscal consolidation and ageing with the challenges of extending pension coverage and ensuring benefit adequacy while attracting increased attention of stakeholders.2 ILO Convention No. 102 sets out that pension benefits should provide at least 40 percent of insured pre-retirement income to those who have contributed for at least 30 years and a reduced or adjusted benefit to those who have contributed for at least 15 years. ILO Convention No. 128 provides higher standards for pensions.3The fears of the impact of ageing on pension system sustainability have triggered a search for solutions to the policy conflicts. Proposals gaining attraction in the Asia-Pacific region (as in other regions) are grounded in:1. the reduction of government liability, which assumes such a move can stand the test of political reality;2. and an increasing role for individualized market solutions, which assumes markets cannot fail, an assumption that is challenged not only by macroeconomic theory but also, and repeatedly, by empirical observation.This debate is not new. After more than 30 years of national and international deliberations, all arguments, be they theoretical-technical, empirical, economic or political, have probably been tried and tested. In earlier debates, it was assumed that changing the method of pension financing, from pay-as-you-go to funding, could circumvent the financing problem. More sophisticated proposals, including full funding or the privatization and institutionalized competition among privatized pension providers, were made, some of them even portrayed as part of a strategy to contribute to the stimulation of economic growth. From more recent debates, it is known that there is no method of avoiding the costs of ageing. In fact, all debate and policy decisions are about how to allocate, in the fairest manner, the costs of ageing among all stakeholders. In this debate, however, the term “fair” is open to interpretation, which is driven by the interests of those involved.While that 30-year debate took place, policy happened. It was mainly in the 1980s that alternatives to traditional pay-as-you-go defined-benefit pensions started gaining influence. This occurred in parallel with a growing prominence of policies that began dismantling the welfare state in favour of privatization of public benefits and services. Following Chile’s privatization of its public pension system in 1981, mandatory defined-contribution schemes were introduced in several countries in the 1990s and 2000s, particularly in Latin America and Eastern Europe. The justification used by the proponents of this policy and paradigm shifts was that it would contribute to guaranteeing the financial equilibrium of the old-age income security system and avoid future increases of costs and contribution rates, notably in the face of ageing societies.In Latin America, 18 countries undertook this type of reform in their old-age pension systems. Some Eastern European countries, such as Bulgaria, the Czech Republic, Hungary and Poland, also adopted defined-contribution elements into their pension schemes. In Western Europe, Germany made a major step in the same direction while maintaining the core principles, but reducing in relative terms the financial volume of its established pay-as-you-go system. Other Western European countries, like the Netherlands, had such concepts already implemented. Similar moves were undertaken in the United States. In all these countries, however, the reforms soon showed their social and fiscal limitations, which resulted in many of the countries adopting policies to reform the previous reforms.3In Eastern and South-Eastern Asian countries, old-age income policy was not (or only marginally) affected by the overall debate. Actually, the region is a public pension policy laggard, which, due to the high speed of its societies’ ageing, is now forced to undergo a compressed phase of “accelerated pension policy”. The good news is that the region can learn from the lessons of the previous3 Following the 2008 global financial crisis, countries began recognizing the risks of defined-contribution schemes with regard to their exposure to investment failures, excessive management costs, limited adequacy of benefits, the limited risk pooling among scheme members and the huge transition cost involved in switching from defined-benefit to defined-contribution schemes. As a result, some of the countries that had switched from defined-benefit to defined-contribution schemes have reversed their previous decisions others have begun intensifying their previous decision – both with open-ended results.4experiences, including the policies pursued or avoided in parallel in other regions.Two very different approaches exist with regards to the generation of old-age income in ASEAN Member States. On the one hand, there are individualized systems based on national public provident funds in Brunei Darussalam, Malaysia and Singapore. On the other hand, there are pooled redistributive defined-benefit systems in the Lao People’s Democratic Republic, the Philippines, Thailand and Viet Nam. Some countries, such as Indonesia, combine the two types. More and more countries have started complementing their earnings-related schemes with non-contributory (tax funded) components, sometimes in the form of demogrants (such as in Brunei Darussalam and Thailand). A demogrant is a grant paid to a group of people based on demographic principles, such as age or sex.Although a few countries are “pension advanced”, most are lagging. In none of the ASEAN Member States have pension schemes reached the same importance in securing old-age incomes as they have in other regional contexts, particularly Europe.The analysis presented in this report identifies a growing recognition of the importance of pensions by governments and populations of the ASEAN Member States. Surveys have demonstrated people’s rising expectations for their government to assume greater responsibility in providing for their old-age income security as well as people’s individual openness to contribute to a public pension system (Jackson and Peter, 2015). At the same time, Member States are working on expanding effective coverage of their existing schemes and launching or conducting policy discussions on new schemes to reach excluded people. All of this indicates a new awareness of public pensions in ASEAN Member States.The current shape of pension systems in the ASEAN region and their role stand in contrast to what the 2008 global financial crisis disclosed: a strong demand for “precautionary savings” across all income groups. This increased demand can be explained by the personal insecurities ignited by, first, the 1997–98 Asian financial and economic crises (when long-term employment growth was suddenly interrupted by employment stagnation or even a decline of up to 5 percent and by a significant restructuring of employment, for example, a strong increase in the number of ownaccount workers, which had been on decline (World Bank, n.d.), with no income security mechanism in place) and then the 2008 Great Financial Crisis. Given this experience, it is not surprising that in a recent International Monetary Fund survey, the majority of the population in selected countries of Asia indicated they would like to rely more on a publicly organized national pension system (IMF,2017).The analysis of old-age security presented here gains additional importance when looked at in the light of the 2030 Agenda for Sustainable Development and its 17 Sustainable Development Goals (SDGs), particularly considering that SDG target 1.3 calls for national social protection systems for all, including social protection floors. To guarantee that no older person is left behind, policy-makers and decision-makers should consider a comprehensive social protection system based on the principle of universality. Equally critical is the Social Protection Floors Recommendation (No. 202), which was adopted unanimously by ILO constituents in 2012 and which calls for the combination of contributory public pensions with non-contributory pension schemes to protect a country’s whole population.The development of an inclusive old-age pension system can address several issues and have a multiplier impact beyond SDG 1 (ending poverty in all its forms everywhere). It can contribute considerably to SDG 5 (supporting gender equality and the empowerment of women) and SDG 105(helping to reduce inequality within and among countries). Income security in old age is essential for achieving all the goals and will contribute to the fundamental commitment to end poverty in all forms. This report was prepared in parallel with the development of a forthcoming report, Old-Age Pensions in Asia, which extends the narrative to China, Fiji, India, the Islamic Republic of Iran, the Republic of Korea, Japan, Mongolia, Sri Lanka and Timor-Leste. Reference to it is made in the discussion here. In addition, recommendations made throughout this report are of a general policy-orienting nature and do not replace the need for detailed country analyses.4 In short, Old-Age Pensions in ASEAN Member States– in its focused narrative – is a stepping stone in supporting countries to make evidence-based decisions towards old-age pension policy and avoid losing time making the same mistakes that have been made elsewhere. At the same time, it points to the importance and urgency of implementing measures to reduce the old-age pension coverage gap, increase benefit adequacy and improve the financial sustainability of pension systems, thus contributing to a comprehensive overall approach towards effective old-age income security policies.4 In many ASEAN Member States, the ILO has undertaken technical analyses on the pension systems with resulting technical recommendations, some more recently, others earlier. These should be consulted in case concrete reform policy advice is needed. Some of the recommendations made here have been extracted from ILO in-country studies.62.PENSIONSOBJECTIVES, TRADITIONS AND SOCIOECONOMIC IMPLICATIONS7This chapter introduces and defines a few general concepts related to social security policies for older persons. It aims to guide policy actors through policy problems related to oldage pension schemes that are common among ASEAN Member States. It addresses the objectives of old-age income security policies that are nowadays generally accepted and put into practice. And it contextualizes old-age pension systems of modern societies and economies as well as discusses implications related to the need for such policies for inclusive growth and socioeconomic development. It overviews generic pension traditions and recalls the essential financing options for universal pension systems. To conclude, it explores the political economy of governments’ attempts worldwide to individualize the “ageing risk”. More on these discussions can be found in the report Old-Age Pensions in Asia.For the sake of clarity, we call a pensioner someone who receives a pension, either as a regular payment or as a lump sum.5 Conditional on citizenship or a labour contract and usually on other parameters, a person can receive such a payment from public or private sources if they have retired from the labour market or due to some exemption. A pensioner is typically an older person, but there are system-specific exemptions.6 In principle, pensioners can receive any pension amount, from very low to very high, including chief executives’ pensions from “pay-packages”. This report distinguishes a pensioner from a “rentier”7, whose money comes from their own or others’investments and who is therefore independent of any age condition or other work contract-related condition. The focus of this report is on “social” or “state-provided” pensions. In other words, the attention is on retirement income, based on public rules, in which workers, employers, the government and civil society or their representatives have a stake jointly or singly.2.1 Pension objectives and traditionsThe purpose of an old-age pension system is to guarantee a person’s regular receipt of income during their “old age”, where old age is defined by societal convention.8 There are different approaches and political attitudes concerning the level of the pension amount to be paid as of retirement age, as discussed here.5 Empirically, schemes exist that pay out lump sums at retirement, either as sole payment or in combination with an annuity. Among the most prominent are the provident funds. Normatively speaking, pay-outs at retirement in the form of a lump sum (solely or in combination with periodical payments) are critical from a social policy perspective as well as prudence of public finance because they are in conflict with recipients’ myopia and financial illiteracy. See the discussion further in this report on provident funds (Singapore; Malaysia and other countries).Provident funds are currently in a process of reducing possibilities of withdrawing lump sums at retirement or for reasons other than retirement. Quite rightly, the ILO normative framework does not consider lump-sum pay-outs a pension; rather “the [old-age] benefit shall be a periodical payment” (ILO Convention No. 102 and No. 128; ILO, 1952).6 See the section on pension types further on. 7 The term often has a disapproving meaning (“rentier class”). 8 This purpose of a pension must be distinguished from the objective of a pension system, which can be defined ashaving “(f)or individuals and government … the core objectives of consumption smoothing, insurance, income redistribution and poverty relief. Indeed (and ideally), assuming mature systems and broad middle classes, pensions are a means of individuals’ consumption smoothing over the life cycle: during a “long” period of active age, the individual “sets aside” resources (either through individual savings or through formal social security pension contributions or income tax payments), which they consume during a “short” remaining lifetime at old-age. If the parameters of this arrangement are reasonably set, one can assume that the individual’s ability to consume pro rata temporis varies only little over the full life cycle (Barr and Diamond, 2010). Here we focus less on system objectives and more on objectives of individual pension design from the point of view of the recipient.8Guaranteeing a standard of livingThe pension amount paid is designed such that it guarantees a person’s standard of living, where the guaranteed standard can be the individual’s standard of living achieved in active age or it can be a standard of living guaranteed for all members of society.This kind of pension amount must be defined and is thus known as a defined-benefit pension, which means that, by law, the benefit at retirement is defined by way of a mathematical formula and, when in payment, must be indexed. If an individual’s standard of living at the time of retirement is to be preserved over the pensioner’s remaining lifetime, indexation must be related to the Consumer Price Index (CPI) or some other relevant price index. If an individual’s relative position within the distribution of wages and salaries is to be preserved, then indexation with gross wages is to be applied.A defined-contribution pension is not calculated on the basis of a formula, meaning that it is not designed to guarantee an individual’s standard of living. Only retrospectively, or expost, can a defined-contribution pension “accidentally” have safe guarded a pensioner’s standard of living.If the aim of a pension payment is to guarantee an individual’s standard of living, the definition of “standard of living” is crucial. Usually, the definition relates to an individual’s wages earned during their active life (before retirement) or another reference period. In this case, the pension amount can be a percentage of the average wages earned by the individual over the reference period,9 as chosen, and there is no general rule as to which reference period should be applied for defining an individual’s standard of living. In absence of external shocks, the whole active life period might appear adequate; but in reality, long reference periods are associated with increased probability that the pension claimant will have been exposed to external influences not under their control, which could have a negative impact on the pension entitlement. Therefore, for schemes that adopt long reference periods, wages earned must at least be indexed (upgraded) to the wage at the point of retirement to account for productivity growth and wage inflation. Using shorter reference periods, such as a few years before retirement, might be unfair in the situation of adverse salary scales; for instance, when wages peak during a person’s prime working age but decline before retirement.The percentage to be applied to the reference wage to calculate the pension, also known as the replacement rate, depends on societal convention.10 In formulating this convention, it must be decided whether the pension is considered to be net or gross. If it is net, the pensioner is not supposed to pay income tax on the pension received. Also, the pensioner must be insured without contributory obligations under health and old-age care. If it is gross, the pension entitlement includes the pensioner’s individual obligation to pay income tax as well as health insurance and care insurance; and the pension amount paid must be accordingly higher than under a net pension. In setting the replacement rate, it is usually argued that older persons retired from the labour market need fewer resources for their daily life than during their active life.Pensions designed to guarantee a standard of living can fail to be socially effective when an individual’s low reference earnings fail to produce a pension level (according to the formula) that is above the poverty line.9 ILO Convention No. 102 prescribes that the pension or income replacement to be received at a certain age should amount to 40 percent of the individual’s reference earnings after 30 years of contributions paid into a contributory scheme, or 20 years of residency in a non-contributory scheme (ILO, 1952).10 As described during the introduction, ILO Conventions No. 102 and No. 128 offer international guidelines on minimum standards in relation to replacement rates.9Guaranteeing freedom from want11This pension is typically designed as a pension that is equal for all. It is a flat amount that guarantees “freedom from want” for all citizens. This means the pension amount is at least above the poverty level, in a narrow definition or it is well above that level, in case a more generous definition of freedom from want is applied. While the type of pension guarantees a standard of living (as discussed in the previous section), the level of this pension must be regularly redefined, in line with the changing needs and perceptions of society. As with pensions that guarantee a standard of living, the defined pension level must consider income tax legislation and any obligation to pay contributions to health and care insurance.This pension is redistributive because it implies potentially high replacement rates for low-income earners and low replacement rates for earners of high primary incomes during active life.Guaranteeing no povertyPension benefits are equivalent to a basket of basic goods and services that is defined by society through legislation. The basket can contain the needs of a standard household, including rent, minimum consumption and socio-cultural needs, or it can relate to the specific needs of a retired household or person. Also, it can distinguish between households residing in different urban and/or rural areas. The value of the basket must regularly be recalculated, based on surveys carried out by the official statistical institution, and the pension-in-payment must be adjusted accordingly.Pensions that“guarantee no poverty”are the narrowest possible interpretation of the freedom-fromwant guarantee. This pension often comes in the form of social assistance provisions, which are paid when the guarantee-a-standard-of-living type of pension fails.12Guaranteeing a subsidy to other, imputed, retirement income This pension, although possibly applying defined-benefit system principles and/or formula, assumes that older households have access to other (imputed) non-pension income and thus the regular pension paid needs to be of a subsidiary nature only. It is neither supposed to cover the living standard nor any minimum household needs as intended by the legislation. The “other” incomes to which this pension is only a subsidy include installments taken from private savings, labour income of continued work after retirement, intra-family transfers in cash and/or in-kind and social assistance. In some situations, governments have implemented pensions of this subsidiary character as a first step, intending to expand benefit levels later, once economic growth, increasing fiscal space and public readiness for providing the required financial means allow them to do so.No guaranteeThis type of pension results from the defined-contribution approach, meaning that the pension level paid at the moment of retirement, as well as all following annuities, depend on the performance of financial markets. In other words, the pension paid can accidentally assume a level that guarantees the standard of living – or not – and assessments to verify whether any guarantees are being secured are possible only expost (after the pensioner dies). Indexation of annuities is possible, but only at the expense of the initial payment level, where the level from which indexation starts is usually actuarially reduced.11 For origins of the term, see the section2.2\. 12 Some developing countries have launched universal pensions, but despite being universal, their low value leavesthem far from being socially effective. Most of them fall in the category of “subsidy to imputed other retirement income”.10Types of pensionsThe previous different pension guarantees apply in the case of lifetime old-age pensions (initiated at the moment of retirement). They can be subject to rules that allow persons to retire early or late (before or after the legal retirement age). Under all approaches, pensions can be actuarially decreased in the event of early retirement and increased for late retirement. Nowadays, it is standard to apply such adjustments.Pensions can be paid independent of any age condition in the event of invalidity, which is a term that usually applies in a labour market context. If a person is “invalid”for work as a result of an accident or sickness – either of which can but need not be work-related, they receive an invalidity pension. The invalidity pension is, by nature, a temporary pension because it is paid as long as the person under consideration is deemed invalid. If rehabilitation measures are successfully applied, a person might lose their invalidity status, in which case the payment of the pension ceases, and the person is expected to return to their prior work or take up new work. If no rehabilitation measures are applied or if they are not successful, then the invalidity pension is paid until standard retirement age, when the invalidity pension is replaced by the standard old-age pension, which may or may not differ in the amount paid.Different from invalidity pensions are pensions that are paid for general disability. Such disability can occur at any point of an individual’s life as a result of birth, an accident, a disease or any other external impact. Measures are usually applied to help disabled persons access the labour market,such as in “protected factories”,13 through medical or other rehabilitation, including specialized training and education, and support through adequate technical means. Disability pensions cease or are reduced, according to established rules, when recipients start earning their own income.The ILO calls the transitory pension payments for widows, widowers and orphans“survivor pensions”. A widow’s or widower’s pension derives from the pension of the deceased person. Typically, the pension amount is defined as a percentage of the most recent pension for that person. Various rules can apply if the survivor remarries or is young, such that they can be expected to take up employment. Survivor’s pensions for a child who loses one or both parents are often paid as a partial or full orphan’s pension, until the child reaches a defined age, often 16 or 18, but in some cases, 25 or 27.14Participation in pension schemes is usually mandated. Once a person fulfills certain criteria based on individual characteristics, they are subject to pension legislation. If participation in a pension scheme is optional or voluntary and a person has opted in, then opting out is usually not possible or only possible under restricted rules.In advanced economies, voluntary participation in pension schemes is typically allowed only if those schemes function as additional pillars in the national pension system. Attempts to base national pension schemes only, or at large, on voluntary participation have been disappointing, even if heavily subsidized by the government. Nevertheless, voluntary participation is a reality for people in informal employment, as with the self-employed in certain countries covered by this report, namely Malaysia, Thailand and Viet Nam.13 T he term “protected factory” signifies an enterprise that takes special account of the needs of employees with disabilities. Usually, the terms under which such enterprises operate are defined by (social) legislation. Maximizing profits is no core operational goal but, nevertheless, such factories – depending on the products produced and on the level of qualification of the employees – must usually stand the test of market competition. Factories are often fostered by NGOs or local communities; sometimes they are partially subsidized. Many forms of such factories exist.14 Widow’s, widower’s and orphans’ pensions are called (not only) in ILO terminology “survivor pensions”.11With regards to adequacy of old-age pension benefits, international minimum rules for fixing the level and other parameters of pensions can be found in ILO Conventions No. 102 and No. 128.Further discussion is included in sections 2.2 and 3.3. 2.2 Contemporary pension systems A generic overview and socioeconomic implications When looking at the current debates and reality of pension systems, one finds various policy elements that stand alone or in combination with others and constitute a “pension system”. The elements encompass: (i) non-contributory social assistance and tax-financed minimum pensions, including demogrants; (ii) mandatory contribution-linked defined-benefit schemes, including occupational pensions; (iii) defined-contribution schemes, such as individual savings accounts, including provident funds, and voluntary supplementary schemes, as well as private or personal pensions; and (iv) non-financial formal and informal family support and private or public support for housing, health and other arrangements contributing to the functioning of society. Pension systems differ from each other in the elements included in a scheme and in the weight these elements are given. They may also differ in other characteristics, such as whether the schemes included are publicly or privately managed. For example, a generic “Bismarckian” pension system typically consists of a (means-tested) public social assistance scheme plus a mandated public contribution-based defined-benefit scheme. A generic “Beveridgean” scheme focuses on the government providing a flat tax- or contribution-financed pension that is sufficient to guarantee freedom from want. By theoretical construction, it would need no social assistance scheme and would leave any additional arrangements of precautionary savings to the individuals, according to their savings capacity and choice. To better describe their respective perspective, while capturing the complexity of pension system design, organizations like the ILO and the World Bank have developed explanatory frameworks based on the concept of a multi-pillar system (figure 1). Figure 1. A: ILO multi-pillar pension modelSource: ILO, 2018c.12B: World Bank five-pillar framework0 pillar A non-contributory social assistance that could be means-tested or take the form of a demogrant.1st pillar A mandatory contribution-linked pension scheme; this may include an additional Pillar 1A, consisting of mandated occupational pensions.2nd pillar A mandatory defined-contribution scheme, such as an individual savings account with flexible options; this pillar includes provident funds.3rd pillar A voluntary additional pillar that could take many forms and be supplementary to Pillars 1 and 2; this pillar would include private (or so-called personal) pensions.4th pillar Non-financial formal and informal support, such as family support, public support for housing and health, etc.Source: Holzmann et al., 2005, slightly amendedThe approach to the contributory schemes is one of the fundamental differences between the two models, in which the ILO understands its state- or semi-state-run Pillar 1 as the societal “anchor institution” while the World Bank has clear lineage towards including more (financial) market solutions.Pillar 0 is relatively similar in both models. It can be both effectively universal (demogrant) or legally universal (means-tested). When means-tested, it should aim at providing a minimum level of income to avoid poverty, including in situations in which a person retiring has failed to accumulate entitlements under the other pillars. For the ILO, universality of coverage can be achieved through a universal non-contributory scheme or by a combination of social insurance and a means-tested or pension-tested scheme.Pillar 1 in both frameworks are mandatory for all persons in formal employment and mostly designed around defined benefits and financed on a pay-as-you-go basis. Although it may be subject to a contribution collection ceiling, it may also be redistributive with the removal of this ceiling and inclusion of a minimum and/or maximum pension benefit.The main difference is found in Pillar 2. Within the ILO normative framework, Pillars 0 and 1 are the fundamental components of any social security system (ILO, 2018c). Pillar 2 has only a complementary function because it can be either mandatory or voluntary and is seen only as a supplementing benefit to the previous two pillars. In addition, Pillar 2 for the ILO can be either a defined-benefit system or a scheme based on accumulated contributions (defined contribution). Within the World Bank framework, the Pillar 2 defined-contribution plan is mandatory, which makes it a central element of the system.Pillar 3 in both systems represents a voluntary component. Both frameworks include individual savings for retirement and are more flexible and discretionary in nature (World Bank, n.d.). However, while the World Bank framework includes employment-sponsored (defined-benefit or definedcontribution) schemes under this pillar, the ILO includes them under its Pillar 2.When the positions of the two organizations are assessed in more detail, the respective multi-pillar system varies in reality and in its normative meaning by the weights placed on the different pillars. While the ILO regards Pillars 2 and 3 as additions to Pillars 0 and 1 (which are the basis of the system), the World Bank framework opens the option to look at Pillars 2 and 3 as alternatives to Pillar 1.13Indeed, within the World Bank framework, Pillar 2 is understood either as an addition to Pillar 1 or as an alternative, such as in the form of a matching contribution system in the informal labour markets. Pillar 3, ideally complements Pillar 2, but in reality, is often seen as an alternative to Pillar 2. Pillar 4, of relatively high relevance for Asian contexts, takes into account schemes in which families still have an important role in the provision of old-age income and may contain benefits and services that are provided by care insurance or similar schemes.It must be stressed that for the World Bank’s framework to function without friction, a number of hidden and outspoken preconditions are required, which include: high labour market formality; wide, deep and transparent financial markets; good governance in public as well as private business administration; and a taxation system consistently reflecting the justified interests of pension scheme stakeholders. In the end, it is the different levels of fulfillment of these preconditions attached to the World Bank framework that lead to considerations regarding different pension system solutions when it comes to country situations and pension policy advice.For the ILO, its multi-pillar pension framework only contains the possible elements of a pension system based on its tried-and-tested standards (Conventions No. 102 and No. 128). Its content does not stand as a norm but as elements of a basket of possibilities, and their applicability and adequacy must be carefully explored in each and every case.Putting aside the discussion of the World Bank and ILO multi-pillar models, social insurance pension systems have always been understood as closely connected with social assistance and ultimately follow Bismarckian principles.15 Their objective is maintenance of the relative income position in old age for members of a scheme, and pension entitlements are strongly linked to earnings-related contributions paid throughout the working life. For people who did not contribute at all or not enough to have a sufficient liveable pension, a means-tested minimum pension would be provided in a Bismarckian system. France, Germany and Portugal are countries that maintain this type of pension system, which are administered by state, semi-state and self-governed institutions.In countries that already have a multi-pillar pension system, the State may provide a flat,universal, minimum pension with the aim of preventing poverty among the retired older population. In that system, individuals rely more on privately provided, occupational-based or personal pensions. Countries that have this type of pension system include Switzerland and the United Kingdom but also, and with a different “colouring”, countries like Japan, Thailand,Republic of Korea and, lately, China.Benefits provided by the different pillars can vary in terms of the amount. Levels of Pillar 0 social assistance pensions are broadly linked to national, regional or local poverty lines.16 In classical Bismarckian social insurance systems, or defined-benefit schemes, the benefit paid at retirement is defined by a specific pension formula that factors contribution payment density, level of wages and other parameters. One of the underlying principles of the defined-benefit schemes is the concept of “actuarial fairness”, which basically holds that “what you pay-in, you get-out”. This concept has many facets, and its strict sense has been repeatedly violated in many mature Bismarckian systems, such as whenever redistributive measures (softening the actuarial fairness principle) were legislated, including special benefit regulations for invalids in cases of early retirement or when years of childrearing are acknowledged in the pension formula.15 See the later parts of section 2.2. 16 This report does not cover related technical and other issues.14To fill the growing gap between“expected pensions”and“pensions paid in reality”by some definedbenefit schemes, governments have introduced or strengthened schemes based on individual accounts, either mandated or voluntarily. These schemes, however, are defined-contribution schemes, under which the pension at retirement will be equivalent to the annuity resulting from the accumulated funds in the individual account. In other words, defined-contribution pension benefits, unlike defined-benefit ones, heavily depend on the interest rate (rate of return if invested in the “markets”) during the accumulation period and the assumed long-term interest rate (rate of return) during retirement.Classification of pension pillars allows for an examination of the demographic and, consequently, financial and political pressures that pension systems and countries face. For countries with a multipillar system, where there is almost universal coverage, the pensions have come to be conceived as a “right”, and pension reforms could experience political backlash. In contrast, in countries with limited pension coverage – which constitute almost all developing countries, a rapidly increasing older population provides the political credit necessary to parties creating or expanding the old-age pension system.Publicly versus privately managed pension schemesRecent widespread retrenchment of governments from providing social security and replacing the financial income gap with private pension solutions or defined-contributions schemes has reduced government liability only preliminarily, at face value. Handing over responsibility for old-age provision to individual citizens has created new problems that require strong government involvement and responsibility for private schemes through the back-door. One of the main problems arising with this type of policy is known as the “market for lemons” problem (Akerlof, 1970), which has been observed in a number of countries undergoing marketization reforms of the pension policy.Analysis of the market-for-lemons problem17 finds that the quality of products traded in a market can degrade in the presence of information asymmetry between buyers and sellers, leaving only “lemons”, or bad-quality items. In the real world, the market for private pension contracts might not collapse as quickly as in the case of lemons because the quality of the pension product bought might only be revealed after many years, possibly only at retirement.18 However, the illusion of bought product quality might vanish (sooner or later)if strong consumer protection institutions exist or when other systems of public communication work as eye-openers.To counter such problems inherent to the private pension market, governments have established complex monitoring and control structures, including through tax offices as well as marketing measures and the like, aiming to guarantee a high quality of private pension contracts offered and to avoid the collapse of the private pension market. Germany is a case in which the Government’s financial liability was formally reduced and people realized that they would receive an insufficient pension from the social insurance scheme in old age and that the gap would not be filled by the additional scheme (Riester pensions) offered. Indeed, the need for government’s responsibility is reinstated.Socioeconomic implications of pensionsPension schemes interact differently with their host economy and wider societal fabric. Social security planners are especially interested to find out whether certain social security designs (or17 A “lemon” in American slang typically refers to a car that is found defective only after it has been bought. 18 See Old-Age Pensions in Asia for further discussion.15traditions19) might be preferable to others and under which conditions. Therefore, the issue to be explored is how those traditions have functioned in reality and whether empirical findings are significant enough to allow informed choices on the scheme design. The ultimate task would be to choose the one social security system that best fits “the society´s” preferences and, at the same time,best serves economic development.Given the empirical relevance, the following explores in more depth the economic implications of the social insurance tradition of pensions20 while we continue to distinguish between: (i) Bismarckian, or contribution-financed, social insurance; and (ii) Beveridgean, or tax-financed, social transfer systems.21 The section discusses their functioning in the context of economic development.Bismarckian pension systemsWithin a decade (the 1880s), Bismarck22 launched a comprehensive system of health, occupational accident and pension insurance that was later called “a lucky strike of history”(Ritter, 2014) – not the least because of its intrinsic economic features.For a proper economic understanding of Bismarckian systems, it is paramount to realize that revenue collected and benefits paid are intrinsically interwoven. This systemic interdependency depends on, as well as triggers, a set of behavioural features of labour market participants, as follows: (i) employers and workers have an interest to maximize productivity in order to maximize wages and thus to maximize benefits; (ii) the system needs formal employment to be administratively able to pay benefits; and (iii) it is, by construction, not universal, and thus to emulate universal coverage, it requires maximum employment. The latter triggers (iv) the need for pro-growth and proemployment macroeconomic policies.Within Bismarckian systems, income redistribution almost solely occurs under social health insurance; otherwise, system-immanent redistribution is close to non-existent. In fact, with a too narrowly set contribution collection ceiling, the system can even have digressive features. A cash redistributive element is social assistance, which is supposed to be paid only in residual cases under this type of system. In other words, the more a Bismarckian system needs to fall back on supporting private households by social assistance payments, the more it loses its intrinsic economic growth features and is considered a failure.The long-standing and worldwide attractiveness and success in many countries of the Bismarckian system can be explained in terms of five aspects: (i) it has a “natural” logic, not the least through its elements of “actuarial fairness”, and is intuitively easy to understand by its members and economic actors in general; (ii) it can be mere pay-as-you-go but also be prefunded, with multiple financing combinations possible; (iii) it is possible to pay benefits without a prior contribution accumulation period; (iv) it can be technically adjusted to demographic and new economic developments; and (v) it can be made redistributive (within limits) while not losing its core, economically supportive characteristics.19 See Old-Age Pensions in Asia for further discussion. 20 Since around three to four decades, the “marketization tradition” has gained prominence, especially among econ-omists. However,detailed comments are not provided here, not the least because this tradition has broadly failed the social test of reality where implemented. Nevertheless, some thought is provided on it in section 3.1, where we indirectly develop our (disagreeing) standpoint. 21 Strictly speaking, Beveridge’s proposal was to finance the welfare state out of flat contributions, which can be regarded as capitation, or a tax. 22 Otto Eduard Leopold, Prince of Bismarck, Duke of Lauenburg, was born in 1815 and died in 1898 and was known as Otto von Bismarck, first Chancellor of Germany (1871–90).16The last point is important because the Bismarckian retirement system operates well as long as socioeconomic and technological developments allow for a broad middle class. It comes to its limits, and then becomes socially dysfunctional, when the middle-class erodes or simply ceases to exist, with wages only supporting absolute benefit levels that are insufficient to avoid poverty in old age. Once benefits fall systematically below the poverty line as a result of too-low wages, they must be topped-up, also systematically, by social assistance. This situation or sequence of events should not become the normal case and thus requires system adjustment through adequate reform.This last feature recently contributed to the questioning of the social rationale of Bismarckian systems: Over the past two decades, an increasing number of countries have developed large low-wage sectors with precarious jobs, while others have not yet created a significant medium-wages sector, or middleclass. If no alternative solutions are developed, the result will be that increasing numbers of these workers will have to work up until their elder ages and will often still live in poverty when retired.Beveridgean pension systemsBeveridge,23a British economist and politician, famously proposed a pension system in which the State guarantees freedom from want for all by way of: all individuals paying (low) flat contributions (capitation tax); and all individuals receiving sufficient (“adequate”) cash (and services) in case of need; there is no means test; and the State covers any emerging deficit.Individuals wanting greater protection than freedom from want are referred to the private sector. It is important to understand that under Beveridge’s proposal, as opposed to Bismarck’s position, finance and benefits are independent of each other. This has two main implications: (i)The level of benefits can and must be decided within every government budget and depends on societal definition. No social assistance is required because the system, by design, guarantees freedom from want. And (ii), the system needs adequate production to generate the taxable resources for redistribution.Strictly speaking, there is, however, no need for formal employment as long as taxes can be collected elsewhere, independently from any employment.24 To be sure, Beveridge’s system needs taxable income (gross domestic product, or GDP) and, therefore, needs employment policies aiming to produce sufficient amounts of GDP as a basis for redistribution, but any pro-employment and economic growth policy is, in Beveridge’s context, fully de-linked from the accumulation of entitlements and the benefit payment system.With these features, the Beveridgean system allows individuals to make decisions irrespective of material constraints. The system is potentially highly redistributive and reacts with the economy only on the demand side, whereas Bismarck’s system, while far less redistributive, also reacts on the supply side of the economy.Considering the system descriptions provided and the discussions surrounding their factual implications within society and an economy, it becomes difficult to say which of the two social insurance system designs (Bismarckian or Beveridgean) is “economically better”. Prima vista, one might prefer Bismarckian because of its clear and proven productivity orientation. But the world is heading towards a multitude of problems that were unknown when both systems were established. In this context, one core question for social policy-makers to consider is whether more equity-oriented and redistributive systems (yet providing benefits clearly above the poverty line) will be the more adequate design in the future.23 Baron William Henry Beveridge was born in 1879 and died in 1963; he published what became his famous report, Social Insurance and Allied Services, in 1942.24 The said feature applies technically, despite the fact that Beveridge argued in favour of full employment policies.17Old-age pension financing options and macroeconomic considerationsBefore discussing various options for financing social security pensions, it is necessary to understand the typical cost developments associated with pension schemes, with the main challenge being that pension schemes mature slowly, typically over many decades. Financing systems must take account of this phenomenon, not the least because the parameters of the systems react differently to the maturation process.There are two basic indicators that describe the evolution of costs, or current expenditure, of a pension scheme,both of which are relative measures:1. the pay-as-you-go contribution cost rate, which describes the expenditure of the scheme relative to the insured earnings of the insured persons.2. the national pension cost: describes the expenditure of a pension scheme or the sum of all pension schemes that might exist in a country as a share of GDP.Indicator 1 describes the real financial burden for pay-as-you-go financed social insurance schemes as well as the hypothetical financial burden for universal schemes financed out of general revenues of the active contributing population, which is measured as a percentage of gross insurable earnings. Indicator 2 places pension expenditure in a national context. Moreover, indicator 1 is equal to the mathematical product of: the demographic ratio, which is identical to the ratio of the number of pension beneficiaries tothe number of active contributors; and the financial ratio, which is the ratio of the average pension to average insurable earnings.This product is equal to the ratio of total pension expenditure over total insurable earnings, as in the following equations 1 and 2:Equation 1Equation 2All pension schemes, whether introduced on a national scale or for specific groups within the population, reflect a characteristic expenditure development over time.25It can mathematically and generically be described as a perfect logistical curve, under a number of simplifying assumptions that are typically not met in reality, thus painting an ideal picture of a scheme’s maturation over 75– 80 years, or roughly three generations. In reality, the logistical maturity curves of pension schemes are never as smooth as in mathematical models. Cost-development curves are observed for informing pension scheme adjustments during different phases of their maturation process. During the first years after the inception of a given scheme, usually only a few pensions are paid; when pensions are paid at all, these are mostly invalidity25 See figure 2.1.1 and figure 5.2 in Cichon et al., 2004 in which the curves presented basically represent variants of the so-called logistical or quasi-logistical functions approaching asymptotically an upper boundary.18pensions. During a second phase, the number of pensioners grows rapidly, and as a result, indicators 1 and 2 rise steeply. There are mainly three reasons for these steep increases: A new cohort of pensioners joins the “pension force” each year. As the coverage of pension schemes increases, these cohorts of pensioners, generally after a time lag of around three decades, also increase every year. If the growth of the pension cohort in the early years after the first pensioners become eligible for benefits exceeds the growth of the active contributing population in the same years, then the payas-you-go cost increases naturally. The average pension entitlements of each early new cohort of pensioners increases in line with a generally rising average number of years of service of the new cohorts entering retirement.The stage of maturity for pension schemes is reached when all pensioners in a given scheme can look back on a full career insured in that scheme. This is the situation at the time of death of the first cohort of workers who joined the scheme when joining the labour force and at a point in time when the population coverage of the scheme has reached its ultimate level. Maturation typically does not occur earlier than about seven decades following the inception of the scheme.And even then, when it reaches maturity, any scheme can only be in a stationary state if legal provisions have not changed substantially during the maturing process, which is not often the case. In anticipation of the demographic impact of ageing, for example, the standard retirement ages might be pushed forward and benefit provisions might be reduced.This analysis reveals that increasing pension costs is a perfectly normal phenomenon expected to happen during the decades-long maturation of a national pension scheme. Rising pension costs are not necessarily indicative of a financial problem.In political reality, the long-term characteristics of pension systems are not always fully understood. This is one of the reasons that young pension schemes, which may pay hardly any old-age pensions for decades while only collecting contributions and accumulating funds, trigger expectations that almost any level of generosity can be afforded. As a result, legislation is designed accordingly, and the public (scheme members) develops expectations that take those generous rules for granted. With time passing and the scheme maturing, however, generous legislation begins taking its toll: scheme generosity might become unaffordable as the scheme approaches maturity, or contribution rate increases and/or benefit cuts might turn out unavoidable. The public usually resents such measures, and schemes can easily run into financial difficulties and end up being politically discredited.26Nevertheless, it may be socially as well as politically rational to pay pensions in“premature”stages to persons who are close to retirement age when the scheme is introduced. In technical jargon, this is called “grandfathering”. Failure to do so has contributed towards undermining trust in new schemes due to contributions being collected from active employed persons while their own parents did not receive any old-age income support.The various designs of pension financing systems accommodate the cost developments in various ways, by specifying who bears the costs, at which amount and when during the maturation process. Generally, the costs can be borne through taxes, contributions coupled or not with income from26 The rationale and political economy of debates are enormously complex – this very report is one element of it – and cannot be addressed in any detail in this discussion.19interest or other returns on capital, or a combination thereof. Tax financing is financially and administratively simple but politically not necessarily so.In the following section, attention is given to a more complex form of pension financing – through contributions. Contribution-based systems are characterized by the actual incidence of contributions, the contribution base and the desired pension scheme funding level.Incidence of contributionsThe relative level of employer and worker old-age pension contributions varies considerably across countries, including within the ASEAN Member States. In most countries, the employer pays half or more of the full contribution, a practice that is aligned with what Convention No.102 establishes in Article 71: “The total of the insurance contributions borne by the employees protected shall not exceed 50 percent of the total of the financial resources allocated to the protection of employees and their wives and children” (ILO,1952). Where social security pensions were replaced through radical reforms with defined-contribution systems, contributions are usually, with few exceptions, solely paid by the employee, as has been the case in Chile and China.27The split between employers and workers does not normally affect total labour costs because the employee ultimately bears the cost by receiving a lower wage than they would otherwise receive and assuming that the tax treatment of employers’ and workers’ contributions is consistent. An advantage of having employers and workers pay at least part of the social security contribution is that it makes them more aware of the cost of the system and gives them a sense of ownership of the scheme. Employers’ shares are especially important because social insurance schemes cannot run successfully without enterprises’ direct administrative involvement.Social security contribution rates can be progressive, meaning that they increase with a worker’s salary, or regressive, which always occurs when a contribution collection ceiling is applied. Furthermore, social security contributions are usually mandated, but they can also be voluntary. In a number of countries, workers meeting qualifying conditions can reduce their social security contribution payments as long as they participate in an individual or occupational pension scheme offering at least the same benefit entitlements.28 This type of arrangement is called“contracting-out”. Often, old-age pension schemes also allow for the voluntary payment of additional contributions.The financing of pension benefits for self-employed workers under a social security scheme is a critical topic in most national systems, not the least because compliance is a notorious problem, but also because self-employed individuals often have to pay both, the employer’s and worker’s contributions.Contribution baseThe contribution base is generally the worker’s gross wage or a certain part thereof. Rather than using actual earnings, however, some systems use a series of wage bands, whereby all workers whose earnings fall within a particular band pay the same amount. Wage bands were initially introduced to minimize record-keeping and simplify employers’ administrative chores. With modern IT systems, such attempt at simplification ends up adding administrative burdens rather than simplifying them.Mandatory social security contributions and thus the contribution base itself are often subject to restrictions relative to individuals’ earning levels. Ceilings are commonly applied on covered27 In China, the system combines a defined-benefit scheme funded fully by the employer, with a defined-contribution scheme funded only by the employee.28 To judge, which is often difficult, see the brief discussion on the market of lemons.20earnings, with no social security contributions required on higher earnings. Among countries in the Organisation for Economic Co-operation and Development (OECD), the ceiling is often between 100 percent and 200 percent of the average salary, calculated on the basis of salaries with or without a ceiling. The purpose of a contribution ceiling is usually to allow for high-income workers to establish occupational pension plans and to have private savings or life-insurance, for instance. Alternatively, if there is no ceiling but a maximum benefit defined, high-income workers receive no additional benefit from contributions made on earnings above what is necessary to receive the maximum benefit. Actually, contributions above a ceiling and not relevant to benefits effectively constitute a redistributive tax instrument.Contribution ceilings must be regularly adjusted over time with general wage developments,although often they are not. In this case the proportion of total earnings subject to contributions of a country’s total wage bill declines rapidly (depending on the pace of inflation). As long as benefits are more or less strictly related to covered earnings, this might not pose a big problem for the financial equilibrium of the scheme, although it might make benefit levels irrelevant from a social and purchasing power point of view. If benefit levels are raised without the ceiling adjusted and kept at socially satisfying levels, they can create massive financial scheme deficits. Thailand’s overall low adequacy of pension benefits paid by its Social Security Fund, for instance, is attributed to the fact that its ceiling on contributions was not adjusted for many years after the scheme’s inception in 1998, and – to maintain financial stability – benefits were insufficiently adjusted. Lack of proper periodic indexation of ceilings and benefits eventually turns pensions, as a source of income, irrelevant to the pensioner.The counter-concept to ceilings are contribution floors on income, subject to mandated social security contributions, which do not require workers with low earnings (below the floor) to contribute.Total contributions can be increased by means of raising the contribution rate or by broadening the contribution base. In practice, both measures are usually applied. The contribution base could be increased many ways, including by raising a given ceiling or lowering a given floor or by making nonwage earnings (wages in-kind) subject to social security contribution. Alternatively, the contribution base could be reduced by relevant measures. Whether such moves have impact on benefit levels would depend on concurrent legislation.A technically though usually not politically “easy” way of broadening the contribution base is to redefine the term insurable earnings, as stipulated by national legislation. In many countries, including the ASEAN Member States, only the basic salary is taxable and subject to contributions, leaving out the remuneration paid on top. In Viet Nam, for example, narrowing the definition of insurable earnings resulted in much lower effective contribution rates and, thus,lower contribution revenue than stipulated by the law. Accordingly, pension benefits became much lower than they otherwise could be. Overcoming this detrimental situation is difficult because workers and employers most often coalesce in this respect because they share an interest in maximizing shortterm disposable income during workers’ active ages.Types of contribution rate regimesThe type of contribution charged by a pension scheme reflects two aspects: (i) the policy choice made concerning preferences for different financial rules of social security; and (ii) the level of funding accepted or aimed to achieve.21The mathematical formulae governing the different pension-financing regimes can be found in Cichon et al. (2004).Table 1 provides a summary of the different types.Table 1.Types of contributions according to financing method and their funding levelFinancing methodContribution rateFunding levelPublic pension schemesPay as you goContributions collected annually It does not lead to the accumulation ofshould almost exactly match the funds, except for a modest contingencyexpected expenditure of the year; reserve maintained to avoid liquidityrate can vary year by year.problems.Partial funding (scaled premium)The contribution rate is fixed The size of the reserve normally increases and maintained constant over a in line with the length of the equilibrium fixed multi-annual period (called period, and a funding target at the end of “equilibrium period”), so that incomes the equilibrium period is fixed, either to and expenditure should be in balance avoid liquidity problems or to pre-finance or equilibrium over this period, taking part of the liabilities. into account the funding target at the end of the period.Partial funding (general average premium)The chosen contribution level It accumulates similar reserves as fullybalances the present value of: total funded schemes (see below), but theexpected future benefits (minus funding level is higher during the initialinitial reserves) and contribution phases of a scheme and lower towardsincome; it stays constant, in theory, the final years (when calculated over aad infinitum.defined long period).Full funding (individual mandatory savings)Earnings related, or otherwise The terminal funding level of the reserve adjusted contribution, is theoretically of the scheme is automatically equal to constant throughout the life of the the total amount of savings in all individual scheme, which is automatically in accounts. financial equilibrium.Full funding (terminal funding)The contribution rate is fixed, such A reserve equal to the value of all accrued that contributions collected over benefit entitlements is accumulated. The a defined period are equal to the value of accrued benefit entitlements discounted value of the benefit includes the value of current pensions entitlements accrued during the and the value of benefit entitlements same period, subject to adjustments earned to date by active and inactive for accumulated surpluses or past members. unfunded liabilities.Private pension schemesFull fundingThe contribution is a flat-rate uniform The funding is equal to the required levelcontribution, usually differentiated by of terminal funding – reserves are alwaysage of entry (in theory, kept stable equal to the present value of all accruedover long periods).entitlements and pensions in payment.Pay as you goThe contribution is flat or income This is a specialty of certain occupational related (which can change annually). schemes that are State mandated andState guaranteed (such as in France).Book reservesThe contribution rate is imputed and It is identical with the reserve underearnings related, generally only paid full funding but invested in theby the employer (in theory, stable enterprise operating the scheme, andover long periods).it is re-insurance obligatory (such as inGermany).Source: Cichon et al., 2004, p. 278, marginally adapted.22With regards to setting contribution rates and resulting benefit amounts, private sector definedbenefit schemes generally charge flat-rate contributions per cohort, while many occupational schemes (defined benefit or defined contribution) and virtually all Pillar1 social security schemes charge an earnings-related contribution in the form of a fixed percentage of insurable earnings. Benefit levels may depend on the amount of contributions paid or on earnings during the active age, although they do not necessarily have to. With major non-earnings-related components in the pension formula, contributions charged as fixed rates can make up for a scheme that is progressive – redistributive from “rich” to “poor”. An alternative instrument for adding a redistributive element to pension schemes is to charge greater contributions for higher-income earners while basing the pension benefit solely on non-contribution indicators, such as earnings. For instance, in its current pension formula, Viet Nam rewards the first 15 years of work of a contributor with a higher accrual rate. Because individuals who are able to accumulate larger contributory careers also have higher incomes, such a measure can be considered as redistributive. By using a pension formula that incorporates the average salary of a group and not an individual as a reference to calculate part of the benefit, China ensures that people with earnings below the average wage have pensions that are higher than if calculated on a pure individual basis of actuarial equivalency. Most social security schemes are de facto scaled-premium systems, where pay-as-you-go systems can be considered a scaled premium system with a minimal equilibrium period of just one year (table 1). Even systems that started out as fully funded and based on a general average premium approach were often turned into scaled-premium schemes when the real financial development of the scheme turned out not to be in line with initial expectations and contribution rates had to be increased successively (the history of the more than 100-year-old German public pension insurance system serves as an example). Figure 2 illustrates the typical evolution of the most important alternative contribution rates. Figure 2. Typical developments of alternative types of contribution ratesNote: PAYG=pay as you go; GAP=general average premium. Source: Cichon et al., 2004, p.281.Determining the level of contributions Even if the parameters of a pension scheme are basically set and a decision has been taken on the23funding policy to be applied, there remains some discretionary space for fixing the contribution rate. Variation enters the scene according to many factors, including: the political need to maintain the stability of the contribution rate over extended periods for reasons of political acceptance; the need to adapt the contribution level and, hence, non-wage labour costs to the prevailing economic situation to stimulate future development; the capacity to invest and manage the accumulated reserves effectively without undue risk of mismanagement or asset depreciation; the capacity of the accessible (mostly domestic) financial markets to absorb the accumulated reserves; and the need for a contingency reserve to cope with unexpected income shortfalls or increases in expenditure.Rate setting in fully and highly funded schemesIf a country opts for a fully or highly funded pension scheme, there is not much room for debate because the contribution rate must be set at a high level to ensure that the later pension paid is to be of a reasonable and decent amount. In fully funded defined-benefit schemes, the contribution rate is the varying factor once the level of benefits has been defined. For defined-contribution schemes, the setting of the contribution rate under mandatory retirement savings ideally should be based on calculations that start with the desired annuity levels at retirement after a full career. In practice, most countries, including the ASEAN Member States, have set the contribution rates according to affordability and relative to economic and financial aspects.Over a period of two to three decades,governments aimed at high levels of funding to increase the national savings rate. This purpose since disappeared,with research on the issue inconclusive: A statistically significant increase of national savings by way of pension funding could not be systematically proven.29Pension funding through investments in government securities is considered to belittle more than pay-as-you-go financing of the overall government sector: While reserves are invested in public securities, non-social security tax rates are kept low because excess government expenditure is financed through social security contributions.Even though the accumulation of a high level of reserve funds can be desirable to promote contribution-rate stability in the context of an ageing population or a maturing scheme, actuarially calculated contribution rates might still need adjustment from time to time. Such actuarial calculations are reaching out far into the future and, because the future is open, the reality is likely to deviate from the underlying assumptions of actuarial calculations to some extent, requiring substantial contribution-rate adjustments now and then. This is especially the case in countries with unstable macroeconomic conditions or limited or no capital markets. History has shown that high inflation, bad investment policies or management or poor pension system governance may contribute towards depleting pension assets.In any case, if a country decides to build up a substantial reserve for economic purposes, it must be29 It is a widespread macroeconomic misunderstanding that savings are a requisite for economic growth; savings are always a result – an outcome of the income production (of GDP), not an input to it.24done during the early stages of a scheme’s existence and maturation, when expenditure is low and the contribution rate is markedly higher than the pay-as-you-go cost rate. One way of doing this is to aim for a general average premium system.30 In such a system, a high contribution level is not needed to cover current expenditure in the early years of the scheme. In this sense, high contribution rates can be considered fair to the early contributing cohorts because they become entitled to future benefits in exchange for their current contribution. In essence, this is the principle of intergenerational equity. If, instead,a scheme operated on the basis of scaled premium or pay as you go from its inception, the first generations or cohorts of insured persons would benefit more from the scheme, in the sense that their “financial return” would be higher than it would be for later generations.The issue with adopting a general average premium scheme for building reserves is more complex. When countries are still in the development phase of their pension system, a general average premium scheme would typically mean higher labour costs than a pay-as-you-go scheme. Hence, it must be carefully assessed whether any possible negative effects of increased labour costs – because general average premium rates are higher than pay-as-you-go rates – outpace the positive effects of intergenerational equity. In the end, this becomes a question of societal choice.Rate setting in pay-as-you-go or scaled-premium schemesIn pay-as-you-go schemes, the contribution rate is fixed every budget year, whereas in scaledpremium schemes, the period of one year is extended to several years (table 1). In other words, a pay-as-you-go scheme can formally be understood as an extreme version of a scaled-premium scheme. The differences between the two financing regimes are floating, which is why a distinction between the two is not always possible.The contribution rate under pay as you go mainly depends on the expected pension expenditure: on the expected number of pensions in payment, their level, the estimated number of new pensioners and their entitlements. The resulting expenditure has to be calculated relative to the expected number of contributors and their average earnings. The pay-as-you-go rate depends also on the contingency reserve to be maintained by the end of the budget year. There is no hard-and-fast rule as to the level of contingency reserve to be maintained.Governments, for various reasons, often try to avoid annual fixation of the contribution rate, and this is why many have opted for a scaled-premium approach. The inherent problem connected with such policy is that governments sometimes wait too long before raising contribution rates because each increase, even if perfectly normal and financially foreseeable, is likely to be politically unpopular. Waiting too long means that the scheme will either run into liquidity problems or need a big contribution hike. Uninformed political propaganda will, most likely immediately, lead to declaring the scheme bankrupt and then calling for radical changes. Policy often reacts by taking “consolidation measures” that entail cutting benefit levels or tightening eligibility conditions. Such ad hoc modifications of the benefits are generally detrimental to the scheme’s public credibility. Thus, to avoid short-term manipulations as much as possible, schemes ideally need three regulatory provisions:1. a clear definition of the actuarial target equilibrium combined with a rule on when the contribution rate must be increased, where necessary.2. a benefit formula that can be maintained at the pension scheme’s stationary state, not only30 A general average premium implies that the constant contribution rate is applicable infinitively, theoretically guaranteeing that scheme expenditure can always be covered by the contributions collected and the funds accumulated in the reserve. It is calculated by dividing the present value of all future benefits (minus the reserve existing at the evaluation date) by the present value of all future contributory (insurable) earnings.25during the scheme’s early maturation stage but possibly combined with transparent transitional benefit provisions for the early pensioner generations (“grandfather clauses”).3. a set of demographic and financial stabilizers stipulating how the financial consequences of adverse demographic and financial developments are allocated between contributors and beneficiaries.Funding or pay as you go: The issue of interest rates, wage growth and employment growthLet i be the prevailing rate of return or interest rate in an economy, w the rate of wage growth, or of insured-earnings growth, and g the rate of growth of employment or the number of insured (Samuelson, 1958):Equation 3Equation 4Equation 5As long as equation 3 holds, funded schemes provide a higher return on contribution payments than pay-as-you-go schemes and vice versa if equation 4 holds. The situation is indifferent when equation 5 holds.It has been argued that the relationship in equation 3 holds for ageing societies, where the rate of wage growth (g) might even be negative, insinuating that the proper old-age system to be chosen should be fully funded. However, the situation is not so obvious in ageing societies with large informal sectors in labour markets and where the labour income share has still much room to increase, even when GDP growth is only modest.These and other considerations make it difficult to base the decision for or against full funding of pension schemes, which are financing decisions that reach out into the future for more than a hundred years. For example, to know which of the above relations actually holds in an economy can only be seen retrospectively. Also, the relationship might change over time. Moreover, if the relationship in equation 3 is taken as a policydatum that cannot be influenced through respective economic policies, this implies that the fruits of economic activities flow in the long run to capital and rentiers. From the pension system analyst’s point of view, it is difficult to comprehend how such policy approach could be held in the long run.When pension funds of ageing societies reach maturity, they will have accumulated the monetary lifetime savings of “the many”. And when these funds’ assets need to be liquidized or sold, there will only be a “few” to buy them. According to standard analysis of market behaviour, this may cause an asset meltdown, which will result in lower-than-expected pensions that can be paid. The current European situation is an example, whereby returns on enormous amounts of liquid savings are at historically low levels, and it appears that huge savings amounts make money “cheap” (interest rates are low and, in some instances, even negative). The same may easily happen to the pension funds of the fast-ageing ASEAN countries but can be avoided by way of learning from the mistakes of the West.Old-age pensions and the informal sector31All ASEAN Member States are confronted with coverage problems resulting from, among other reasons, the dimensions of their informal sectors,32 which significantly affect the share of informal employment relationships –some more so, others less so.The challenges on how to close or reduce the coverage gaps in informal employment or how to include the population working under informal employment conditions and mostly working in the31 This chapter took inspiration and evidence (partially updated) from Hinz et al., 2013. 32 Coverage problems are not specific to Asia but are a worldwide problem, including the most advanced social democ-racies. Maintaining high coverage is a never-ending task.26informal sector in old-age income security systems have been subject to many years of national and international debates, trials and approaches. Expanding social protection, including to informal economy workers, reflects recent commitments at the global and regional levels to extend social security coverage to all workers in the informal economy and to promote their transition from the informal to the formal economy. This is in line with the ILO Social Protection Floors Recommendation, 2012 (No. 202) as well as the ILO Transition from Informal to Formal Employment Recommendation, 2014 (No. 204).According to the guidelines33 endorsed by the Seventeenth International Conference of Labour Statisticians in 2003, informality is defined as “all economic activities by workers and economic units that are – in law or in practice – not covered or insufficiently covered by formal arrangements”.Informal employment refers to all employment arrangements that do not provide individuals with legal or social protection through their work, thereby leaving them more exposed to economic risk than others, whether the economic units they work for or operate in are formal enterprises, informal enterprises or households. To further classify workers within formal or informal employment, a distinction is made according to the categories of the International Classification of Status in Employment (ICSE-93). Based on this classification, informality concerns two groups of workers, as shown in table 2 and figure 3.Table 2. Informality conceptsPersons employed in the informal sectorEmployers and employees in informal enterprises Own-account (self-employed) workers in their own informal enterprise Contributing family workers working in informal enterprises Members of informal producers’ cooperativesPersons in informal employment outside the informal sectorEmployees in formal enterprises not covered by social protection through their workPaid domestic workers not covered by social protection through their workContributing family workers working in formal enterprisesSource: ICLS, 1993.Employment in the informal sector and informal employment are concepts that refer to different aspects of the “informalization” of employment and to different targets for policy-making. While the informal sector refers to informal enterprises, informal employment refers to informal jobs. In this sense, employment in the informal economy can be defined as the sum of employment in the informal sector and of informal employment found outside the informal sector. The two concepts are not interchangeable, but they are both useful for descriptive and analytical purposes and are hence complementary (figure 3).Figure 3. Conceptual framework: Informal employment33 For more information on the lines referencing results of the Seventeenth International Conference of Labour Statisticians, see ILO, forthcoming b and ILO, 2013.27Note: (a)=As defined by the Fifteenth International Conference of Labour Statisticians (excluding households employing paid domestic workers); (b) households producing goods exclusively for their own final use and households employing paid domestic workers. Cells shaded in dark grey refer to jobs, which, by definition, do not exist in the type of production unit in question. Cells shaded in light grey refer to formal jobs. Unshaded cells represent the various types of informal jobs. Informal employment: cells 1–6 and 8–10. Employment in the informal sector: cells 3–8. Informal employment outside the informal sector: cells 1, 2, 9 and 10.Source: ICLS, 2003.Workers in the informal economy differ widely in terms of income level, regularity and seasonality; their status in employment (employees, employers, own-account workers, casual workers and domestic workers); sector of employment (services, agriculture, industry); type and size of enterprise; location (urban or rural); employment protection (type and duration of contract); annual leave protection; and social protection.The possibilities and strategies used for extending pension coverage depend on the prevailing types of informality. For example, reducing informality is not likely to occur in many situations by “formalizing labour”via social security measures only. Instead, increasing formal employment can be achieved by the formalization of businesses,which is particularly important for small and mediumsized enterprises (SMEs), and by measures that go beyond the scope of technical social security policies and/or administration. In other situations, the extension of old-age pension coverage will depend on the capacity of systems to increase compliance within the frequently not (or not fully) compliant formal sector. Regional statistics point to a significant number of enterprises that are formal but use informal work, or contra legem (against the law). This, in turn, can be a result of lack law enforcement, together with the lack of inspection capacity. Lack of trust in the social security institutions and lack of social security culture among the population are also essential elements contributing to a lack of compliance with social security obligations.There has always been the expectation that old-age benefit entitlement through individual contribution payments would trigger incentives in a society,which would automatically result in increased pension coverage. This expectation prevailed in the Bismarckian systems as well as (but less so)the Beveridgean systems. But evidence to date shows that incentives alone are not sufficient to increase coverage, and pension systems must be well organized, with their functioning depending on good governance, including strict and sustained law enforcement. For example, first and naive reform attempts (mainly in Eastern Europe and Central Asia) to counter low pension coverage or to stem falling coverage rates that occurred in the aftermath of the collapse of the Soviet Union aimed at or included giving incentives, which consisted of strengthening or establishing better links between contributions and benefits in mandated and earnings-related old-age pension schemes. Success was limited due to weakened state structures and because the reforms were often accompanied by cuts in benefit entitlements and provisions, raising doubts about the reliability of the schemes. As a result, coverage declined even further in some of those countries (Holzmann, Hinz and Tuesta, 2013).Coverage issues are subject to broader economic considerations. This is why the design of pension reforms, to the extent that they successfully support coverage extension and formality, is of economic relevance. Stagnant or even growing informality hampers economic development because it excludes workers and employers from access to formal productivity-enhancing measures.Countries like the Republic of Korea and Japan have introduced (over their social security history) specific policies that promote the extension of coverage to groups that were, at an early stage,28not formalized. European countries undertook efforts to integrate groups, such as self-employed persons or farmers, by simplifying procedures or subsidizing part of the costs. The inclusion of domestic workers happened either through the creation of specific regimes or through legal inclusion in existing social security schemes. Overall, the success of each of those policies varied with the degree of effectiveness of administration and law enforcement.DemograntsDemogrants are flat universal benefits, usually paid out of taxation on the condition of residency and age, such as a universal flat pension paid to all residents at and after a certain age. Many countries implemented such a scheme or are currently experimenting with one. They are prominent among development economists, not the least because of their alleged transparency and simplicity to administer. Examples of countries that have adopted demogrants are Brunei Darussalam, the Netherlands, Thailand (box 1), Timor-Leste and, the latest, New Zealand (Holzmann, Hinz and Tuesta, 2013).Box 1 Old-age allowance in ThailandFaced with an ageing population, a vast informal economy and a young contributory system that only recently started paying old-age benefits, the Thai Government first introduced tax-financed, means-tested, old-age pensions in 1993. The old-age allowance was designed at the national level but differed widely in the interpretation of implementation guidelines and application of eligibility criteria at the local level. Inefficiencies and leakages led to more than half of the intended target group being excluded from a pension. Reforms to the targeting methodology were made in 1995, 2002 and 2005, which resulted in increased representation of older advocates and community members in the committees in charge of the selection and disbursement processes. As a result of this and with increased budget, the number of beneficiaries went up steadily in the years that followed.When the Government dropped ineffective targeting procedures altogether in 2009 and made the old-age allowance available to all people not in receipt of any income or other pension, the coverage increased dramatically. Overall, old-age pension coverage was at an estimated 83 percent of the older population in 2016. This high coverage, mainly achieved through the old-age allowance, is an almost unique achievement among countries at a similar stage of economic development in the ASEAN region. The coverage gap of 17 percent is likely to be a result of two factors: (i) selfselected abstinence due to modest benefit levels, especially given that the share of recipients is significantly smaller in richer households, and (ii) administrative obstacles faced by some of the eligible individuals.The old-age allowance benefit is tiered according to the recipient’s age. People aged 60–69 receive 600 Thai baht (THB) per month, while those aged 70–79 receive THB700 per month, those aged 80–89 receive THB800 and those 90 and older receive THB1,000. The old-age allowance by itself is unable to sufficiently protect older persons from income poverty, considering that the average benefit level corresponded to merely 10 percent of the minimum wage in 2016. The benefit level remains far below the national poverty line (at THB2,647in 2014). Any reform would have to ensure adequate old-age protection through a combination of contributory and non-contributory schemes (see also the country brief at the end of the report).Source: ILO, 2016f.Introducing universal benefits or simplified means-tested social assistance to take care of afterretirement security interests of workers in informal employment may increase macroeconomic demand, but unintended effects might also have to be considered. Some analysts argue that in countries with large informal sectors, incentives to join and remain in the formal sector can be reduced, while formal sector workers’ propensity to work informally might increase as a result of29their increasing income tax burden. This would especially have to be expected in countries with large informal sectors, in which only a small formal middle-class exists that could be effectively income taxed. Collection of additional indirect taxes requires a system, which in the informal as well as the formal sectors often only works insufficiently. Societal tensions might occur, not the least because of distributional and equity issues, and such tensions would probably not decline even if governments implemented measures to increase fiscal space.It is therefore not surprising that proposals of coverage extension through non-contributory basic schemes raise fiscal concerns, not only among the ASEAN Member States and in ministries of finance but also among other groups. Where such steps were undertaken, such as in Thailand, tax resistance of the middle-classes was high and, as a result, the benefits paid were of a low subsidy nature. And their levels remain below the poverty line.In sum, many factors contributed to the concerns related to tax-financed schemes. The cost of coverage extension through basic tax-financed schemes, the need to control the fiscal costs of existing national earnings-based defined-benefits schemes and the effects of sustained informality on productivity and public revenue have led to a narrative of a downward spiral where: the introduction or strengthening of basic provisions can lead to constraints in employment and increased informality, which possibly increases incentives for labour to leave the formal sector, which would worsen the fiscal position of the public pension scheme, and finally could lead to benefit cuts, which in turn increase the need for better basic benefits (Holzmann, Hinz and Tuesta, 2013).Under standard economic reasoning, this narrative is theoretically not implausible. But the wish might be father to this deduction. Implementing or reactivating social policy has always required governments to break-free and cut the Gordian knot.34 Developments under this narrative could be avoided through measures to extend the available fiscal space (Ortiz, Cummins and Karunanethy, 2017) and by careful design of tax-financed basic systems, including through appropriate colegislation and administration.Still, successful widening of the fiscal space depends on a growing and stable middle class willing to bear the additional costs. At the global level, however, socioeconomic developments over the past two to three decades have contributed to an increasingly insecure middle class rather than the opposite (OECD, 2019). In Asia, limited evolution of the middle classes has been subject to what has been called “productivist welfare” (Holliday, 2000). Despite variations between countries and territories and while focusing on Japan, Hong Kong (China), Singapore, Republic of Korea and Taiwan (China), this model can be summarized as a “growth-oriented state and subordination of all aspects of state policy, including social policy, to economic and industrial relations” (Holliday, 2000), in which market outcomes are the determinants of welfare. Indeed, at least until the 1997–98 economic and financial crises,welfare policies in the region were dominated by a strong opposition to a State’s welfare redistributive policies and an emphasis on the role of the family.34 An intractable or extremely complex problem.30With only few exemptions, concerns relative to tax-financed, flat, universal basic old-age pensions have represented serious obstacles to their introduction in the ASEAN Member States and beyond.35 Countries have tried various approaches to strengthen voluntary participation of formal and informal sector workers in formal old-age pension and savings schemes. According to Holzmann, Hinz and Tuesta (2013):“The key challenge is to develop a design that will motivate lower-income groups– which require powerful, immediate and readily understood incentives to overcome their inherent consumption preference and liquidity constraints–to direct their limited resources to retirement income. Such a system must also be attractive to informal sector workers[…]while not increasing the incentives for workers to leave the formal sector.”While such design would probably meet much sympathy of the middle- and higher-income groups,successfully developing these schemes for low-income workers in the formal or informal sectors has met with limited success in many countries.36 These groups do not have sufficient income to participate in such schemes, and required subsidies would need to be very high, making the attempt costly,which would put the measure into question and make alternative policy options, such as demogrants, fiscally more attractive.Alternatively, an approach that has been adopted with mixed results in a growing number of countries is matching contributions.Matching contributionsMatching contributions are paid by the sponsor of a pension scheme on top of the contribution paid by the members of the scheme. Although the classical Bismarckian split of contributions into employee and employer portions can be regarded as a matching system, the term “matching contribution” is mainly used in the context of voluntary pension schemes, whereby the scheme member is not mandated to join and it is an incentive measure for workers to join and remain in the scheme.For a few decades, high-income countries have had an increased interest in this approach after public schemes’ benefit levels and coverage were reduced in response to ageing and the growing (so-called) implicit public debt burden. In such contexts, matching contribution systems were promoted to redress coverage gaps, both in terms of legal coverage rates and adequacy of benefit levels.In principle, matching contributions may be provided for public programmes or by the sponsors of private occupational plans and could be defined-contribution or defined-benefit systems. In practice,there appears to be a great variety in design approaches. For all systems, the matching design feature has the ultimate goal of increasing individual savings levels (Hinz et al., 2013).The attractiveness of the matching contribution system has been seen in its following two features (Hinz et al., 2013): A matching contribution system provides more tangible incentives for individuals to participate in pension funds than the more traditional approach of mandatory participation and provision of preferential income-tax treatment. This is especially so for low-income groups who do not pay income tax and for individuals who may not participate in the formal labour force and thus35 And indeed, the fate of the Beveridgean system in the United Kingdom after World War II gives reason to make careful decisions.36 Singapore might be considered an exception through its Workfare Income Supplement (see the country brief for more details about the programme), which has helped to improve coverage of its Central Provident Fund and incentivize less-educated Singaporeans, particularly in the older age groups, to stay in the workforce. Singapore is a city-state with modern, well-developed governance and information systems, which make it relatively easier to reach low-income, self-employed workers than in other countries.31receive no advantage from tax-based incentives. The close association of matching contributions with individual defined-contribution savings accounts, rather than defined-benefit schemes, more appropriately addresses those populations that these arrangements seek to reach, especially in lower-income groups or in developing countries, because they have irregular earning patterns over the life cycle or are predominantly engaged in the informal sector.At the same time, however, these two features highlight problems often related to matching contributions. First, they are costly, especially for people with low incomes,and thus raise the question of why public resources are not channeled into public-mandated defined-benefit schemes. Second, they potentially contribute to fostering informal structures as much as universal coverage schemes might. In other words, matching contributions could be seen as conserving existing social security structures rather than overcoming their deficiencies in order to make them more equitable and inclusive.Also, the administration and governance of matching-contributions and defined-contribution schemes, whether public or private, appear to be simple at a first glance. But while they might be lifting immediate responsibilities from government’s shoulders, they require complex systems of monitoring and control of financial markets that are kept under the perspective of securing workers’ entitlements. In the end, they might not only reintroduce government responsibility through the back-door but also unintentionally change financial markets’ operations and functionalities with economic implications, whether to the worse or to the better being an open issue.Social securitySome scholars on pension reform have recently stipulated growing concerns with respect to the future of classical social security solutions for the region through “[t]he marginalization or even the demise of pay-as-you-go, earnings-based (Bismarckian) public pensions in countries with a high level of informality”(see Holzmann, Hinz and Tuesta, 2013, p. 24). Although there are opposite views (Scholz, 2015), the logic of the argument cannot be ignored. One of the major barriers obstructing the expansion of coverage in many ASEAN Member States has been the ineffective enforcement of the labour and social security regulation and lack of compliance. This partly leads to the large share of informal employment in both the formal and informal sectors in some countries (box 2).Classical social security solutions for old age have shown, over many decades and in many countries worldwide,much resilience against the ups and downs of history and economies,not the least because of their inherent flexibility and immediate accessibility by legislation and government action. And they will behave similarly in the ASEAN region. These solutions have repeatedly proven to be important elements of the success stories of social policy (Scholz, 2015) and indeed one of the easier-to-handle social policy approaches. They have also proven to be valuable governance instruments of “last resort” whenever direct and immediate government action and responsibility were required and where private market solutions systemically failed in securing incomes for those in need. Naturally, these schemes must be developed further and possibly adapted in the context of policies increasingly relying on old-age demogrants-cum-matching contributions. In any case, matching contributions would hardly be more easily handled administratively than classical Bismarckian systems.32Box 2 Lack of enforcement and compliance as a major barrier to expanding coverage In many ASEAN Member States, labour and social protection inspection mechanisms are weak. In some countries, like the Lao People’s Democratic Republic, there is no systematic mechanism to enforce compliance, such as regular inspections or penalties. A recent ILO study on the inspection system in Indonesia and Viet Nam identified deficits in various aspects. In both countries, several agencies are involved in the inspection process. Weak coordination between the agencies leads to overlapping responsibilities and inefficiencies,resulting in low compliance among enterprises. Limited human, financial and administrative resources often result in few inspections. Information and communication technology systems are either nonexistent or ineffective to monitor and facilitate compliance and inspection processes, and the lack of well-trained personnel impedes proper enforcement of regulations in the informal and formal economies. Inspection mechanisms or processes are often not tailored to the situation of economic units and workers in sectors with a high likelihood of informality. In Cambodia, Indonesia and Viet Nam, inspectors with the social security agency do not have the authority to impose sanctions for noncompliance but need to refer to a higher authority for final decisions. This reduces compliance among businesses because they know the referral process takes a long while.Source: ILO, forthcomingb.3334Photo: @Réhahn Croquevielle3. OLD-AGE PENSIONPROBLEMS AND POLICY TRENDS IN THE ASEAN REGION35For the analysis of the pension policy landscape in the ASEAN Member States, it is important to remember that they are extremely heterogeneous in their economic, demographic, political and social structures. For instance, Singapore and Brunei Darussalam have the fourth- and fifth-highest GDP per capita in the world, respectively, while Cambodia, the Lao People’s Democratic Republic and Myanmar are among the countries with the lowest levels of economic development. Moreover, the level of institutional maturity, particularly when it comes to social security policies and institutions, is often at polar opposites.Almost all the ASEAN Member States have been characterized by high levels of economic growth, coupled with significant poverty reduction in the past few decades Despite this progress, 14 percent of the ASEAN population still lived on less than US$1.25 a day in 2015 (ASEAN, 2018). Inequalities within and across countries are substantial; and 4 percent of workers live in extreme poverty (less than US$1.90 per day), while 14 percent of workers live in moderate poverty (more than US$1.90 per day but less than US$3.20 per day) (ILO, 2018d).3.1. Overview of challenges affecting ASEAN countriesInformal and vulnerable employment in ASEANThe ASEAN region is associated with a high prevalence of workers in vulnerable employment (self-employed and unpaid family workers) who are usually characterized by informal working arrangements(ILO, 2016d). More than 149 million people, representing 47.5 percent of total employment, are still in vulnerable forms of employment. The situation is more aggravated in certain countries, with vulnerable employment presently affecting more than half of all workers in Cambodia, the Lao People’s Democratic Republic, Myanmar, Thailand and Viet Nam (table 3) (ILO, forthcomingb; ILO, 2018e).Table 3. Share of workers in vulnerable employment in the ASEAN Member StatesCountryWage or salaried workers (employees)EmployersOwn-account workers2010 2018 2010 2018 2010 2018Brunei Darussalam91.22.8Cambodia 29.8 48.8 0.2 0.45.7 50.4 32.9Indonesia 39.8 49.3 2.9 3.5Lao PDR15.7 19.5 0.5 0.6Malaysia Myanmar74.8 74.8 3.8 3.537.53.0Philippines 54.2 62.5 3.9 3.7Singapore 85.5 85.4 5.3 4.8Thailand44.5 49.8 2.7 2.9Viet Nam 33.7 43.4 3.4 2.039.5 34.4 49 51.7 17 17.6 35.530.1 27.7 8.6 9.3 31.8 31.4 43.3 39.5Contributing family workers2010 20180.319.417.917.812.934.828.34.34.224.011.86.10.60.52115.919.415.0Vulnerable employment201020186.069.850.857.347.383.880.021.321.859.541.933.89.29.852.847.362.754.5Source: ILO, 2018e; Ministry of Manpower (Singapore), n.d.36Despite the gradual reduction of the share of vulnerable and informal employment in total employment in the past few years (table 3), the presence of vulnerability will likely continue to be relatively high in most countries in Asia. Projections indicate that, in absolute terms, the numbers are likely to increase in the coming years, with 155 million people in ASEAN countries in vulnerable employment by 2021.Demographic changes will be an additional factor influencing the extent of social protection, particularly with the rapidly ageing populations and expected trends towards increasing urbanization. ASEAN regional integration processes (inside ASEAN and with other countries), as well as trade agreements, with their expected impact on labour mobility and movement of goods and services, will also have a substantive influence on the further development of social protection for Member States.A common central feature among most ASEAN Member States, important for any system analysis and pension policy recommendations, is the prevalence of informal employment. In the ASEAN region, 78 percent of the employed population engages in informal employment (78.4 percent for women and 77.7 percent for men, as shown in (table 4). Within the Asia and Pacific region, the South-Eastern Asia and Pacific region has the second-largest share of informal employment, at 75.2 percent, after Southern Asia.Excluding the agriculture sector, the share of informal employment decreases to 63.9 percent, which is still high. Individual countries have informal employment to varying degrees. According to latest estimates, the share in total informal employment ranges from around 30 percent in Brunei Darussalam to more than 90 percent in Cambodia and the Lao People’s Democratic Republic (ILO, 2018e).Table 4. Informal employment in ASEAN and subregions as a share of total employment, by sex, 2016Region or subregionASEAN Asia and Pacific non-ASEAN Eastern Asia South-Eastern Asia and Pacific Southern AsiaInformal employmentNon-agricultural informal employmentTotal Male FemaleTotalMaleFemale78.0 77.778.467.467.966.866.3 69.260.857.861.251.375.263.987.877.6Source: ILO, forthcomingb; ILO, 2018e. Figure 4. Composition of informal employment (including agriculture), by employment status, latest available year (%)Source: ILO, forthcomingb; ILO, 2018e. 37Among the 78 percent of employment that is informal in the ASEAN region, nearly 60 percent is in the informal sector, around 10 percent in the formal sector and nearly 8 percent in households. Among those in informal employment, own-account workers and employees make up the biggest groups, with 36 percent and 38 percent, respectively. Employers account for 2 percent of total informal employment, and (unpaid) contributing family workers represent 23 percent. The Asia-Pacific region is also characterized by huge gender gaps in employment, evidenced by the low female labour force participation rates found particularly in Southern Asia, where the gaps are among the widest in the world. In addition, women are more represented in vulnerable forms of work, particularly unpaid family work (ILO, 2016d). Ageing Asia An unprecedented demographic transition is taking place globally,37and population dynamics in Asia are prominent drivers of this process. In numerical terms, most of the global increase in the population aged 60 or older will take place in Asia (figure 5). Figure 5. Population aged 60 years or older, by region,1980–2050 (thousands)Source: ILO, 2014d. Moving from 962 million in 2017 to 2.1 billion38 in 2050, older persons (aged 60 or older) will constitute one out of every five persons worldwide (table 5), which will surpass the proportion of children and adolescents aged 10–24 years. In Asia, an increase of 549 million to 1.3 billion by 2050 is projected and, out of each world-average cohort of 100 older persons, around 61 will live in Asia by 2050. For ASEAN Member States, the nearly 64 million older persons in 2017 is projected to increase 2.6 times by 2050, reaching 168 million (DESA Population Division, 2017b).37 The number of people in the world aged 60 or older is set to increase from around 962 million in 2017 to more than 2 billion in 2050 – more than a doubling of the current population levels (DESA Population Division, 2017a). This number of older persons will represent 24.2 percent of the world population in 2050, up from 12.2 percent in 2017.38 Followed by Africa from 65 million to 220 million, Latin America from 71 million to 200 million and Oceania from 6.5 million to 13 million. In percentage terms, Africa will see a more than a tripling of the population aged 60 or older (340 percent), Latin America and the Caribbean (280 percent), followed by Asia (250 percent) and Oceania (200 percent).38Table 5. Number and percentage of persons aged 60 or older in ASEAN Member States, 2017 and 2050World Asia SouthEastern Asia Brunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet NamNumber of persons aged 60 or older in 2017 (in thousands)Number of persons aged 60 or older in 2050 (in thousands)Change between 2017 and 2050 in %Older persons Older personsin % of totalin % of totalpopulation, 2017 population, 20509622632080459116.212.721.35492461273175131.812.224.263973167877162.49.921.034154352.98.028.71129 22743434 3074 5043 8023 1115 11691 106163730 617291472 9647 11544 21417 2638 22954 32429230.47.116.9171.48.619.2239.26.316.1213.89.723.1128.99.418.5166.97.614.2136.619.540.193.316.935.1205.511.128.3Note: Data are based on UN population projections.Data source: DESA Population Division,2017b.The total dependency ratio [(number of persons aged 0–14 plus the number of persons aged 65 and older) divided by the number of persons aged 15–64] of Asia is expected to increase from 47.3 (2015) to 55.8 (2050), whereas the respective old-age ratio is expected to increase from 11.2 to 27.8 over the same period (DESA Population Division, 2017b).39For the region, the United Nations dependency rate estimates are as follows:Eastern AsiaTotal dependency rate2015:= 39.5 2050:= 69.8Old-age dependency rate 2015:= 15.5 2050:= 46.3Southern AsiaTotal dependency rate2015:= 53.6 2050:= 48.6Old-age dependency rate 2015:= 8.3 2050:= 19.7of which India:Total dependency rate2015: = 52.2 2050: = 47.7Old-age dependency rate South-Eastern Asia40 Total dependency rate2015:= 8.6 2050:= 19.8 2015:= 48.5 2050:= 53.8Old-age dependency rate 2015:= 8.8 2050:= 23.939 This implies an increase of the youth dependency ratio of 4.2 percentage points, the effect mainly stemming from India, which (like Africa) will be lagging in international demographic transition.40The estimates for the South-Eastern Asia region can be taken as representing ASEAN. All estimates represent the medium-variant forecasts. For details, see: https://population.un.org/wpp/Download/Standard/Population/, hereunder tables (i) total dependency ratio 1 and (ii) old-age dependency ratio 1.39One of the core characteristics of the ageing process in Asia is its pace, particularly when compared with countries that have already aged in other parts of the world. While countries like France or the United States needed, respectively, 115 and 69 years to move from an “ageing” phase to an “aged” situation,41 Japan needed only 25 years, and the Republic of Korea took 20 years (World Bank, 2016a). Similar paces are expected to occur in countries like Viet Nam and China, with respective ageing periods expected to pass 20 and 25 years. Even countries that still have younger demographic structures are expected to have fast-ageing processes,such as the Lao People’s Democratic Republic at 20 years and Myanmar at 25 years. The only relative exception is the Philippines, where this transition is expected to take 35 years (World Bank, 2016b).Fast-paced ageing is mostly the result of a general strong decline in fertility rates (table 6). In Asia, on average, the fertility rate declined from 5.6 children per woman in the 1950s to 2.1 in the late 2010s, which is a much sharper decline than the one observed for the global average. For South-Eastern Asia, the decline was even sharper, as seen in table 6, with average fertility per woman projected to decrease to less than one third of its initial value in the period between 1960 and 2050.Table 6. Fertility rates in ASEAN Member States since the 1950sTotal fertility (children per woman)1955–601985–902015–202045–50World Asia South-Eastern Asia Brunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet Nam4.93.42.52.25.63.52.11.96.13.62.31.96.93.51.91.77.06.02.52.05.73.42.31.96.06.32.61.86.43.72.01.76.03.82.21.87.34.52.92.26.31.71.31.46.12.31.51.66.23.92.01.9Source: Various UN population projections; DESA Population Division, 2017b.These observations are hardly novel, and many measures have been taken in the past to address the upcoming income (in) security problem of the rapidly growing older populations. Nonetheless, not all countries have pursued consistent and sustained old-age income security policies. With several late though important exceptions, expansionary pension policies typically have been regarded as detrimental to economic growth, which was deemed as needing to “come first” in such countries as the Republic of Korea. It is only recently that countries in Asia appear to be changing their social policy attitudes – while realizing that they might run out of time.4241 Defined as the number of years needed to increase the population aged 65 or older as a percentage of the total population from 7 percent to 14 percent.42 See, for example, the ASEAN Declaration on Strengthening Social Protection adopted at the 23rd ASEAN Summit in 2013.40Evidence from10 Eastern Asian countries signals a shift in peoples’ attitudes in relation to retirement and with regards to changes of perception towards the traditional role of families in securing oldage income security. Families still have an important role in retirement security in Asia,but this role has shrunk in recent years and is expected to shrink even further. Research shows that only a small minority of workers in countries like Malaysia, the Philippines, Thailand and Viet Nam believe that family support ideally would be the major source of retirement income. A majority of workers in a 2015 study were supportive of pension reforms, including increasing standard retirement age, and a more active role of the State in encouraging savings or longer work and savings periods among workers, even though they understood that this might represent individual sacrifice (Jackson and Peter, 2015).Indeed, the pace of ageing observed in ASEAN countries is, on average,faster, with the ageing phenomenon occurring at lower income levels than more mature, high-income countries had while establishing their welfare states, including old-age income security systems.43 In other words,the countries do not have the same long time span available during which to accumulate wealth and assets, as well as build social security institutions, before undergoing the ageing process.To what extent the ASEAN Member States’ pension policy options, including expansion of fiscal space, will be constrained by their current and foreseeable political, administrative and, importantly, economic capacities will be explored in the following sections of this report.Box 3 Statistical development in ASEAN Member States and social security policiesA word on statistics: ASEAN statistics on demographic, economic and labour market data have reached consistent levels of quality, although much improvement is still possible. To address common social policy goals, potentially with the aim of converging policies, the establishment of a common framework for statistical reporting on social spending (and revenue) –regular social budgeting plus core legislated rules – is due.Improvement of statistical reporting appears to be necessary because ASEAN should not make similar mistakes as, for example, the European Union, where social policy has for two to three decades been used by national actors as an instrument for cost competition between European Union Member States, and beyond, for example, reducing labour costs by reducing contribution rates. This makes no sense in ageing societies because they cannot avoid shifting more national resources to the older generation. Squaring the circle is not possible as long as decent societies are committed to remain civilized. Rather, to foster social policy in the region, ASEAN should seek to find agreement on certain common social policy goals. This could include harmonizing legal parameters (like retirement ages and their indexation) and the income replacement levels of social benefits (pensions especially), plus the harmonization of other social institutions.Harmonization of social policies on the basis of objective data would also be one of the suitable measures to counter social policy stalemate resulting from “ageing before being rich”. In other words, the implicit production costs of social spending could be harmonized, which would leave ASEAN Member States with a fair common platform on the basis of which to organize economic competition and income growth.43 For the past three to four decades, the term “welfare state” has lost much of its earlier meaning. The term should more adequately reflect realities and be replaced with the term “social investment state”. Under the notion of a welfare state, social benefits were considered the societal harvesting of the fruits of economic growth, whereas a social investment state now perceives social benefits as inputs into (investments in) people to enhance economic growth. For more detailed discussion, see ILO, 2018b.41Inequality in ASEAN countriesFruits of the unprecedented economic growth in the region have not been spread evenly across Asia’s citizens. Gini indices are, with few exceptions, higher in international standards, particularly when compared with most European countries or Canada. The same happens when they are compared with average indices in OECD countries (table7).44 In a few countries,statistical evidence shows distributional deterioration. For instance, for Indonesia, the Gini coefficient increased from 32.7 in 2004 to 38.1 in 2017. For a number of countries, no such inter-temporal comparison is possible. Among ASEAN Member States, Thailand and Malaysia are two examples in which policy efforts appear to have improved the situation (World Bank, 2018a).Table 7. Gini coefficients in ASEAN Member States and selected countries, latest available yearCountryASEAN MEMBER STATES Brunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet NamOTHER SELECT COUNTRIES China Fiji Japan Rep. of Korea New Zealand Sri Lanka France Germany United Kingdom United StatesCanadaOECD averageIndex valueYear……38.1201736.4201241.0201538.1201540.1201540.4201836.5201735.3201638.6201536.7201332.1200831.6201233.3201439.8201632.7201531.7201533.2201541.5201634.0201331.82014Source: World Bank, 2018a; Singapore Department of Statistics, n.d.; OECD, 2016.44 The higher the index, the higher the inequality. While strict country comparison is not possible due to differences in methodologies, the coefficients presented provide robust indication of countries’ overall situation regarding inequality.42Even when compared with other countries in the region, like Japan and the Republic of Korea, the ASEAN Member States’ values are still higher. Countries like Malaysia, Singapore and Philippines present values that are closer to those of the United States,a situation that calls for significant social policy action. The Lao People’s Democratic Republic and Viet Nam Gini coefficients are at an intermediate level, but each country’s average income level is still very low. In these countries, the minimal policy programme must be to equitably lift everyone’s incomes and avoid deterioration of people’s relative income positions. While improvements related to the Gini coefficient can only be achieved through a bundle of welfare-oriented measures, including labour remuneration and, more generally, societies’ formalization. Implementation of social security institutions must be pursued on a step-by-step basis, while making maximum use of the momentum of future growth.Poverty in ASEAN countriesThe 10 ASEAN Member States vary widely in their social policy achievements. Brunei Darussalam and Singapore, which have high GDP per capita and high labour productivity in regional as well as international comparison, have not considered it appropriate in the ASEAN context to measure poverty and publish poverty rates (table 8).While not much poverty might exist in the two countries in absolute terms, they obviously do not consider the concept of relative poverty applicable, as is standard in the European Union and other regions, for measuring the development of the degree of social inclusion of the poor strata of the population over time45 or of persons at the bottom of the income and wealth distribution. Only Cambodia, Indonesia, Thailand and Viet Nam publish poverty rates on a regular basis, while the Lao People’s Democratic Republic, Malaysia, Myanmar and the Philippines provide such information to ASEAN only on an irregular basis, as seen in table 8. For the Philippines, it takes place at three-year intervals.Table 8. Population below the national poverty line in ASEAN Member States, 2006–15 (% of total population)Country20062007 2008 2009 2010 2011 2012 2013 2014 2015Brunei Darussalam CambodiaIndonesian.a.n.a.- 30.1 17.8 16.6n.a.n.a.29.9 22.9 15.4 14.2n.a.n.a.n.a.n.a. n.a.n.a.21.119.818.9 16.3-13.513.312.512.0 11.4 11.311.2Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet Nam---3.632.1-26.6-n.a.n.a.23.4 20.915.5 14.827.6--3.8--- 26.3n.a.n.a.20.5 19.114.5-24.0-- 23.2-24.0--1.7--0.625.623.6------25.2--21.6n.a.n.a.n.a.n.a. n.a.n.a.16.913.2---7.214.212.611.19.88.47.0Note: n.a. = not applicable; - = not available. The figures for different countries should mainly be read over time, to be interpreted in their development because countries use different poverty definitions.Data source: ASEAN, 2017.45 Poverty is a multidimensional social phenomenon, and its measurement is difficult and must make statistical shortcuts. Concepts of measuring relative income poverty, such as individual (household) poverty related to a predefined societal average, always create positive poverty rates, independent of the referenced average income level achieved. Relative poverty is a measure to reflect societal coherence as developing over time.43Most poverty rates in ASEAN Member States show a significantly falling tendency. While strict country comparison is not possible due to methodological differences between countries in calculating poverty, it is still possible to see the poverty trends over time. Many countries, such as Cambodia, Indonesia, Malaysia, Thailand and Viet Nam, have falling poverty rates up to 2015. Others, such as the Lao People’s Democratic Republic and the Philippines, have remained at a similar level. The use of World Bank poverty lines, such as US$1.90, shows a similar trend. Overall, poverty rates are only one of the indicators used for measuring the level of success and degree of comprehensiveness of social policy in a country. Many other indicators must usually be taken into account to get a full picture. These are distribution of GDP on labour (labour income), distribution of GDP on capital (capital income), profits and their taxation, comprehensiveness and quality of education, social housing and property ownership, to name a few. 3.2 Old-age pension systems under construction in ASEAN countries: Achievements, deficiencies and incompatibilities Only about 30 percent of Asia’s total working-age population is covered under public or private pension schemes, such as paying contributions. And only around 55 percent of persons older than retirement age receives some form of regular or one-time lump-sum pension payment (ILO, 2017d). This number hides significant disparities, with several countries having no gaps in coverage (China, Japan) and others having coverage below 10 percent (Cambodia, the Lao People’s Democratic Republic). Among the ASEAN Member States, the percentage is smaller, with 31.5 percent of persons aged 60 or older covered on average. In relation to other regions, average coverage in South-Eastern Asia is the second lowest and ahead only of Southern Asia(at 23.6 percent). Eastern Asia and Oceania have higher coverage levels, at 77.3 percent and 74.1 percent, respectively (ILO, 2017d). Figure 6. Percentage of population older than the statutory pensionable age in Asia and the Pacific receiving an old-age pension, latest available yearNote: Statutory pensionable ages for all ASEAN Member States can be viewed in table A2 in Annex I.Source: ILO, 2017d. In a context in which (i) an extremely fast demographic transition occurs in parallel with unprecedented economic and societal transformation; (ii) the size and role of families shrinks; and (iii) the process of urbanization and industrialization changes the roles of families and traditional institutions of community solidarity (such as the religious institutions), the time of retirement is becoming a time of anxiety for many people in ASEAN Member States (Jackson and Peter, 2015). For example, the survey (Jackson and Peter, 2015) found that a large proportion of workers worry about 44being poor and in need of money. This proportion goes higher than 80 percent of those surveyed in such countries as Indonesia, Republic of Korea, the Philippines and Viet Nam.With public pension systems not yet having the dimension and adequacy levels required to provide retirement income security, the pressure on families is expected to grow. With lower fertility rates come a smaller number of adult children in a position to bear the costs of securing their aged parents. Even half a century ago, a woman in the Philippines or Cambodia would give birth to nearly seven children, now the number is down to three. Today, a couple just starting their working careers in China, for example, will possibly have four older parents as dependants. This situation of insecurity is reflected in what Miller (1981) and Brody (1990) had earlier denominated as the “sandwich generation”, in which young adults are pressed between the need to provide support to their children and their parents.This situation, unless adequately addressed by macroeconomic and macro-social policies, will most likely prevent sustained further development and expansion towards a secure middle-class. The resulting social insecurity is compounded by persistent poverty of various scopes, such as among children, women, rural inhabitants, people with disabilities and especially older persons, all acting as a threat of potential socioeconomic failure to the members of the still-too-small middle-classes in many countries in the region.In an atmosphere of economic insecurity, societies have difficulty in concentrating their energies and efforts on solutions to the most urgent problems, chief among them the pressing environmental problems.Due to young persons entering the labour market, it should be expected in theory that old-age pension coverage rates of the labour force will gradually increase over time; but persistent informal sector inertia has thus far prevented significant progress, with only a few exceptions, such as China. Policy action towards establishing or strengthening social security institutions is needed for Asia’s population to have sufficient financial resources in their old age. This can ultimately contribute to formalization of ASEAN Member States’ societies.Transforming these findings proactively into redistributive social policies, including substantial universal old-age income security systems, would, most probably, positively affect the region’s economic future (OECD, 2012a; see Chapter 5 for more information) and support Asia’s societal, institutional and political long-term stability. Such policy orientation would clearly bolster countries’ ability to contribute proactively to their international obligations, such as those related to the environment and climate change. The United Nations has urged governments to “design innovative policies specifically targeted to the needs of older persons, including those addressing housing, employment, health care, social protection and other forms of intergenerational support” (DESA Population Division, 2015, p. 1).While embracing existent or initiating new, proactive social security policies is due, it is to be conceded and to be taken into account as a condition to such new policies, that many institutional arrangements already exist. Indeed, all ASEAN Member States have legislation that addresses all or a set of standard social policy areas as defined by ILO Convention No. 102, and almost all have pension legislation(see table A1 in Annex I, column “old age”).These findings must not be over-interpreted, however. In many cases,they go along with only limited coverage – partially as a result of high informality of labour markets because of limited public administration,lack of a “social security culture” and a general mistrust in the functioning of public45institutions. In parallel, too-low benefits, no or only irregular or ad hoc indexation and generally low average per capita incomes limit the fiscal space available for effective social and pension policies (ILO, 2014d). This is also expressed by low levels of tax or GDP ratios due to political choice, including low allocations to social expenditure.Legal old-age pension coverage in ASEAN Member StatesLegal coverage of pension entitlements (box 4) varies widely among ASEAN Member States. The higher rates were found for Brunei Darussalam, Malaysia, the Philippines, Singapore and Thailand (ILO, 2017d, table B9), all with an estimated 100 percent of legal coverage. Even countries like the Lao People’s Democratic Republic (at 80.5 percent) and Indonesia (at 69.7 percent) present a high rate of legal coverage. On the other side of the spectrum are Cambodia and Myanmar.Box 4 Pension coverageAchieving full pension coverage and adequate levels of old-age income is a core policy goal for nearly all countries. Maintaining coverage is an equally challenging and permanent task. As documented by the ILO (2017), legal coverage has advanced with development and growth in income, and there is a strong relationship between the level of per capita income and individuals’ participation in formal pension systems.Legal coverage, however, is not always synonymous with mandated participation, and growing average incomes are, by far, not sufficient incentive for all members of society to join or to be able to join pension schemes, especially contributory systems. For these and other reasons, there remain considerable differences among countries at similar levels of development in pension coverage and in the way in which participation in pension systems has evolved in different settings. The differences in experiences and outcomes indicate that context, the design of the system and the path of its development have a central role in the dynamics of pension coverage and benefit levels (Hinz et al., 2013).Since the beginning of the century and accelerated after the Great Financial Crisis of 2008, countries that had long achieved full coverage became at risk of growing problems in controlling their earlier high coverage rates and in maintaining adequate income replacement ratios in old age (Vaughan-Whitehead, 2015).These trends show the limits of pension policies resting predominantly upon “strengthening incentives” for individuals to join private or public pension schemes. It is obvious that such policies run a high risk of failure in countries with labour markets characterized by low wages or large sectors of labour markets with low wages. Therefore, they must be complemented or, where applicable, even be replaced with strong government-supported and strictly administered public schemes, which can be contribution based or tax financed (or a combination of both), depending on prevailing formal versus informal labour market structures, including – especially – wage levels and wage distribution.Hinz et al. (2013) argued that in “higher-income countries, coverage levels under mandated schemes remain high, but fiscal pressures caused by the generous benefits provided to early cohorts and exacerbated by rapidly falling fertility rates are now imposing the need46for reductions in future benefit levels. Such reductions will require a substantial increase in supplementary retirement savings if income replacement rates are to be maintained.”However, when incomes of the middle-class stagnate and job precariousness increases (as experienced over the past four decades or so in most higher-income countries and as expected to continue broadly in the future), policies aiming solely at increasing retirement savings will fail to solve the upcoming old-age income crisis. What is necessary in addition is substantial support by pension system reforms aiming at strengthening a basic (tax-financed) pension layer that is sufficient to secure older peoples’ freedom from want. This clearly goes beyond mere social assistance-type or subsidy solutions to other income-types.People in South-Eastern Asia and Eastern Asia appear to have a clear perception of the upcoming old-age income problems in the region: The combination of strong past income growth (although widely bypassing many persons) with the fast-falling fertility rates have left many of the people who made that strong growth possible with insufficient income in their old age.Source: ILO, 2017d; Jackson and Peter, 2015; Vaughan-Whitehead, 2015; Hinz et al., 2013.Legal coverage rates very much depend on applied methodology and quality of available data. Generally, there is agreement that legal coverage rates can only be indicative where high rates imply that legislation is designed such as to incorporate “all” (such as universal coverage) and a low or no rates indicate discrimination.Table 9. Active contributors to an old-age pension scheme in the ASEAN Member States (as share of working-age population and labour force), by sex, latest available yearCountryYear Share of active contributors in the working-age population (15–64 years)Share of active contributors in the labour force (aged 15+)Brunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet Nam…2010 2015 2010 2010… 2015 2015 2015 2015TotalMale……0.00.07.6…1.3…28.132.4……21.4…48.1…33.6…20.6…FemaleTotalMaleFemale…………0.00.00.00.0…10.5………1.6……23.643.239.350.2……………30.9………61.7………31.9………23.5Source: ILO, 2017d.47Old-age pension models and their adequacy In terms of type of old-age pension programmes, most ASEAN Member States pursue a social insurance approach, often in combination with a provident fund and/or non-contributory pensions,such as universal or social assistance (figure 7). From the perspective of long-term financial and social sustainability of pension schemes, legal retirement ages and contribution rates are generally considered low, with a few exceptions, indicating a need for reforms. For provident funds, the contribution rates seem generally to be significantly higher. Whether they are sufficient to provide decent pensions at retirement is an open issue, especially in times of worldwide low interest rates (see the country briefs). Figure 7. Overview of old-age pension schemes in ASEAN Member States, by type of scheme, 2015 or latest available yearNote: All ASEAN Member States have opted for public contributory pension schemes, complemented by other types of schemes (or are en route to developing them). Cambodia and the Lao People’s Democratic Republic have contributory schemes only, although Cambodia’s scheme is currently operating only for public sector employees. Indonesia, Malaysia, the Philippines, Singapore and Viet Nam have a combination of contributory and non-contributory means-tested schemes. Thailand has a combination of contributory and non-contributory pension-tested schemes. Brunei Darussalam and Myanmar have a combination of contributory and non-contributory universal schemes. Myanmar currently operates a defined-benefit pension only for public sector employees and a new social pension categorically targeted at people aged 85 and older, with plans to expand it to other age groups over time. The country plans to extend pension coverage to currently uncovered people, like private sector workers, in the near future.Source: ILO, 2017d. Effective coverage and adequacy Most important is that all older-aged persons have sufficient income at their disposal for a decent living. Formal public pensions are considered a major component of total old-age income. In ASEAN Member States, the share of persons receiving a pension (table 10) varies significantly. For Brunei Darussalam and Thailand, it is high, with four in every five older persons covered. Singapore coverage is close to three-fourths, while for Malaysia, the Philippines and Viet Nam, coverage rates range between one fifth and one third of older persons. Malaysia and Indonesia have coverage below 20 percent, and in Cambodia and the Lao People’s Democratic Republic, effective coverage remains below 6 percent. All ASEAN Member States, in spite of coverage levels, have considerably varying income replacement ratios. 48Table 10. Old-age pensions beneficiaries in ASEAN Member States, latest available year (as % of reference population)CountryYearTotal MenWom- Contribu-entoryNon-contribu- Reference popula-torytion (men&women)Brunei Darussalam 2011 81.7………81.760+Cambodia20153.2…………55+Indonesia2015 14.0…………56+Lao PDR20105.6…………60+/55+Malaysia2010 19.8……16.23.655+Myanmar*…………………Philippines2015 39.8 53.2 29.021.917.960+Singapore2018 76.8……48.927.965+Thailand2016 83.0……8.274.855+Viet Nam2015 39.9…………60+ / 55+Note: The differentiation of “contributory” and “non-contributory” must be taken with care because the underlying sets have been assumed disjointed, which might normally not be the case (whenever contributory and non-contributory pensions accumulate). *=There is no available data for Myanmar. Because only civil servants are entitled to pensions and the tax-funded (non-contributory) scheme covers only persons older than 90, one might conclude that the coverage rate of Myanmar would stay within the group of countries with the lowest coverage among ASEAN Member States.Data source: ILO, 2017d.Countries that have reached broad coverage levels typically have established tax-funded schemes or social pensions to extend coverage rapidly to population groups with low contributory capacity. Again, documented statistical rates are only indicative. Details depend on the quality of pension and demographic data available as well as on the type and degree of maturity of pension systems. Young schemes, for example, which require time-bound accumulation of entitlements and stipulate a formal retirement age legally, might not be able to pay the pensions. These schemes are still maturing and will only later reach out to a maximum of beneficiaries.A full picture of old-age pension coverage is only possible when pension-benefit (income) replacement rates are also taken into account. For these to be calculated, a stable methodological and statistical framework is necessary, usually requiring a longer (year-long) preparatory phase of establishing the binding legal rules,the reference incomes before retirement to be used and other information. These works are especially necessary for international comparisons, where a common framework – including a “common currency denominator” –must be agreed upon to avoid the comparisons becoming meaningless. Such preparation was, unfortunately, not possible for this report. Thus, the replacement rates quoted here can only be of an indicative nature.46 Establishing an informative reporting system on pension policies as a first step for a wider reporting system on social policies at large remains an urgent task for all ASEAN Member States (and beyond), where each Member State should be able to take advantage of existing cooperative structures (see Chapter 5).For a few ASEAN Member States, effective coverage rates below 100 percent do not imply that persons not covered by public pensions have no old-age income at their disposal. In Brunei Darussalam and Singapore, for instance, those without a public old-age pension can also be covered,such as selfemployed persons who may or may not have run a business throughout their active age and are now46 The discussion on replacement rates provided by provident funds may be taken as an indicator for the dependency of results on (more or less) complex assumptions. See section 3.2.49receiving a pension from that business, from accumulated private savings or from their property. Some may even choose to continue working and earning. In low-income ASEAN Member States, however, it can be assumed that many uncovered older persons are forced to continue working into their old age and must depend on intra-family transfers and support or simply live in poverty. For the Philippines, Thailand and Viet Nam, a large proportion of older persons – or 40 percent –continues to work after age 60 (Knodel et al., 2015). In addition, workers are mostly working in agriculture or informally, which is a further indicator of their vulnerability when reaching old age.For the time being, the role of family support in old age remains significant in ASEAN Member States. In Thailand, 79 percent of older people and 67 percent in Viet Nam, for instance, reported receiving income from their family (HAI, 2016). According to the Global Aging Institute,47 40 percent of retirees in the Philippines depend financially on their grown children (they are net recipients of income from children). Within the ASEAN Member States covered by the Global Aging Institute research, the Philippines is followed by Thailand (at 32 percent), Singapore (at 29 percent), Viet Nam (at 27 percent), Indonesia (at 25 percent) and Malaysia (at 20 percent). After comparing the results of the various surveys, the Institute researchers concluded that the role of the family is in decline. For example, financial dependency of elders on their children declined from 28 percent to 20 percent in Malaysia and from 39 percent to 29 percent in Singapore (Jackson and Peter, 2015).Nevertheless, the need of older persons to rely on intra-family transfers is considered too high; plus, those transfers broadly appear to be considered inadequate or insufficient. Surveys in the Philippines, Thailand and Viet Nam found that only around a third of older persons consider transfers from children to be their main retirement income source, with work and social protection collectively taking a bigger role. In terms of the scale of support, most older people in Thailand receive transfers from their children that are below the national poverty line (HAI, 2016). In the Philippines, half of the retirees in a HelpAge International study (2016) endured “some” or “considerable” difficulty in managing their expenses, despite financial support from their children. An obvious reason for the inadequacy of intra-family support is the high level of poverty and economic vulnerability that the surveyed populations endure, which reveals that if the “active” family is poor, its older members typically are poor, too.48Support from spouses tends to constitute only a small share of income for older people, and if available, this support varies with gender. While most older men were married, between 43 and 67 percent of older women in the five surveyed countries were widowed.Data on public spending on pensions support the view that none of the ASEAN Member States have sufficiently high effective income replacement rates can, on average, (table 11, column “old age”) to solve or reduce the problem of older persons’ dependency on intra-family transfers or other sources of additional income throughout old age. Viet Nam is the top pensions-spender of the countries included in this report, with 5.5 percent of GDP being spent on old-age programmes, followed by Thailand, at 2.2 percent, which suffers from too low effective old-age pension coverage in combination with generally too low benefit replacement rates.47 The survey was conducted in 2014 and administered to representative samples of workers and retirees in Indonesia, Malaysia, the Philippines, Singapore, Thailand and Viet Nam.48 One can turn around the argument: In these societies, broadly, when the parents are poor, their children typically appear to be poor. This points towards suboptimal education systems and a low upward social permeability. There appear to be strong barriers to poor people moving into the middle class. This problem, however,was not subject matter for this report.50Table 11. Public social protection expenditure, by branch in the ASEAN Member States, latest available year (% of GDP)CountryYear*Total HealthSickness,Childrenmaternity,employmentinjury, disabilityOld-ageEmployment Generalandsocialunemployment assistanceBrunei Darussalam20112.32.0Cambodia 2009-20111.81.3Indonesia 20151.10.1Lao PDR 20131.21.0Malaysia 2012Myanmar 2011Philippines 2015Singapore Thailand2015 2015Viet Nam 20153.82.91.00.32.21.64.23.53.71.26.30.8………0.00.10.20.1 0.7(2010)1.00.1 (2010) 0.0(2010)0.20.10.00.90.10.0 0.7(2014)0.20.10.60.90.00.71.2 0.5(2011)2.20.3 (2010) 0.0(2010)5.5……0.10.2… 0.8 (2013)…0.10.0 (2013) 0.4 (2013)…0.00.0 0.5 (2013) 0.3 (2013) 0.7 (2013)0.1 (2011)0.10.10.3Note: The figures in this table are only of an indicative nature. The reason that the “totals” do not add up to 100 is that none of the countries included in this report runs a systematic statistical social budgeting programme as, for example, proposed by Scholz et al. (2000) and as run on a regular basis by the European Union, the EU Member States and the OECD. Therefore, in its 2017 World Social Protection Report, the ILO used all accessible information from international and national institutions (IMF, ADB,government ministries and statistical offices) to provide maximum information to readers, at the expense, however, of that information being partially inconsistent. The contents of the table are based on information available in 2015–16. Allocation on programmes is based on ILO staff estimates. *= year of data unless indicated in cells in other columns.Data source: ILO, 2017d.In summing up, in countries where public pension system reforms are due, core elements of reform will have to include three main measures: (i) increase of retirement age; (ii) increase of resources,such as contribution rates and public subsidies; and (iii) increase of benefits’ income-replacement rates. These measures will have to be complemented by substantial investment in pension administration, such as staffing, information technology, general service orientation and legal enforcement of the existing provision, such as inspection and registry services.Provident funds A provident fund49 is a form of institutionalized mandatory individual savings account whereby each participating individual has an account in which deposits and returns accumulate on an individual basis. Supervised and often run by the State, provident funds are intended to pursue social security purposes. Payments into provident funds are made from monthly wages,through payroll deductions, where a maximum income ceiling can apply and employers can be made to contribute as well. The contributions can be exempt from taxes, in which case, the benefits in payment are usually taxed. The administrative costs of running the provident fund scheme can be sponsored by the State out of general taxation. A provident fund falls under the defined-contribution type of social security system, not defined benefit. Withdrawals from a provident fund’s individual accounts can serve various purposes, including purchase of retirement annuities, housing, education and similar (Sherraden, 1997). Individuals at retirement age can only expect to receive (either as a lump-sum49 This section takes advantage of Karunanethy, 2017.51pay-out or monthly pension payment, or a combination of both) what they and their employers had contributed to the fund over their active lifetime, plus any accrued returns on those savings or investments purchased with the savings.50In sum, there is no guarantee that provident fund participants will receive sufficient income to prevent poverty in their old age.Provident funds are usually not considered social security. At best, they might be judged as providing social security of a secondary nature or only in combination with other approaches. Institutions like the ILO, the International Social Security Association (ISSA) and the OECD have for years taken skeptical positions of the social rationale of stand-alone provident funds (Sherraden, 1997). After the marketization revolution in social security thought (Dixon, 1999), these viewpoints had changed in many ways. For the ILO, however, they remain broadly unchanged. This is due to both the recent scandal surrounding a major Eastern Asia provident fund but more fundamentally because of the negative social results of earlier social security reforms in many countries around the globe that aimed at more funding and were pro-financial markets (thus proving earlier professional concerns, of which many stakeholders have become increasingly aware).In the wave of de-colonialization that took place after World War II,Great Britain implemented provident funds in many of its colonies, whereas at home, the Labour Government introduced a Beveridge-inspired social insurance pension. Provident fund memberships were offered to government employees and formally employed wage earners. The original aims were to provide only for old-age income, disability and death (survivorship of dependants). The most famous current example is Singapore’s Central Provident Fund, founded in 1953, which became a model to many other countries (Sherraden, 1997).The financial set-up of a provident fund (fully funded by its members) is often considered as posing no risk to the financial solvency of a government and is thus expected to improve a country’s financial and economic stability. Not surprisingly, these features, including comparatively simple administration, were appealing to the British colonial authorities. They continued to remain so to the governments of the countries following their independence from Britain as well as to contemporary financial market actors.Indeed, many stakeholders in developing countries see them as useful instruments for other economic, political and social gains. Among the ASEAN Member States, Malaysia and Singapore operate provident funds. Countries in ASEAN and worldwide have more or less successfully used their provident funds: To (i) accumulate assets and capital to drive economic growth; (ii) moderate inflation;(iii) generate real income growth by increasing contributions during times of rapid expansion or, conversely, during recessions; to (iv) improve employment through labour costs cuts by decreasing employer contributions; (v) finance government investments;51 (vi) accumulate foreign reserves; or (vii) increase rates of home ownership (Sherraden, 1997). As an incentive to increase labour market participation and productivity due to the strict link between earnings and contributions. In this respect,provident funds share similarities with classical Bismarckian systems. In the World Bank’s Five-Pillar Framework,provident funds come under Pillar 2: A defined-contribution scheme, such as individual savings accounts with flexible options (Holzmann et al., 2005).50 P rovident funds are, by their design, financial intermediaries. Usually, however, they are not banks because they arenot licensed to create fresh money through loans to enterprises and private households. 51 Most provident funds buy and hold government securities or shares of government-owned companies.52 To allow workers to tap their mandatory savings to cover spells of unemployment52 and thus reduce the risk of moral hazard often associated with traditional unemployment insurance,as practised, for example, in Fiji and some Latin American countries. To implicitly strengthen traditional social values of continued reliance on family, kinship and communal ties rather than substituting them with publicly provided formal social insurance that could erode such social structures.Without doubt, Singapore’s Central Provident Fund first had a significant role as a financial intermediary in its rise as an economic powerhouse. Later, it increasingly fulfilled its intended social financial obligations to its members, as stipulated. Nevertheless, international organizations, such as ISSA and the ILO, consider provident funds as transitional, at best, and as simplistic forms of social security, and they advocate for countries to transform them into social insurance schemes (Sherraden 1997; ISSA Committee on Provident Funds, various meetings53) or, at least, to bolster them with underlying tax-financed forms of universal benefits (see Chapter 5). The reason is that provident funds face several issues that question their capacity to meet social security minimum standards as specified in the international norms systems of the ILO or the United Nations and when countries feature less reliable governance than in Singapore’s modern urbanity.54 Some of the issues involving provident funds worldwide include (Sherraden, 1997): There is no pooling of risks: individuals basically self-insure themselves through their own savings. There is no redistribution, neither inter- nor intra-generational.55 Coverage is often limited to government employees and people formally employed, so theinformally employed and people living in the subsistence sector are usually left out. Returns on savings may be lower than inflation, especially in countries where financialmarkets are not well developed or stable, which erodes the real value of savings and pension payments because there is no indexation either to inflation or wages.56 Low-income earners are unlikely to generate sufficient income to meet basic consumption needs in retirement. Where members are allowed to personally invest their savings, this only feigns maximization of individual utility as a result of freedom of choice and independence because institutional investors, such as insurance companies, pensions and sovereign wealth funds, are normally able to obtain higher returns at lower risks than individual retail investors could. They might turn socially inefficient when beneficiaries, unused to managing large sums of money, undertake imprudent financial decisions and effectively squander their old-age savings.52 This specific feature is not applicable in Singapore. 53 Later ISSA meetings on provident funds switched their focus to ensuring that they are well managed. This has, how-ever, other than insinuated by Sherraden (1997), not changed the organization’s principal skepticism relative to the social security capacities of provident funds. 54 For reference, the reader is advised to visit the Central Provident Fund website. 55 Singapore employs (to a certain extent) the principles of risk pooling (through the national annuity scheme, the Central Provident Fund’s LIFE and through health insurance premiums paid towards MediShield Life) and income redistribution (through the Workfare Income Supplement to add on to the Central Provident Fund contributions of lower wage earners by the Government). 56To counter this problem, provident funds have implemented progressive interest rate structures, with lower-balance members and older members getting a higher interest rate than other members (such as the Singaporean Central Provident Fund). Such measures also introduce socially welcome redistributive elements to provident fund schemes.53In practice, these principal problems often materialize or are aggravated by the fact that provident funds or respective governments fail to stipulate automatic annuitization57 of accumulated funds at retirement. This omission may have been justified in the early days of provident funds because no sufficient finance-mathematical and administrative-statistical expertise was readily available that could undertake the calculations on a mass basis. Such limitations hardly exist anymore, and a minimum of financial literacy of provident fund administration would allow for an appropriate application of standard software programmes. Therefore, most members of many provident funds are nowadays trapped by “big money” illusions, whereby they insist on the full pay-out of their accumulated funds. Most often, after years of contributions and for various reasons,provident fund members want to have the result of their long-lasting savings efforts at their disposal.This practice is well entrenched in most provident fund members’ behaviours, making it difficult for governments to change it. Empirical evidence (EPF,2015b) indicates that many retirees spend the paid-out lump-sum amounts too quickly or re-invest the monies inefficiently and/or in a costly manner and thus run out of their reserves well before death – necessitating they turn to society for help. Other than in individual cases of long survival after retirement,where it might be impossible to stretch the paid-out lump-sum meaningfully over the remaining lifetime, this behaviour holds more often than not.These and other factors combined may result in old-age poverty, despite accumulated savings, which has to be countered by society at large either directly through the provision of taxfinanced social assistance-type benefits or indirectly through private donations, such as charity or community help, with all possible unpleasant implications (for donors and beneficiaries alike).Annuitization of pension funds or savings implies that members take advantage of insurance principles, whereby even longevity “beyond limits” would be covered, and no one would be endangered of running out of their monthly pension payment. Of course, annuitization also implies that “big amounts” accumulated may turn into only “tiny monthly pensions”. The level of pension payments after annuitization would depend mainly on (i) the assumed life-expectancy (mortality table) of a scheme member at retirement and (ii) the assumed interest rate. In the event of annuity indexation (in line with consumer price indices), the initial payments might even be lower.To overcome provident fund members’ resistance to annuitization or to improve their management capacity of a full pay-out of the accumulated amount, governments (provident fund administrations) in many countries have implemented programmes to improve people’s financial literacy – thus far with only limited success. This is not surprising, considering finance mathematics remains a difficult field.Acknowledging original design gaps and challenges in providing income security in old age, some provident funds started introducing features in their system that, to a certain extent, added principles of risk pooling and income redistribution. Additionally, some countries added mechanisms that protect their members from potential negative impact of significant fluctuations in the financial markets. Singapore has been one of the front runners in this regard with the integration of several innovations towards improving income protection (box 5).57 A nnuitization refers to the conversion of an amount into a series of periodic payments, either for a specific period of time or for life. From a rational social policy perspective, annuitization is clearly preferable to paying out lump-sums. In Asia, with the “productivist” welfare state concept prevailing, paying out lump sums may be considered an opportunity for the pensioner to invest the monies in productive projects and potentially contributing to economic growth.54Box 5 Central Provident Fund of SingaporeThe Central Provident Fund (CPF) aims to ensure financial security in old age by providing lifelong income and addressing housing and healthcare needs. It is a defined-contribution savings scheme that is complemented with the CPFLifelong Income for the Elderly (LIFE) and Workfare Income Supplement schemes as well as government subsidies in housing and healthcare, top-ups and other tax-funded social transfers targeted at people in need of support.The Central Provident Fund employs (to a certain extent) the principles of risk pooling (through the pooling of interest income into a common fund as well as for health insurance contributions paid towards MediShield Life) and income redistribution (through the Workfare Income Supplement to add on to the contributions of low-income earners by the Government). CPF LIFE is a mandatory scheme that provides lifelong retirement income. Central Provident Fund members can use their savings to pay for premiums for MediShield Life, which is a universal healthcare insurance scheme that incorporates risk-pooling for healthcare needs.The Central Provident Fund has a progressive interest rate structure, with lower-balance members and older members getting a higher effective interest rate. The base interest rate on monies in the Retirement Account is 4 percent, but an extra 1 percent of interest is given on the first 60,000 Singaporean dollars (SGD) of balances, and an additional extra 1 percent is given on the first SGD30,000 for members aged 55 or older. All interest is stable and risk-free and consistently above returns for market instruments of a comparable duration.In the Singapore system, members can receive up to 6 percent interest on their savings balance, with higher interest on lower balances of older members. The legislated floor interest rate in the CPF Ordinary Account is 2.5 percent, while the current minimum interest rate for monies in the Special, Retirement and MediSave Accounts is 4 percent. Both are above long-term inflation rates of approximately 2 percent.Asia: Aged before rich? The issue of pension affordabilityAsia might well be ageing before it becomes rich. The explicit assumption is that pension savings are needed as a requisite for economic growth, stimulated through investments to be financed out of personal savings. The implicit concern is that neither now nor in the future can sufficient resources (public or private) be made available for financing adequate old-age incomes.58The accumulation of savings,however, is no requisite for state and enterprise investments. Contrary to popular belief, banks do not need private savings for handing out loans to investors. Rather, banks are able to create money out of nothing, which is the very core of bank operations (Deutsche Bundesbank, 2017; Werner, 2014).59 The financial innovation that stood at the beginning of this process triggered, historically, the explosion of capitalism in the United Kingdom in the late eighteenth century.Even where governments realize that any accumulated pension reserves will be depleted or where, as a result of ageing, personal savings accumulated in private accounts might be at risk of declining, this does not imply that no monies for financing growth will be available in the future. In such situations, any prominent role of financial intermediaries can, and must, be replaced by (re-)fostering instrumentation of banks in their role as money creators.6058 See the Old-Age Pensions in Asia report for further discussion (ILO, forthcoming). 59 It is increasingly understood in the literature that private banks, by handing out loans, create money, including for in-vestment. Banks do not need savings to finance investment. In this respect, many textbooks in economics are flawed – thus stimulating the myth of savings as a precondition for investment. 60 The famous macroeconomic equation I = S (investments = savings) only holds expost, not exante. Actually, it is an expost accounting identity usually completely misinterpreted even by economists.55That savings are not a requisite for growth but rather its result explains why many of the growthmiracle countries could run their social security systems with the pay-as-you-go approach while registering record growth rates, be it pre-World World I United Kingdom, United States, Germany, France and others or post-World WarIIGermany and in Europe at large, Japan, the Republic of Korea and, most impressively,in China recently. All these countries accumulated high individual and partially institutional savings – but as a result of growth, not as requisite to it. Thus, ageing societies are not doomed to run out of monies required for financing growth-stimulating investments. And while the role of provident funds might decline in Asia, the role of banks might have to be restrengthened.Such reshuffled policy mix will be necessary to improve labour productivity. While it is evident that better education and training is an indispensable ingredient to enhancing labour productivity,it will not be sufficient to maintain or re-accelerate economic growth: It must be co-supported by public and private investments. Both of these can be financed through efficient and effective banks.61This policy mix is a precondition for strengthening labour productivity in the short run at the enterprise level as well as in the long run,through the widening of the available fiscal space for public education, health and work-place safety measures and the like.Governments must seek to expand their available fiscal space to thus significantly contribute to covering the income needs of their ageing populations. Next to accepting their redistributive role, as required for maintaining inclusive and prosperous societies, governments must also re-assume responsibility as a major investor along with the private enterprise sector. Growth will not come from further unleashing the market forces but from a revival of an additional active role of the State as investor, redistributor and standards-setter.Under such conditions, this report holds that future growth, even if reduced, will be sufficient to allow for proactive social policies, including effective pension policy. With this perspective, the notion of ageing before becoming rich loses much of its gloomy implications for future social policies. That notion cannot be taken as a pretext for no or stagnant and defensive social policy. To the contrary, the process of building Asia’s welfare states must pursue a double track: proactive involvement of the State and enterprises in economic growth policies and simultaneous extension of social security (institutions).By embarking on this policy approach, the design of social policy and pension schemes will assume a decisive role. Much care must be taken in choosing the right institutional settings, such that pension systems are supportive to reoriented macroeconomic policies and not contradictory. Pension schemes (and other social security schemes) should be designed to enhance – not hamper – labour productivity (see section 2.2).61 According to International Monetary Fund(IMF) calculations (2017).As a result of ageing, future economic growth will only be marginally affected in the region. A maximum of minus 1 percentage point for the so-called post-dividend countries (among them China, Republic of Korea, Japan, Thailand) will be almost offset by a close to plus 1 percent maximum in the early dividenders (among them, India and Indonesia). But even a 1 percentage point lower growth path would not make a big difference to a country with stagnating population growth and especially the many people with low incomes; indeed, a 5 percent annual growth for all – if inclusive and equitable (including low income earners and the very poor) would, from a social policy point of view, be preferable to a path of 6 percent per annum. that is unequal and exclusive. It would be a complete misunderstanding of the economic effects of social security to assume that social security hampers growth. In the IMF analysis, a more comprehensive redistributive analysis and argument is missing, including the potential growth effects of such policy, which, inter alia, would mean increasing the labour income share in GDP as well as a careful choice in the design of social policy, including pensions.563.3 Social protection and pension trends in the ASEAN regionThe ASEAN Charter of 2008 is a document that legally binds its signing members to advance regional cooperation and economic integration. It sets regulations and values on the regional level and in matters of common interest that serve as foundations for the Member States to develop and achieve legislation, policy frameworks and enabling policy environments in many areas, including social protection and social security to older persons.The following sections of the ASEAN Charter elucidate the commitment of Member States towards a more prominent social protection agenda:Article 1 – PurposesTo alleviate poverty and narrow the development gap within ASEAN;To enhance the well-being and livelihood of the peoples of ASEAN, by providing them with equitable opportunities for human development, social welfare and justice; andTo promote a people-oriented ASEAN in which all sectors of society are encouraged to participate in, and benefit from, the process of ASEAN integration and community building.Article 2 – PrinciplesRespect for fundamental freedoms, the promotion and protection of human rights, and the promotion of social justice (ASEAN, 2008).In line with the ASEAN Vision 2020 that was adopted by leaders in 1997 and after signing the Cebu Declaration in 2012, efforts towards an ASEAN Community united in peace, stability and prosperity were initiated. The ASEAN Community entails three pillars, each with its own plan or blueprint: (i) the ASEAN Political Security Community; (ii) the ASEAN Economic Community; and (iii) the ASEAN Socio-Cultural Community.For this report, some of the action areas covered in the ASEAN blueprints can serve as basis for the development of national policy frameworks: (i) cooperation in political development, including promotion of peace and stability in the region by supporting poverty alleviation and narrowing development gaps; (ii) SME development; and (iii) advancing and prioritizing education, investing in human resource development and promotion of decent work, social welfare and protection (ASEAN, 2009).Other ASEAN declarations, signed and ratified by all Member States, contain explicit remarks regarding social security systems. For instance, in 2009, the ASEAN Human Rights Declaration emphasized:“Every person shall have the right to social security, including social insurance where available, which assists him or her to secure the means for a dignified and decent existence” (ASEAN, 2009, Article 30).In 2013, ASEAN leaders agreed on one of the most important institutional frameworks for the region: the Declaration on Strengthening Social Protection (ASEAN, 2013). The document sets basic principles and contains guidelines on advancing the social protection agenda and it lists strategies and mechanisms to help countries build or strengthen effective systems. Its contribution to ASEAN regional integration processes should be highlighted because it addresses core issues of social and economic development of each Member State and the community (box 6).57Box 6 ASEAN Declaration on Strengthening Social Protection: A milestoneThe 2008 Great Financial Crisis highlighted the role of social protection in mitigating the risks of unfettered markets and preserving economic and social stability in crunch periods. Between 2009 and 2012, a number of global forums advocated for expanding social protection, including the United Nations, the Group of 20 countries and the 101th International Labour Conference, which adopted Recommendation No. 202.The ASEAN Member States were no different. During the same period and as part of the regional integration process, they advocated for improved social protection and the progressive extension of coverage to all, following a life‐cycle approach. This led to the adoption in 2013, during the 23rd Summit in Brunei Darussalam, of the ASEAN Declaration on Strengthening Social Protection by the 10 ASEAN Heads of State. In adopting the Declaration, they pledged the completion of social protection floors as a priority to achieve growth with equity.Committed to actualizing this Declaration, the Member States agreed in 2015 on a regional framework and plan of action for the implementation. Increased social protection is also a core priority of the 2016–20 Senior Labour Officials Meeting’s Work Programme. The Member States are now defining a monitoring framework to measure progress in extending social protection, using relevant SDG targets and indicators. This instrument will be used to gauge compliance of the Member States with the 2013 Declaration.Over the past six years, the ASEAN Member States, via the ASEAN Secretariat, have enhanced their collaboration on social protection with the support of the ILO. In particular, ASEAN has requested ILO expertise and referenced to its standards for policy-oriented research on such topics as current and future trends of pension systems, social protection of migrant workers, the challenges of extending coverage to workers in the informal economy, financing social protection and monitoring social protection progresses.Source: ILO, 2017d, based on documents provided by the ASEAN Secretariat.The documents rest upon a phase of impressive economic development of the ASEAN Member States since the organization’s foundation in 1967. The ASEAN region’s GDP per capita has increased by almost 33 times since then.62 Life expectancy at birth has increased from 56 to 71 years, or by about one year every three years, on average. Between 2006 and 2015, the measured share in the population of people living below the poverty line fell significantly, with impressive results especially in Brunei Darussalam, Malaysia, Thailand and Viet Nam (DESA Population Division, 2017b; World Bank, 2018a; IMF, 2017).Figure 8. GDP growth rates in ASEAN Member States, 2008–17 (%)62 GDP per capita grew from US$122 in 1967 to US$4,021 in 2016.58Data source: World Bank, 2018a.ASEAN has achieved much progress towards the goals of its founding document, which contains strong social connotations and commitments (ASEAN, 1967). At the same time, old-age income security policy, in terms of the systematic establishing of social security institutions, has remained undecided over many decades, giving a scattered picture (see table A2 in Annex I).Coverage rates, in terms of contributing employed persons, and replacement rates, in terms of average benefits paid as a portion of average wages or average nominal productivity, are generally low among ASEAN Member States. Although the situation has improved in terms of legal coverage (see table A1 in Annex I), the ASEAN region still greatly relies on intra-family transfers and mutual help for aged people, which contrast to the fact that ASEAN Member States are demographically ageing (the number of old persons with no children, or only one or two, will considerably increase in the future). This reality requires more financial resources to be channeled towards the older generations as well as strengthening or establishing new forms of community help.ASEAN old-age pension trends at the country levelUnder economic aspects and socioeconomic indicators(table 12), ASEAN Member States can be grouped as follows:1. Group 1 consists of the two “outliers” – Brunei Darussalam and Singapore, which have high per capita incomes surpassing the incomes of all other Member States by between 3 and 50 times, such as Malaysia vs Brunei Darussalam and Cambodia vs Singapore. Brunei Darussalam and Singapore also present the best rankings in terms of the Human Development Index.2. Group 2 consists of Malaysia, Thailand, Indonesia, Philippines and Viet Nam, within which Malaysia and Thailand are the most and Viet Nam the least developed.3. Group 3 consists of Cambodia, the Lao People’s Democratic Republic and Myanmar – all with per capita GDPs of less than US$2,500 per annum. These are the three countries that rank lowest in terms of the Human Development Index. One common element in all of them is the large proportion of people still working in the agriculture sector.Table 12. General social and economic aspects of ASEAN Member StatesCountryHuman Development Index (rank), 2017GDP growth rate–average 2013–17GDP per capita (current US$), 2017Brunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet Nam39-1.2282911467.113841165.138471397.42457575.299451487.112991136.6298993.460306832.865941166.22343Data sources: UNDP, 2018; World Bank, 2018a;email correspondence with Singapore Department of Statistics, 2019. 59Group 1: Brunei Darussalam and SingaporeBecause Brunei Darussalam is an extraordinarily resource-rich and economically stable country, it is in a unique position in relation to other ASEAN Member States to secure social protection for its older population. It has made great strides in the past decades. Notably, it has achieved full legal and high effective pension coverage (more than 90 percent) through a combination of different schemes. At the core of high coverage is its tax-funded, and not means-tested, universal Old-Age Pension, which constitutes the overall system’s first or basic tier. The second tier is the contributory Employees Trust Fund, which is a provident fund providing benefits for public and private sector workers. The first and second tiers are complemented by the Supplementary Contributory Pension scheme, to which the Government offers matching contributions, conditional on an income test (box 7).Box 7Brunei Darussalam’s old-age flat-rate demogrant: An ASEAN experience with universal oldage social pensionsA universal old-age pension is available for every resident of Brunei Darussalam upon their 60 birthday. Persons born in the country must have at least 10 years of residence immediately before claiming the pension, while persons born outside must have lived there for at least 30 years immediately before claiming the pension. The universal old-age pension provides 250Bruneian dollars (BND) (approximately US$200) per month and the amount is adjusted on an ad hoc basis.Granting adequate benefits while sustaining financial stability will be the greatest challenge in the medium and long terms for the country. Indeed, Brunei Darussalam is ageing; however, the demographic situation is not dramatic: With high per capita GDP and growing life expectancy, the pension financing issues should be manageable, mainly by way of increasing the retirement age to 65 and making more use of the social pension, the first old-age pension pillar, as a policy instrument. That Brunei Darussalam implements a flat tax-financed pension can be supportive to that end, not the least in case of persistently low interest rates in the future. Through this additional tool, social policy can be organized on an equitable basis by shifting more national resources into the scheme. Given the population ageing, the Government must sooner or later explore the need to implement and finance a system of old-age care (see Chapter 5 and the country brief for more details).Source: www.pensiondevelopment.org/51/brunei.htm, www.pension-watch.net/country-fact-file/brunei-darussalam/[accessed 13Aug. 2019].Singapore, with its high GDP per capita, has thus far quite successfully relied on its Central Provident Fund for executing its social policy. The Central Provident Fund is one of the oldest and most developed old-age schemes in Asia. In its function as a financial intermediary, the Fund has contributed to Singapore’s growth and is deemed financially viable. Most of the Fund’s efforts in recent decades focused on ensuring adequate income to retirees to maintain their pre-retirement standard of living. Its adequacy is continuing to improve – six in 10 members who turned 55 in 2018 were able to meet the Basic Retirement Sum, which provides for a life-long pay-out in retirement that covers basic needs. The Fund’s progressive interest rate structure, with lower-balance members and older members getting a higher effective interest rate, introduced socially welcome redistributive elements to the scheme. For example, the base interest rate on monies in the Retirement Account (meant for retirement pay-outs) is 4 percent, but an extra 1 percent of interest is given on the first SGD60,000 balance, and an additional extra 1 percent is given on the first SGD30,000 for members aged 55 or older.60At the same time, Singapore is the ASEAN Member State with the”oldest” population; it is projected that,as of 2040, the share of its resident population aged 60 or older will likely increase to around 32.5 percent, from 20.5 percent in 2018. Thus, as the number of older persons increases (and if everything is kept constant), an increase in their share of consumption will also take place.The Central Provident Fund contribution rates, totaling 37 percent (CPF, 2016), are high and adequate for a productive high-income country, which contrasts positively with the situation of underfunded social policy in many neighbouring countries. The allocation of funds for the different Central Provident Fund programmes varies by age, and only a fraction is being withdrawn for retirement benefits, while the largest withdrawals are for purchase of property.63 Thus, despite the high level of contributions, almost one third of Singaporeans older than 60 years declares to be financially dependent on their grown-up children, a situation that clearly reflects a benefit adequacy problem (Jackson and Peter, 2015). This situation can be traced back to two core reasons: The first is that benefits are (for a significant group)withdrawn in the form of one or several lump sums, including for the purpose of co-financing the purchase of property at younger ages, which collides with the need to accumulate funds for old-age. Second, with underdeveloped or widely non-existent skills to financially manage these amounts, many older members are left unprepared to face the contingencies of (long) life duration and possibly other uncontrollable factors of life. Recognizing these detrimental factors, the Central Provident Fund began in January 2013 obliging its members with at least SGD60,000 (US$44,000) accumulated in their Retirement Account at age 65 to purchase the life annuity offered by the Central Provident Fund Board.In addition, for reasons outside their control, a considerable proportion of active members is unable to set aside the minimum sum for meeting basic needs in their old age. This is particularly relevant for women who often occupy lower-paid jobs and generally participate less in the labour market and thus face more challenges to accumulate sufficient savings for old age. To better manage this reality, the Government launched the Silver Support Scheme in 2016. It provides a quarterly cash supplement to the bottom quintile of Singaporeans aged 65 and older (box 8).Box 8Singapore’s Silver Support Scheme: Tax-financed and means-tested old-age income supplementSince 2016, Singaporeans aged 65 or older,who had low incomes throughout life, little or no family support and live in public housing apartments automatically qualify to receive quarterly tax-financed cash pay-outs under the Silver Support Scheme to supplement their retirement income. The scheme targets older persons in the bottom 20 percent of the income distribution.Eligibility criteria for the Silver Support Scheme consists of (i) household per capita income of 1,100 Singaporean dollars (SGD) or less; (ii) total accumulated Central Provident Fund contributions of up to SGD70,000 by age 55, and, if self-employed, annual income of up to SGD22,800 between the ages of 45 and 54; (iii) residency in a five-room or smaller public housing flat (Housing and Development Board, or HDB flat; and (iv) no ownership or no spouse having ownership of a five-room or larger HDB flat or other real estate property. The amount paid varies according to the number of rooms in the HDB flat that the older person lives in and can range from SGD300 to SGD750 per quarter. Older persons living in smaller flat types get63 S ee https://www.cpf.gov.sg/Members/AboutUs/about-us-info/cpf-statistics [accessed 13 Aug. 2019]. Regrettably, theCentral Provident Fund does not publish income replacement rates of withdrawn (annuitized) pension amounts. Although housing withdrawals of US$2.4 billion were higher than retirement withdrawals at $2.1 billion in 2018, they did not constitute a majority of the SGD8.9 billion in contributions paid over the period (direct communication with Singaporean authorities).61higher pay-outs. The Silver Support Scheme covered 153,000 older persons in Singapore in 2018 (Ministry of Manpower, 2018). An additional element that needs to be considered is the fact that, in an ageing society,health requires a growing share of available resources. In 2011, Singapore allocated 0.7 percent of its total public expenditure to income security for retirement and 3.5 percent to healthcare (ILO, 2017d).In comparison with advanced countries that have ageing societies, the share of public expenditure on old-age income security can be considered low, indicating that most likely these numbers will be growing in the future. One proposal that could be considered by Singapore is to transform its current Silver Support Scheme to a universal non-contributory social pension similar to Brunei Darussalam’s demogrant (see Chapter 5 and the country brief for more details). Figure 9. Social protection expenditure,including health care, in ASEAN Member States, latest available year (% of GDP)Note: Data sources include the Asian Development Bank, International Labour Organization, International Monetary Fund, GSW and the Organisation for Economic Co-operation and Development (see ILO,2017d, Annex table B17 for details).Data source: ILO, 2017d. Group 2:Indonesia, Malaysia, Philippines, Thailand and Viet Nam Indonesia’s GDP per capita levels are at a rate of around 7 percent of Singapore’s levels. Its population amounts to around 40 percent of the ASEAN total and is, by far, the biggest of all ASEAN Member States. Indonesia faces shortfalls in pension coverage, with only around 14 percent of persons aged 60 and older receiving a pension in 2015 (ILO, 2017d), especially among people who had worked outside the formal and wage-paying sectors. But also, coverage in the formal sector remains low, primarily due to lack of compliance with legislation on mandated participation. In 2015, the percentage of active contributors was estimated at only 7.6 percent of the working-age population (ILO, 2017d). With the share of informal employment at 80 percent and 77 percent of the workforce employed in the informal sector (ILO, 2018d), any policy of coverage extension requires particular attention from and towards this group, including compliance of businesses in the formal sector. Most persons among the covered population in Indonesia belong to the highest income quintile. While its population structure, shape and speed of ageing is similar to the other countries in this 62group (not especially alarming for the foreseeable future), the most pressing challenge for Indonesia is the implementing of a tax-financed pension for older persons who have no contributory pension entitlements. Orientation along the lines of solutions found in Thailand is advisable. Currently, the Social Assistance for Older Persons (ASLUT) scheme covers less than 0.2 percent of the population aged 60 or older, corresponding to an extremely low level of public expenditure of around 1 percent of GDP in 2015 (ILO, 2017d).As in other countries in the region, one of Indonesia’s core problems of social policy is the lack of a coherent and strategic vision for old-age income security. The country has a combination of social insurance and provident fund. While the social insurance replacement rate is only 30 percent after 30 years of contributions, the provident fund cannot fulfill its intended meaningful role due to a combination of detrimental factors, including the high rate of withdrawals before retirement,the low levels of savings accumulated and the absence of automatic annuitization. In addition to this dysfunctional system, some businesses provide occupational pension plans to their workers,which benefited 3 million workers in 2011 (ILO, 2014a). Workers may take advantage of stipulations of the Labour Law entitling them to certain benefits at the termination of employment.As in Malaysia and Thailand, the lack of a coherent and comprehensive approach to the old-age income security problem dealing with coverage, adequacy and financial and social sustainability in a holistic way hinders the country from encountering long-term solutions for these problems. Organizationally, the vast extension of the country, similar to the Philippines, is, and will remain, a major challenge in comparison with other countries (see Chapter 5 and the country brief for more details).Of all five countries in the second group, Malaysia has the highest GDP per capita, equaling only close to one fifth of Singapore’s GDP. It runs an old-age pension system similar to the one in Singapore and thus relies mainly on a provident fund. After more than six decades, the country has begun transforming its pension system from a single-pillar to a multi-pillar system. However, the system faces growing challenges. Around 37 percent of the labour force stubbornly remains uncovered by the Employees’ Provident Fund; they represent mostly self-employed workers who are entitled to contribute to the scheme on a voluntary basis only. But only about 1 percent of them actually do so.64 Other largely uncovered groups are unpaid family workers and labour migrants (World Bank, 2014).One of the main issues with the Employees’ Provident Fund is that the system allows members to make several withdrawals before reaching retirement, which significantly reduces the accumulated amounts available to them when they exit the labour market. In the absence of regulations requiring annuitization on a compulsory basis, only 100 members, or 0.13 percent of the total active contributors reaching age 55 in 2014, chose the periodical payment withdrawal scheme (EPF, 2015). About half of those who select a lump sum withdrawal exhaust their savings within a period of five years (EPF, 2015), creating additional and serious problems of old-age income insecurity, to be borne in the end by the society directly or indirectly (box 9).64 The percentage covered by the Social Security Organization under its working injury, employment and invalidity schemes is different from the numbers in the Employees’ Provident Fund. The Social Security Organization had more than 6.9 million active contributors in 2018, representing 47 percent of the total working force.63Box 9 Little protection against longevity risks in the Malaysian Employees’ Provident FundSome 50 percent of workers reaching the age of 55 and withdrawing their Employees’ Provident Fund savings in 2014 exhaust their money within five years (EPF, 2015). In 2015–20, life expectancy at age 55 in Malaysia reached 24 years and is projected to increase to 28 years sometime between 2050 and 2055 (DESA Population Division, 2017b). The Fund is the largest old-age pension scheme for the private sector in Malaysia, including the self-employed. Thus, it becomes crucial to assess the withdrawals and annuities rules to improve its old-age insurance capacity.A few explanations for the high level of lump-sum withdrawals at a relatively early old age of pensioners are possible: First, on the individual level, it is hard for pensioners to have an idea of their remaining life expectancy. This negatively affects individuals’ capacity to plan and tends to cause a preference for present consumption rather than long-term needs. Members of the Fund prefer to withdraw their money at a relatively early age and count on “family-internal annuity processes”, whereby informal financial support throughout old age is provided within the family unit. Second, financial literacy concerning the Fund’s rules and possibilities for withdrawing funds as well as general trust in the system are generally low, which prompts members to opt for the lump-sum pay-out as soon as possible, considered a simpler alternative for accessing their lifelong savings. And third, the financial products available in the Malaysian financial markets might not be the most adequate to address faults in or develop annuities products for the Fund’s participants.There are several schemes in Malaysia offering social assistance to the older poor that could be labeled the country’s first pension pillar. However, this fragmented pillar fails to cover all older people in need. Its main programme is the means-tested and tax-funded Bantuan Orang Tua, which offers basic transfers to an increasing number of persons aged 60 or older. Coverage under the Bantuan Orang Tua is extremely limited. In 2010, it covered only 3.6 percent of people older than the eligible age (ILO, 2017d). In 2016, the benefit of 300 Malaysian ringgit (MYR) corresponded to 33 percent of the average national minimum wage (ILO, 2018d).It appears that after more than six decades, Malaysia has begun to transform its pension system from a single-pillar towards a multi-pillar system. The population structure and its shape of ageing, together with changing family and labour market structures, are similar to the critical situation in Thailand and will exacerbate the challenges of low coverage and replacement rates. The Government is thus advised to implement measures to regain the general public’s trust in the provident fund and would be well advised to pursue exploring social pension solutions as an additional tier, as in Thailand (see Chapter 5 and the country brief for more details).The Philippines’ GDP per capita – only slightly below Indonesia – is at almost 6 percent of the level in Singapore. The Philippines has one of the oldest pension schemes in the region, which is mutating into a four-tier system of protection, even reminiscent of European solutions. Of all the ASEAN countries, the Philippines operates the most complex system, requiring good governance and well-developed pension literacy of scheme stakeholders (workers, employers and the Government). These requirements seem not to be fulfilled always and everywhere; replacement64rates are among the highest in the world, while less than 40 percent of older persons can count on receiving a pension at all. This rate plunges to 20 percent for women and rises to 53 percent for men.The system’s first tier is a social assistance benefit targeting destitute older persons and covering close to 17.9 percent of persons aged 60 or older in 2015 (ILO, 2017d). The second tier of social insurance providing defined-benefits pensions legally covers all private sector employees and own-account workers. In practice, coverage is mostly limited to persons working in the formal sector, with less than 22 percent of the working-age population contributing to the system. The third tier consists of the mandatory deposits maintained in the PAG-IBIG Fund or the Home Development Mutual Fund, both of which become available at retirement.For the foreseeable future, the population structure and shape of ageing remain favourable. Nevertheless, family support for older persons must remain strong, while alternatives will also need to be found in the face of urbanization and persistent labour emigration. As in Indonesia, the country’s extension is a challenge for administration and governance. To address the low coverage, the Government is advised to streamline social assistance and explore replacing social assistance with a universal, flat-rate, tax-financed pension, along the lines of solutions found in Thailand. For the Philippines, this might mean – more than in the other ASEAN Member States– a major break with the prevailing old-age income security paradigm, but the option is still worth exploring (see Chapter 5 and the country brief for more details).Thailand, with average GDP per capita only slightly above 10 percent of the Singapore level, has reached a remarkable degree of pension coverage through its universal old-age allowance, at around 83 percent of all persons aged 60 or older in 2016 (ILO, 2017d). Yet, the level of the allowance is insufficient to guarantee a decent living, even in combination with other forms of old-age income. In fact, 42 percent of persons aged 60 or older have insufficient income to maintain their pre-retirement living standards (ILO, 2016c). To this too-low level add that the old-age allowance is not automatically indexed to inflation or per capita wages, or a combination thereof, leading to a progressive erosion of the benefit’s purchasing power. As a result, the 2016 benefit level amounted to 10.3 percent of the minimum wage (ILO, 2017d).Adding the pension of the system’s social insurance pillar, or the second pillar, managed by the Social Security Office to the old-age allowance improves the situation but not by much. The social insurance pension theoretically allows for a 45 percent replacement rate after 30 years of service, which conforms with the ILO-defined minimum replacement rates (ILO Conventions No. 102 and No. 128). But the suboptimal governance of three other scheme parameters prevents pension adequacy: The contribution ceiling established at the creation of the scheme, in 1998, was never increased to take into consideration general wage developments. As with the old-age allowance, no pension indexation mechanism was established into force. Being a “young” scheme, the first retirees were not able to accumulate entitlements basedon long contributory careers, which, in the absence of grandfathering of provisions, led to relatively low replacement levels for many people.The simultaneous combination of all three factors generated (and still generates)a lack of trust in the scheme among contributors (workers and employers) as well as lack of understanding of the social and economic values of the social insurance pension system. Under growing public pressure resulting from the general disappointment with the system’s inadequacy, the Government recently launched an additional mandatory provident fund on top of the old-age allowance and the Social65Security Office pension – instead of first rectifying the parameter inconsistencies in the system. Therefore, the decision to add a new provident fund move lacks a harmonized policy, one that considered the pension system as a whole and developed comprehensive solutions that brought together, consistently and congruently, the different pillars to establish overall pension system adequacy, sustainability and affordability. Without considering its impacts on the existing pillars, the establishment of a new scheme might fail to lead to positive outcomes.One of the main challenges for the future will be to harmonize the combination of the flat, universal, tax-financed old-age allowance and the contributory system for the private sector that currently includes the social insurance pension and a mandatory provident fund. Harmonization should include institutional rationalization and parameter reforms of the contributory system, increasing the benefit levels of the old-age allowance and extending the contribution base into the informal sector. One other aspect requiring additional efforts regards the percentage of active contributors within the total working-age population, which the ILO recently estimated at less than 34 percent. This reflects the formal-informal split in the labour market, including challenges of compliance with relevant legislation. Despite the positive economic developments in recent decades in Thailand, the 2016 labour force survey findings indicated a level of informal employment greater than 60 percent, which was due to the large share of workers in the agriculture and fishing sectors (ILO, 2016c).Finally, design consistency of the pension system must also be improved to put it in line with other social security institutions, like health. Given Thailand’s rapidly ageing population, the Government must urgently address the issues of implementing and financing a system of old-age care. A more decisive social policy is recommended to overcome the growing urban-rural split of the society (see Chapter 5 and the country brief for more details).Viet Nam has a GDP per capita close to 80 percent of the Philippine level. Like other ASEAN countries with similar economic development levels, Viet Nam faces the dual challenge of improving the financial sustainability of its pension system and expanding coverage to the large informal sector.In these respects,two problem areas are underlined here: The first is the issue of coverage, both in terms of the number of contributors – around 23.5 percent of the total labour force contributed in 2015 (ILO, 2017d) – and of the number of older people receiving a pension,which was less than 40 percent coverage in 2015 (ILO, 2017d). There are several reasons for these issues. One is the usual fact that coverage is almost completely limited to persons working in the formal sector, while almost 58 percent of workers are in informal employment (ILO, 2018e). Almost 20 percent of people in informal employment work in the formal sector or formal enterprises (PSA, 2015), indicating a challenge of compliance with the social insurance laws. Workers in the informal sectors are meant to be covered by a voluntary scheme,which currently does not even reach 0.5 percent of them (ILO, 2017d).With the current low scale of coverage provided by the tax-funded schemes, only 17 percent of older persons were covered by non-contributory social pensions in 2014 (ILO, 2017d). A large majority of the population is left unprotected by old-age pensions.The second problem area is the system’s medium-to-long-term financial sustainability. Even though a large number of older people do not receive a pension, the system costs, which corresponded to slightly more than 5.5 percent of GDP in 2015 (ILO, 2017d), are relatively high in comparison to other ASEAN Member States (table 15), although still low in a global comparison. Simple, static analysis of the unchanged population structure indicates that achieving full coverage would increase these costs to roughly 8 percent of GDP (table 15). In this context, increasing the retirement age will be66unavoidable in the future to improve pension system financial sustainability.While social pension benefit levels are low, at only 6.6 percent of per capita GDP, the replacement rate of the contributory pension scheme is high, at 60 percent, albeit featuring only small redistributive elements that contribute to rising economic inequality. Despite this relative theoretical high replacement rate, because insurable earnings are based only on the basic salary, there is a potential adequacy problem(see Chapter 5 and the country brief for more details).Group 3:Cambodia, the Lao People’s Democratic Republic and MyanmarCambodia’s GDP per capita is at 20 percent of what it is in Thailand but only 2 percent of the level in Singapore. In economic terms, it is the poorest ASEAN Member State. With only a tax-financed public sector pension scheme for civil servants, the country’s pension system is one of the least developed in the region. The country is making some efforts to improve, although it has several steps to take to implement a comprehensive social security system.There are still six Cambodian citizens aged 15–54 for every person aged 55 or older. Within 50 years, this ratio will fall to 1.6 persons (DESA Population Division, 2017b). In the absence of a pension system, this will lead to an unbearable increase of the financial and physical burden on families. Many of them, even under assumed future economic growth, will likely be living in situations of socioeconomic vulnerability. A comprehensive pension system that rewards the important social contributions made by older persons to private households and local communities is overdue. It is important to take into account that any contributory scheme will most probably have low coverage because the prevalence of a typical work arrangements and short duration contracts turn such schemes dysfunctional in terms of coverage as well as benefit levels.65 For long-term benefits, such as old-age pensions designed under standard rules, the Cambodian labour market will allow only a slow accumulation of years of service to be taken into account for retirement benefit calculation. Thus, traditional social insurance is not likely to provide adequate income security to Cambodia’s older population66 (ILO, 2018a).Initiatives are underway to extend coverage of the workforce by introducing a social insurance scheme for private sector workers and to introduce a social pension, at least for impoverished persons. The Government should aim at solutions similar to those in Thailand. Demographically, Cambodia is in a comparatively favourable situation: While undergoing an ageing process, the demographic situation will remain far from dramatic until 2050. The financing of old-age income should be, under mere demographic aspects, no major problem, and for this reason, the Government would be well advised to explore issues of old-age care at an early stage (see Chapter 5 and the country brief for more details).The Lao People’s Democratic Republic’s GDP per capita is estimated to be twice as high as what it is in Cambodia but still only around 4 percent of the Singapore level. The country has a contributory pension system in place for private sector workers, with coverage that is low due to the small size of the formal labour market.The country faces two major challenges: First, even when disregarding jobs in the agriculture sector, roughly 60 percent of jobs are in the informal sector. Second, the prevalence of informal employment is high, even in the formal sector. Informal employment in the formal and informal65 Even the core of Cambodia’s formal sector is informally employed for almost half of the year, reflected by the average contribution period to the scheme,at 7.2 months per year.66These remarks mutatis mutandis apply to all countries and systems in similar situations, including in the most advanced economies.67sectors or enterprises account for 79 percent of all employment. If agriculture jobs are included, the ratio goes up to 94 percent, at 96 percent for women (ILO, 2018e).Accordingly, the number of contributors is extremely low. In 2010, only 1.3 percent of all employed persons contributed to the social security system. This implies that the number of future beneficiaries of the contributory pension scheme will be small. In 2010, 5.6 percent of persons older than the statutory pension age received a contributory pension (ILO, 2017d). As yet, no social pension exists that could be used to address the coverage problem, which is a major cause of old-age poverty.As in Cambodia, the Lao population is ageing, but the foreseeable situation is not dramatic in demographic terms, with respect to the implications for old-age income financing. For this reason, the possibility for introducing a social pension should be carefully explored. Also, the Government would be well advised to explore issues of old-age care at this early stage of the projected demographic trends. Reforms will be necessary to keep the public sector scheme financially sustainable. A recent actuarial valuation by the ILO documented the conditions under which the pension benefits branch, particularly the one for civil servants, is financially unsustainable. Parameter reforms are necessary to achieve long-term financial sustainability of the system, which must be carefully balanced with the requirements for social sustainability (see Chapter 5 and the country brief for more details).In terms of GDP per capita, population structure and ageing, Myanmar is in a similar situation as Cambodia and the Lao People’s Democratic Republic. The country’s old-age dependency ratio has increased from 6.8 young persons per aged person in 1983 to 8.8 in 2014. At this pace, it will take the country only 25 years to become an “aged society”. In comparison with Cambodia or the Lao People’s Democratic Republic,the Myanmar population is much bigger. The majority of older people, particularly in the rural areas, are poor in terms of income and assets. The vast majority of older people, at 94 percent,is economically active (DOP, 2016). Although formal employment may increase in the future, informal employment is currently widespread, corresponding to 82 percent of total employment, which allows only a small fraction of the workforce to contribute to a formal pension scheme. As in other ASEAN Member States, there is also lack of social security coverage for persons working in the formal sectors. For instance, around 19 percent of workers in the formal industrial sector and almost 16 percent of those in the formal services sector are informally employed (ILO, forthcoming b).Not surprising, poverty is widespread among the country’s older population. In the absence of a comprehensive state-supported social protection system, older people rely on their children and strong familial ties for material support (Knodel, 2012). Most older people have four to five living children; only 6 percent are childless (Knodel, 2012). But as the fertility rate drops in Myanmar, the increasing burden of care will fall on a decreasing number of children (Clay et al., forthcoming).Following the political and economic opening in the early 2010s, the country began building the main elements of social security, albeit only covering small segments of the population thus far. The first step was the Social Security Law in 2012, which includes a provision for an old-age pension under the responsibility of the Social Security Board. It is expected to cover private sector formal workers in companies with five or more workers, but it has not yet been implemented. The National Social Protection Strategic Plan, launched in 2014, includes two programmes for older people, including provisions for a universal social pension for those aged 65 or older. A regular payment condition on an age threshold of 90 years was recently introduced. Coverage is obviously low, but the Government intends to lower the age threshold and thereby increase coverage when fiscal space expands and administrative capacity is improved.68Myanmar’s evolution of social security definitely depends on successful economic development. Nevertheless, it would be wrong to put economic development first. Already more efforts are needed in terms of resource allocation to the social and pension sectors and the development of respective administrative capacities. Myanmar has the unique opportunity to avoid the development of piecemeal solutions and, instead, to look at the problem of securing old-age income in a comprehensive way through articulated contributory and non-contributory schemes,thus progressively reaching the entire older population while avoiding overlaps and inefficiencies. Social policy options, like in Cambodia and the Lao People’s Democratic Republic, are pending (see Chapter 5 and the country brief for more details).6970Photo: @Réhahn Croquevielle4. PENSION POLICY RECOMMENDATIONSFOR ASEAN MEMBER STATES71In light of the theoretical aspects presented previously and the old-age policy challenges and trends in ASEAN Member States, this chapter presents a series of recommendations and options. Governments and national stakeholders are asked to consider them as solutions for coping with the challenges and thus ensuring that all citizens within the ASEAN Member States can enjoy a decent life during their old age. The chapter unfolds in three sections: The first presents a series of recommendations without a specific country focus, and the second shifts to country-specific recommendations. Despite the concreteness of some recommendations, there remains a need for detailed country analyses. Finally, due to the importance of fiscal space for income security in oldage, the third section is dedicated to this topic, describing some of the options available for countries and presenting some general estimations on how costs of pensions could develop in the region.4.1 General considerations on pension policy-making for ASEANWith ongoing urbanization and with fertility and mortality rates dropping, the concept of extended families harmoniously looking after themselves has become more of a romantic concept of the past that cannot be maintained under the conditions of modern capitalist production in the ASEAN Member States. Families’ capacity to support their older population is being squeezed, creating serious risk of increased vulnerability and poverty in the midst of societies that are becoming richer on average. If the concept of “family”, which is one of the core strongholds of any society worldwide, is to be maintained and re-strengthened in ASEAN Member States, then families need relief from intra-family burdens rather than additional ones. As a standard measure, such relief can be organized by way of a formal pension system fully covering all members of society.Although still growing faster than most of the so-called advanced countries, ASEAN Member States have realized that they will have to cope in the future with more modest economic growth rates – implying lower rates of return in pension savings schemes. Societies will no longer be able “to hide behind high rates of return seemingly producing effortless income”, including for pensioners. They must now prepare for explicitly factoring these costs into their economy’s production costs. If ASEAN societies want to keep older people out of poverty, there is no alternative –governments must take quick action.In this respect, it might be helpful to look at the bigger, global picture. Actually, the costs of oldage income security are not exclusive to the ASEAN countries but must be borne, more or less simultaneously, by almost all countries around the world. China, as a big stand-alone country, has already proven that a comprehensive pension policy is possible without affecting its international competitiveness. The ASEAN countries have the chance to follow suit while coordinating the related cost effects by means of their international organization and, possibly, through common exchange rate policies. ASEAN Member States, having already a common platform, could find agreement among themselves that the costs of ageing be kept excluded from regional competition. They could undertake appropriate motions or declarations through the channel of international organizations.For example, ASEAN governments could agree to set targets and goals in terms of coverage and benefit levels. ILO Conventions No. 102 and No. 128 could serve as an internationally tried and tested platform to start from. With increasing percentages of older people, governments that disregard these expenditures may see enormous consequences in terms of poverty at the country level and could put at risk all progress reached over recent decades.The experience in the region shows that solutions exclusively based on contributory systems do not lead to substantial expansion of coverage. Considering the urgency and the large presence of informal employment, the only recourse for countries that want to achieve quick results in terms of72coverage expansion are tax-funded solutions, for instance, through the use of demogrants (such as in Brunei Darussalam and Thailand).Earnings-related social security should keep performing the central role of ensuring adequate pensions, particularly in terms of replacing working-related income. Social insurance mechanisms should be a central element of those schemes. One of the advantages of such social schemes is that they allow for the introduction of redistribution, which offers social policy flexibility if needed. For example, they can be used to address the specific needs of workers structurally trapped in vulnerable situations.The term “social policy flexibility” is not intended here as an invitation to change social insurance schemes permanently – indeed, social policy and their underlying schemes need stability. Rather, it refers to how these schemes are under permanent observation and, when necessary, can be reformed. Women can particularly benefit from such an approach to social insurance because it allows the scheme to integrate design elements that can correct some of the imbalances created by markets and labour markets. Doing the same under defined-contribution schemes is technically close to impossible because tight system limits preclude any adjustment.When considering both tax-funded and contributory schemes, countries should avoid a silo approach to policies. They should instead pull together the different pillars and sources of funding in an integrated policy framework. To respond to the challenges of national labour contexts, the design of a contributory system cannot be disconnected from the non-contributory system and vice versa. Attempts to disconnect these two sides of the same coin have precluded systems from maximizing their objectives.When systems are consistently approached, they offer more effective solutions not only in terms of coverage but also in terms of adequacy. For instance, if flat tax-funded pensions are part of a wider system, even a small number of years of contribution payment can top-up the flat pension and thus increase the overall monies available to pensioners in old age. Under such an arrangement, the flat pension will give someone with years in vulnerable employment a sense of not having worked in vain – because a pension will be available in old age – while efforts to join formal employment, even if only for a short period, would still be rewarded through a contributory pension topping-up the tax-financed component. These considerations are particularly important in countries in which most jobs in the formal sector are offered in industries where the average working age is young, implying that workers must find new work as they mature, which will most likely be in the informal and farming sectors.At the same time, countries should invest more to understand the reasons why workers are not covered and place more effort on the development of solutions that allow the participation in schemes by workers under their particular circumstances. This includes turning informal enterprises formal and, more generally, intensifying policies towards strengthening employers’ social sense of responsibility in organizing formal social security for older people. Examples from other regions could be an inspiration, such as the processes of administrative simplification for SMEs or collective registration for self-employed persons.Policies of whatever nature to enhance coverage are often not successful without continued efforts to increase the levels of compliance with relevant legislation. This includes critical review and, where necessary, improving the existing legislation. In many countries (not only in those covered by this report), there are substantial coverage gaps for workers employed informally in formal enterprises, in addition to workers in the informal sector. Increasing the social security and labour administration73capacity to enforce compliance with regulations is essential, as is its capacity to adapt to new forms of work in the formal economy. It was not within the scope of this report to determine whether informal work in formal enterprises may be a result of inadequate and cumbersome legislation and administrative rules – nevertheless, this should be kept in mind as a possibility that hinders effective coverage extension.Countries that have achieved considerable levels of coverage must now focus on benefit adequacy. This requires a more comprehensive approach, not only looking at the amounts of the flat-benefit social pensions but also ensuring the extension and adequacy of the earnings-related social insurance component. Particularly relevant will be the capacity to assess the complementarity and potential synergies between these two elements.Adequacy is generally a multidimensional concept. A benefit may be considered adequate by the society at large but not by a beneficiary. The reverse situation may also hold: There are situations in which benefits are considered too high by the society, such as when special societal groups take advantage of their assigned or achieved privileged roles, and no equivalent rules or privileges exist for the rest of the society. These are usually complex issues when formulating policies aiming at equitable and inclusive societies and also beyond the scope of this report.We understand adequacy mainly in terms of minimum conditions that must be fulfilled for which the stipulations of ILO Conventions No. 102 and No. 128 (as well as Recommendation No. 202 in regards to flat pensions) can be taken as reference. A benefit is understood as adequate at retirement when it can reasonably (on the basis of objective statistics and/or quantitative rules) be expected to allow for a decent living in old age. In pension legislation, this goal is usually determined by a simple formula for defined-benefit schemes or through the rules governing the calculation of an annuity undefined-contribution schemes.Thus, adequacy is foremost a benefit-level issue in the moment of retirement. Most, if not all, ASEAN Member States must improve their pension benefits in this respect.A measure of equal importance is to index pensions regularly, once payment begins. The indexation of pensions should not be discretionary because, as international experience shows, this is not only a sign of bad or suboptimal governance but it tends to mess up the intrinsic system logic of pension schemes. And it often ends in inconsistencies and unequal and unjustified treatment of scheme members.In the absence of indexation mechanisms, the purchasing power of benefits disappears over time, thus destroying the pension system’s main objectives. ASEAN countries need to assess what are the most adequate mechanisms and the best benchmarks to use as a basis for indexing their pensions in payment. International practice mainly rests on annual, semi-annual or quarterly formulabased indexation, either in line with the CPI or with national or pension system average wages, or a combination thereof. More elaborate methods are sometimes linked with labour market and/ or productivity developments. These approaches, however, are likely to replace transparency with sophistication, which, depending on the circumstances, may undermine the general public’s trust in the system.For countries that have opted to rely on defined-contribution solutions, based on individual savings accounts under the management of provident funds, we do not recommend radical changes. Recognizing the limitations of those approaches in terms of old-age income security, especially for a society’s lower income strata, the report recommends considering a flat universal pension as a first74tier or pillar of the overall system. We briefly discuss here (but do not come to a decisive conclusion) the additional option of re-structuring the existing provident funds, possibly into two schemes, in which one scheme would maintain the current provident fund features and the other would be a defined benefit-based pooling mechanism financed by some part (to be determined) of the overall contributions, thus introducing an element of redistribution into the contribution-funded system.For countries with defined-benefit schemes, one of the principal shortcomings of social insurance has been the capacity to promote timely and required reforms. Reforms are often not politically popular in the short run but necessary to guarantee the preconditions required for a scheme’s long-term survival as a societal institution. Reforms usually include increasing the retirement age and reducing entitlements, at least with overly generous replacement rates, such as high benefit provisions. These reforms usually aim to balance the interests of beneficiaries with those of contributors, such as limiting the otherwise unavoidable increase of contribution rates, stabilizing them or even reducing them.Sometimes, reforms aim at reducing government subsidies for financially ailing pension schemes. Delaying decisions on any reform too long can have serious implications in terms of the long-term sustainability of schemes but also it can create unfair situations between generations. In the short run, current generations may be benefiting from the postponement of decisions, while in the longer term, those current generations (whose life expectancy has been increasing and will most probably continue into the future) and their children’s generation will probably be exposed to much more drastic measures. Instead, if decisions are taken at an early stage, reforms can be of a more gradual nature, thus allowing people to adjust and adapt to the limited impacts that these decisions will have on their life.Countries included in this report should reconsider and/or take a“watch list”position to implementing reform measures, adjusted to their specific social and economic conditions. In countries with a large gap between the rich and poor households and with only a small middle class, reform measures must first be directed towards the poor while not ignoring the interests of the middle class – a difficult task, indeed, requiring careful preparation of legislation and administration.In the ASEAN Member States, the dominant problem is not so much generosity of benefits but the underfunding of schemes, represented by inadequately set revenue parameters: too low contribution rates and, where applicable, too low and non-indexed contribution collection ceilings. In this context, the appropriate definition of insurable earnings is one of the often disregarded but most important elements concerning schemes’ financial sustainability and benefit adequacy. A low ceiling or the use of a definition that covers only low base salaries (thus representing only a small percentage of total earnings while possibly excluding more important additional components of remuneration) will mean the insurable earnings are too low in relation to the earnings effectively earned. Pensions will thus usually replace only a too-low percentage of the effective earnings and have only limited impact in terms of guaranteeing minimum levels of well-being (financial and other) in old age.The increase of revenue, such as the increase of contribution rates, is also overdue in all provident funds, as addressed in this report. In most schemes this change is urgent; in only a few schemes it is less urgent. Generally, the assumption that decent pensions can be financed (whether prefunded or pay-as-you-go) by single-digit contribution rates is wrong. Over the past two to three decades, a worldwide illusion was nourished, fueled by high growth and high returns on evolving financial markets. When lower growth rates become the new normal, contribution rates must be increased substantially to an old normal level. Pensions are expensive. If contribution rates of around 25 percent are considered too high for the ASEAN Member States’ poor households, then there is no75choice but to either subsidize the contributions of the poor by way of tax-financed government transfers or, as suggested previously, to introduce a flat demogrant-type pension for all.The issue of underfunding also holds for provident funds with comparatively high contribution rates, like the Singaporean Central Provident Fund.67 With a total contribution rate of 37 percent of employees’ wages, the Central Provident Fund is used to cover several types of social expenditure, such as co-financing the acquisition of property, saving in preparation for old age and covering the unforeseeable needs of health care. The Government has implemented several measures to help workers maximize benefits from the Fund through improvement of human capital and social transfers, like the Workfare Income Supplement scheme. Nevertheless, due to the inherent nature of a defined-contribution scheme, even with high contribution rates, benefit levels in old age may not always be adequate to meet post-retirement needs in the long term, especially for workers earning below the median income. The Silver Support Scheme supplements the retirement income of older Singaporeans who had low income throughout their life and have little or no family support.Most ASEAN countries run their old-age income protection systems with too-low legal retirement ages, and most are underfunded in the sense of providing only insufficient replacement rates. Simple parameter adjustments, such as increasing the legal retirement age, would help to reduce financial pressures substantially. Nevertheless, additional resources will have to be shifted into old-age provisions inevitably. Provisions entail not only old-age income security, in a narrow sense, but additional resources for health and old-age care. In these respects, all ASEAN countries face similar challenges.Trust is a central element in the future of social security. It is particularly important with pensions due to the deferment between the payment of contributions and receiving benefits. It is important to understand that not only must members trust the scheme but the society at large must also trust it. If the scheme is universal, “members” and “society” are basically identical; however, some members of society are typically excluded or prefer to be excluded from the scheme, especially economic and/ or intellectual elites and possibly certain civil servants may be excluded. Their support in the wider societal discussion, including governance and management, is important. Otherwise, schemes directed at the poor might become poor schemes.Developing an overarching culture of transparency, including the maximum possible alignment of different schemes, is essential to gain the trust of the various schemes’ members. Improving administrative effectiveness and efficiency will also contribute to the creation of trust. While demanding compliance with the statutory scheme rules from their members (individuals and employers), social security institutions, including pension scheme administrations, must see themselves as serving their members’ interests.Voluntary schemes, including matching contribution schemes, have shown only limited success. In many countries,the effort was a failure. The core reason is that the implied minimum personal contributions are often subjectively judged too high by contributors while, objectively, they are too low for “producing” a decent pension in old age. The combination of these two – subjective and objective – components appears to be triggering low compliance with offers by private insurance providers and governments. It is often the middle and upper-middle classes and higher67The Central Provident Fund is not the sole source of income security for Singaporeans. It is complemented by measures to improve human capital and help workers earn more during their working lives, as well as various, though limited, government transfers, including a social assistance scheme.76who can participate and benefit, while people who are poor are being left behind. For most Asian societies,matching contribution schemes appear to be a waste of money due to their favouring those who do not need it. Any tax-financed matching contributions thus should better be used for channeling them into tax-funded demogrants.Creating and maintaining a “social security culture” is no easy task. It requires permanent, institutionalized effort. More fundamentally, its creation requires changes in Asia’s prevailing productivist development model, which has been economically successful over several decades, albeit at various degrees in the different countries of this report. Nevertheless, it is fundamental to recognize that this model, if technically understood as running permanent trade surpluses relative to the rest of the world, results in capital exports requiring at least one country to accumulate national debt in relation to the rest of the world. This country has in the past been the United States and, less so with volatility over shorter periods, countries in Europe. The trade-surplus policy of Asia (its productivist orientation) has begun encountering growing international resistance. The United States has declared that it is no longer ready to accumulate debt the same way it had over so many years in the past, and Europe has decided to run trade surpluses the same way as Asia. With the main economic regions of the world now oriented towards policies aimed at, or actually running, trade surpluses, this can only end up in an international conflictual situation – unless other regions, like Africa and/or Latin America,are ready to replace the United States in the role of the debtor.Social security expansion, increasing coverage and adequacy of pensions could be part of a macroeconomic policy to ameliorate the conflictual international trade dynamics: Increasing national pension and other social security benefits would increase national imports through the impact of standard elasticities between national demand and imports and thus contribute to closing the trade surpluses and reducing capital exports. For Asia, this could turn into a win-win situation: International trade tensions would be reduced, and national mass incomes, including social benefits, would increase. At its core, this switch would imply to dispose of the employed productivist model based on low labour costs and international demand. It would be replaced by a model that rests upon the productivity effects of high(er) labour costs, while international demand would still be a trigger for production but would be compensated by imports resulting from national demand.Hence, ASEAN should initiate international talks on the interdependencies of its social security requirements and the international trade implications. This is considered necessary to avoid “free rider” positions of the other big players, such as Europe and the United States.Expansionary social policy will require additional fiscal space (more tax and/or contribution revenue). To some extent, the additional fiscal space will come quasi-automatically because popular social programmes have thus far found support from the general public everywhere in the world. Extension of fiscal space is much easier if internationally coordinated because it minimizes the risk of private sector investors (“funds”) declaring such policy detrimental to growth, thus triggering their withdrawal of investments from national markets, and, potentially, creating exchange rate crises.The report has cited a number of technical possibilities for extending fiscal space, but it here recommends hedging any such policy against detrimental international reactions by way of proactive international coordination with major trade partners and/or competitors.774.2 Recommended priority areas per country and steps of coverage and adequacy reformBRUNEI DARUSSALAMThe legal retirement age is 60, at which age remaining average life expectancy is around 21 years (in 2015) but increasing to about 24 years by2050 (DESA Population Division, 2017b). For such long life expectancies, the legal contribution rate of 5 percent (by employee and employer each) is too low (by far) and still not sufficient after introducing the supplementary pension financed out of, at least, 3.5 percent by employee and employer each. Well-organized increases of the retirement age to 65 would help to solve half the problem. The Government could consider making the increase over a period of 10–20 years.An increase of the contribution rate of at least 8 percentage points or an equivalent tax-financed subsidy68 would be required for a relatively decent pension level at retirement. Any options to withdraw monies early should be reduced to a minimum, and at retirement, funds should be annuitized as a rule. The latest developments in Singapore in this regard could be an interesting source of inspiration.As a complementary policy, the retirement age of 60 could be indexed with the growing life expectancies of men and women (see box 7). Increasing the flat tax-financed pension significantly would require tax reform.With other required social programmes, such as public health or health insurance and elder care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach.As for the socioeconomic framework for pension policy, the Government must continue diversifying the economy in order to reduce its dependency on physically limited oil and gas reserves. The Government should coordinate its own production cost levels with those of Singapore, given that the Bruneian dollar is pegged to the Singaporean dollar.CAMBODIAThe country has no pension provision for the private sector – making it, along with Myanmar, an outlier in the region. Cambodia’s population structure is young and ageing fast but, according to United Nations population projections,it will not reach a critical “aged” or “super-aged” status by 2050 or even soon after. Thus, implementation of a national pension system, in line with the Constitution, is an issue of political will, economic development and strategic decision in terms of design: funded or non-funded, Bismarckian or Beveridgean, hybrid or tax-financed flat universal, plus individual private options.A pension system is available for the public sector and is fully funded through taxation. The legal retirement age is 60, at which time remaining average life expectancy is around 23 years at least (as of 2017). This average is projected to increase to at least 29 years by 2050. It is characterized as “at least” because public sector employees and civil servants are assumed to live longer than the average life expectancy of the whole population (DESA Population Division, 2017b).68 Numerous actuarial studies by the ILO prove that, as a rule of thumb, pension contribution rates equivalent to around 25 percent of underlying salary or income (or a corresponding mix of contribution rates and government subsidy), irrespective of fully funded or pay as you go, are required to finance a decent pension at retirement that would comply with the rules provided by ILO Convention No. 102, which is a minimal norm. For funded schemes, this rate increases with declining wage levels (in pay-as-you-go systems) and interest rates and rates of return (in funded systems).78In 2015, a government technical working group led a series of discussions and studies aiming at the development of a more effective and efficient social protection system for Cambodia. This work led to the approval of the National Social Protection Policy Framework 2016–2025, which includes: (i) development and implementation of an old-age pension scheme for the private sector; (ii) implementation of an old-age pension scheme for non-poor workers in informal employment; (iii) unification of pension schemes under one management system and expansion of coverage of civil servants; (iv) transfer of the civil servant scheme from a non-contributory to a contributory system; and (v) development of a legal framework for the management, monitoring and implementation of all public pension schemes to ensure consistency of pensions.The launch of the pension scheme for workers and employees in the private sector was announced for 2019. The plan is for the scheme to be based on a partially funded mechanism, funded equally by workers and employers and integrating a grandfathering provision to ensure that persons already near to retirement age can enjoy some sort of income protection in old age from the early stages of the scheme. The second element relates to the implementation of a voluntary, old-age pension scheme aimed at covering workers in the informal sector who have unstable income sources. The Government is also studying the possibility of launching a tax-funded pension for persons with income above the poverty line, most likely using the IDPoor69 as a reference.The financial strategy of the Government to apply these measures in the near future relies on reforming the pension scheme for civil servants, aiming to change from the current non-contributory and publicly funded system to a contributory one, and employing a contribution rate that is consistent among all the workers in public, private and informal sectors.Reforms will be based on a future study that will specify technical aspects related to contributionbased old-age pension systems and discuss the most suitable retirement age for the system to ensure labour productivity and the financial sustainability of the pension system as well as the types of schemes that are appropriate for the Cambodian context. The portability of benefits for the targeted population (workers in the public and private sectors and self-employed) is one of the concerns regarding the draft of the new law in terms of ensuring that the system is not a barrier to labour movement.The pensionable age should be 60 for men and women, with an option to increase it gradually (much) later. Effective implementation of the system has been hindered by limited administrative and economic capacities. And for the time being, the wide split between the public sector pensions and the non-existent private sector provision contributes to hardening Cambodia’s unfair income and wealth distribution. It is advisable to hedge civil servants’ privileged position without affecting their loyalty to their administrative duties, including the establishment of a pension scheme for the private sector. An increase of civil servants’ legal retirement age would be advisable as well.Because setting up a contribution-based pension system for the private sector takes time, the Government could consider complementing the process in the meanwhile with a tax-financed and flat-rate basic tier, which could easily be integrated with the private sector system when it is ultimately operating (box 10).69 I DPoor provides regular up-to-date information on poor households to a large number of government and non-gov-ernment agencies to help them target services and assistance to the poorest and most vulnerable households. These agencies can also use the IDPoor information to target poorer communes or villages, according to their relative poverty levels (see http://www.idpoor.gov.kh/en/home).79Box 10 Addressing the foremost challenges in the Cambodian old-age income security agenda: Cost projections for providing a universal floor of old-age protection Establishing a contribution-based pension scheme takes time and would certainly have low initial coverage rates in Cambodia, mainly due to the high density of a typical work arrangements and short-duration contracts in the labour market. Traditional social insurance could not provide income security in the short and medium terms for Cambodians who contribute, on average, only seven months per year to a pension scheme, even when employed in the formal sector. This indicates that even the core of the formal sector engages in informal employment for almost half the year. According to National Social Security Fund data, workers in the private sector, including a majority of women in the garment sector, frequently leave and re-enter employment throughout the year. In recent years, the average duration of formal fixed-duration contracts was reduced from one year to three months(ILO and AFD, 2018). Figure 10 shows that the bulk of workers covered by the National Social Security Fund is aged 19–45. The formal workforce is not only slowly accumulating funds for old-age due to their formal career breaks but they also stop contributing at an early age, which significantly compromises the amount of pension they will receive upon retirement (ILO and AFD, 2018). Figure 10. Distribution of Cambodian workers contributing to the National Social Security Fund, 2015Source: ILO and AFD, 2018. Given such a challenging national context, the best and perhaps only way for Cambodia to provide citizens with old-age income security in the short and medium terms is through the establishment of a universal floor for old-age social protection. Such a floor can be achieved with the adoption of a demogrant or social pension scheme. It would provide all citizens at (and older than) a certain age with a flat-rate, tax-financed pension, thus ensuring them income security regardless of previous links to the formal labour market.The proposal of such a scheme was recently technically backed by an ILO actuarial study, which projected the cost of a universal flat-rate pension at the level of the median national poverty line (ILO, 2017b). According to the study and despite ageing-induced growing numbers of beneficiaries,80the costs would absorb only slightly more than 1 percent of GDP in the short and long runs as long as the pension is regularly indexed with the CPI and the economy grows in real terms. Under this policy, it would be expected that the expansion of the contributory system in the long term would have a central role in ensuring the adequacy of pensions. Figure 11. Cost of a monthly universal pension in Cambodia, 2018–2117 (% of GDP)Note: Monthly pension amount of 75,000 Cambodian riel (roughly US$19), adjusted to wages before retirement and to inflation after retirement.Source: ILO, 2017b.The Government should integrate grandfathering possibilities under an insurance-based system. Long up-front savings periods or long periods required for benefit eligibility should be avoided, unless accompanied by rules that allow for a pay-out of pensions to well-defined older cohorts from the onset of the scheme. Such design improves the general public’s trust in the system, not the least because it demonstrates effective functioning while improving the living conditions of eligible persons and households. Even with other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach. The Government will need to continue relying on co-support of intra-family transfers for some time. In practice,these transfers will be effective only if general income levels increase and the tide of economic growth lifts all boats equally. Even when and where this might be realized, Cambodia will be exposed to further urbanization,making it increasingly critical for families to assume their classical role of mutual help. As for the socioeconomic framework for pension policy, the Government must continue diversifying the economy to reduce its dependency on a shallow textile industry or exploitation of offshore oil reserves. Tourism is becoming a major generator of income. Because the country is small, the economy will not have much room to go on an independent development path – the best way to prosper is to adjust to the economies of its bigger neighbours.81INDONESIAThe legal retirement age is 56 but is scheduled to gradually increase to 65 by 2043 (ILO, 2014a). For now, the remaining average life expectancy is 21 years but is projected to be 15 years when the retirement age changes in 2043 (DESA Population Division, 2017b). By adjusting the retirement age, the Government has moved to lighten the burden of future pension financing. With 3 percent of earnings, the collected contribution rate is, nevertheless, far too low, unless the Government subsidizes schemes in the future out of taxation. Further increases of the contribution rates appear overdue; the target should be an overall rate of around 25 percent of earnings to be replaced in old age. This would also be required to increase the replacement rate after 15 years of contribution payment, which is currently just 15 percent. At the same time, however, such tax-financed increases might be perceived as ambivalent of nature because only around one quarter of Indonesia’s older population receives a pension payment. Whether recent legislation effectuates further entitlements and pension payments to bigger shares of the older population remains to be seen.Alternatively, or as a complementary policy, the Government could introduce a flat demogrant for all citizens aged 60 or older (or at an appropriately chosen age), such that fiscal issues remain at controllable levels (at the level of an adequately defined national poverty line). For Indonesia, regional poverty lines would probably be adequate financially and politically (box 11).Whether the retirement age applicable to the demogrant should be indexed the same way as it is in the current system, is to be debated. Clearly, it is an option but would not be considered urgent.Box 11 Fiscal implications for the adoption of an old-age demogrant in IndonesiaOne of the main recommendations for increasing the coverage rate of Indonesia’s old-age pension system is the introduction of a universal floor of social protection for older citizens in the form of a demogrant or social pension scheme. It would ensure adequacy of benefits for new pensioners as well as for those without former links to the labour market.In 2014, the ILO estimated the fiscal implications of a universal tax-financed flat pension for Indonesia (ILO, 2014a).As seen in figure 12, if set at the level of the national poverty line and paid to persons aged 65 and older, the scheme would start with a fiscal volume of 0.6 percent of GDP and, if adjusted to the CPI, would be close to 0.5 percent of GDP 50 years later. If indexed to salary growth, the costs would rise to 2 percent of GDP in 50 years. If the Government decides to set 60 years as the eligible age, the costs of covering all persons aged 60 or older would correspond to 1 percent of GDP in the beginning and, if indexed to CPI, would cost 0.7 percent of GDP in 50 years. If indexed to salary growth, costs would rise to 2.6 percent of GDP in 50 years. Based on the illustrative findings of this exercise, it becomes evident that under no reasonable setting would such a policy be financially unfeasible.82Figure 12. Costs of providing a universal old-age pension in Indonesia to persons aged 60 or 65 or older, 2014–62 (% of GDP)Source: ILO, 2014a. Implementing and regularly increasing the flat tax-financed social pension, assuming it would be fixed at significant levels, would require careful assessment and further analyses of public finance in Indonesia. In this sense, a significant reshuffling of public resources or even some level of tax reform could be required. Finally, taking into account the diversity of existing schemes that deal with old-age income security in Indonesia, including the social insurance fund, the provident fund and the work termination entitlements, as specified in the labour law, it is highly recommended to comprehensively reform these schemes. This could be achieved through one consistent policy solution that not only addresses pensions but also the benefits of other social security branches, such as unemployment insurance and the benefits as stipulated by the labour law for redundancies. With other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach. For a transition period in which reforms are being prepared and implemented, Indonesia will need to continue relying on co-support of intra-family transfers. However, despite further urbanization and with fertility rates declining only slowly, old-age pension reforms will probably not be of a critical nature because co-residency and similar arrangements will continue to be possible. As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy and supporting the nation’s education system.LAO PEOPLE’S DEMOCRATIC REPUBLICIn terms of average wages, the Lao People’s Democratic Republic is the poorest ASEAN country. It provides a public sector pension scheme as well as a private sector one. But the coverage is extremely low. In 2010 (latest available data), only 1.3 percent of the working-age population was contributing to a pension scheme, while only 5.6 percent of persons at or older than the retirement age received a pension payment (ILO, 2017c). The legal retirement age is 60 for men and 55 for women, at which remaining life expectancy is around 15 years for men and 24 years for women. If the retirement ages remain unchanged, these lifespans will increase to about 19 years for men and 24 years for women in 2050 (DESA Population Division, 2017b).83For such long life expectancies, the current legal joint contribution rates of 11.5 percent, by employers and employees together in the private sector, and 16.5 percent in the public sector, are too low (ILO, 2017c). As a rule of thumb, the target total contribution rate should be in the order of 25 percent of protected earnings for both schemes.Given the extremely low coverage, it would be advisable to reform the small social assistance schemes and turn them into a demogrant scheme for all, to be designed in a way that it can be combined on an equitable basis with the existing social insurance schemes for the private and the public sectors (box 12). Demography should not be a hindrance to implementing such a scheme. The population is ageing, but the resulting changes in the population structure are not considered dramatic until 2050 or soon after. The country’s dominant old-age financing problem is not of a demographic nature but almost solely of an economic one. To address the most urgent evidence of old-age-related poverty, the Government should institute a demogrant for all persons at and older than 65 years and possibly index this age with general life-expectancy to avoid overstretching the available fiscal space.Box 12Policy dialogue with the Lao People’s Democratic Republic: Two scenarios of cost projections for an old-age demogrant and other recommendations As part of a social protection national policy dialogue assessment within the relevant government bodies and the ILO and based on the conclusions of that dialogue, scenario calculations were made in 2017 for the establishment of targeted old-age social pensions in the Lao People’s Democratic Republic. Such pensions would cover workers outside the National Social Security Fund pension system, who represent the vast majority of Lao formal workers (at 86 percent) as well as workers in the informal economy and persons without prior links to the labour market(ILO, 2017c).The first scenario for introducing social pensions (figure 13) consisted of gradually establishing a tax-financed and pension-tested demogrant scheme. The pension, defined at 50 percent of the poverty line,would be paid to persons aged 70 or older from the first year, extended to those aged 65 or older after five years and then extended further to those aged 60 or older after 10 years. Under this scenario, the benefit costs of the scheme would start at a meagre 0.21 percent of GDP and reach only 0.66 percent of GDP in its maturity stage, by 2030. Projections of the seemingly more financially conservative second scenario, under which eligible age begins at 75 and gradually falls to 60 by 2029, indicates lower costs in the short term but similar costs to the first scenario in the medium term, by year 2030 (ILO, 2017c). These estimates are consistent illustrations of the low costs associated with ensuring old-age income security on the national level in the Lao People’s Democratic Republic. Figure 13. Cost projections for two scenarios of old-age universal social pensions in the Lao People’s Democratic Republic, 2017–30 (% of GDP)Source: ILO, 2017c. 84With regards to the country’s old-age income security system, the Lao assessment-based national dialogue on social protection highlighted another important policy and implementation issue: The financial sustainability of the pay-as-you-go scheme for the workers in the public sector is dangerously weak and, as the Government continues to merge public and private sector schemes, the risks associated with the lack of reserves of the former will possibly spread to the latter and threaten the newly merged social security system. The main recommendation of the assessmentbased national dialogue would be to ensure solid actuarial design and sufficient fund reserves before merging the public and private (State Authority of Social Security and National Social Security Organization) old-age pension schemes. This would ensure a sustainable financial state for the overall pension programme and protect the Social Security Organization’s reserve fund, which was accumulated by contributions from employers and workers in the private sector and explicitly designated to ensure the long-term financial sustainability of private sector pensions (ILO, 2017c).With other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach.For a transition period in which reforms are prepared and implemented, the only choice is to continue to rely on the co-support of intra-family transfers. As it appears over the foreseeable future, neither declining fertility rates nor urbanization will be obstacles to co-residency and similar arrangements in the future, despite more and more households with only one or two children in the next decades.As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy to reduce its dependency on shallow textile fabrication or similar industries. As a small country, the Lao People’s Democratic Republic will not have much room to go on an independent economic development path – the best way to prosper is to adjust to the economies of its bigger neighbours. In this regard,the Lao People’s Democratic Republic is in the same situation as Cambodia.MALAYSIAMalaysia’s demographic situation and expected future development are not as dramatic as those of Singapore and Thailand, but trends are somewhat similar. The legal retirement age has been increased to 60 years from 55 for the Employees Provident Fund and from 58 years for the Public Sector Pension Fund. For the time being, this increase appears sufficient, but further increases might be necessary in the future, depending on economic, productivity and labour market developments. The average life expectancy at age 60 was around 20 years in 2017and is projected to increase to 23 years by 2050 (DESA Population Division, 2017b). The provident fund contribution rates of 23–24 percent70 by private sector workers and employers and of 17.5 percent under the public scheme appear to be high in comparison with the low rates in several other ASEAN countries. However, the withdrawals are not limited to old-age,and thus what goes into the old-age provision is definitely too low to finance a decent income after retirement.This situation is aggravated by the fact that other and early withdrawals can be made from the retirement account. At present, only about 70 percent of the theoretically available monies,on average, remain for the time of retirement (EPF, 2015). Malaysia will have to face further increases in70 For employees earning less than MYR5,000, contributions are 11 percent by the employee and 13 percent by the employer, while for employees earning more than MYR5,000, contributions are 11 percent by the employee and 12 percent by the employer.85contribution rates, possibly in combination with increases in the legal retirement age. As with other countries, any options for the early withdrawal of monies from provident fund savings should be reduced to a minimum; at retirement, funds should be annuitized as a rule. Malaysia should take advantage of the discussions and innovations going on in Singapore.As a complementary policy, the Government could consider a flat tax-financed demogrant for all citizens aged 60 or older, at the level of an adequately defined national poverty line. In a first step, the retirement age for the demogrant could be tentatively chosen to keep the public financing from being overstretched. Implementing a flat-tax financed pension at levels to eradicate poverty would require tax reform aiming to increase Malaysia’s fiscal space.With respect to the option of deviating a certain share of collected contributions into a new social insurance scheme, see due reference made to the consideration for Singapore.In 2017, self-employed workers constituted 18 percent of total employment. The Government has tried to counter the low coverage of the self-employed group but with limited success because only about 1 percent of self-employed persons participates (World Bank, 2014). A number of private retirement schemes were introduced in 2012 to increase pension coverage among workers outside the mandated schemes; they had 250,000 active members in 2017. Even though take-up in the private retirement schemes has been increasing, with the Government providing annual tax relief for individual and employer contributions and a one-time MYR1,000 incentive to qualified youth members, the coverage of self-employed workers remains low (PPA, 2017).With other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach.One option to lift the financial burden from the provident fund would be to change the sickness and maternity benefits, while maintaining the contribution rate,in such a way that these programmes are financed through taxation in the future.As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy to reduce its dependence on the physically limited oil and gas reserves and palm oil and to coordinate its own production cost levels with those of neighbouring Singapore.MYANMARAfter decades with marginal formal social security,under which only about 1 percent of the population had been participating, the country’s pension system is still in status nascendi. Thus, pensions are uncommon as a source of old-age income. Most older people depend on support from their children, and a significant number of them still work.With the Social Security Law of 2012 (see the Myanmar country brief ), the public pension scheme and social insurance were revised, setting the legal retirement age at 60. The public sector scheme is financed through taxation, whereas the social insurance pension is financed out of a total contribution rate of 6 percent,by workers and employers together. However, the invalidity, superannuation and survivor insurance benefits have not become effective yet, and thus there is no experience that can serve as evidence to whether the new social insurance will be a success. Nonetheless, to finance a decent pension that is in line with ILO norms, the contribution rate will need to be increased in the future, depending on whether the system is intended to be fully-funded or fully pay as you go.7171 Myanmar’s Social Insurance System covers medical treatment under the Health and Social Care Insurance Benefit for retired persons who paid at least 180 contributions before their retirement. Also, persons who retired from work can voluntarily register under the system, in which case they will be entitled to receive the same cash benefit.86The country’s population is also ageing, but its structure and probable future development is not especially dramatic in the foreseeable perspective. Life expectancy at age 60 is currently estimated to be 17 years, which is expected to increase to 18 years in 2050 (DESA Population Division, 2017b). However, these macro-demographic data are of little relevance for the debate on Myanmar’s pension system because its coverage rate is only 3 percent (see the country brief ). The system lacks attraction also because paid-out pension benefits are low. Increasing both coverage and benefit levels is a function of administration and governance and of future economic developments. The Government should urgently start planning the development of a comprehensive pension system that guarantees income security in old age as a necessary element to counter the foreseeably growing financial vulnerability of older people in Myanmar. Without doubt, expanding the number of older people benefiting from a pension must be a priority, and reform efforts should mirror this focus (box 13).Box 13 Myanmar: Recommendations for increasing old-age pension coverage Given Myanmar’s large informal employment rates and low contributory and savings capacity of the majority of the population, it is essential that the country’s old-age pension system encompass a scheme that provides tax-funded social pensions for older people to ensure old-age income security. In addition to working as a safety net, the social pensions could also work as replacement income for older people who have been able to start contributing but will need to wait longer before accumulating enough to receive decent minimum pension pay-outs. The demogrant can be pension tested, which would be a more cost-effective alternative to universal pensions. This means it would only cover persons who are not already beneficiaries in the old-age pension schemes. Costs of a pension-tested demogrant would vary, depending on the indexation of pensions and would decrease with time, assuming that the national economy keeps growing. ILO estimates show that providing a flat-rate pension to older persons would cost less than 1.5 percent of GDP if the eligibility age started at age 60. It would cost less than 1 percent of GDP if it started at age 65 (figure 14). Figure 14. Projection of costs of a pension-tested demogrant in Myanmar for persons aged 60 or older and 65 or older, 2017–97 (as % of GDP)Note: The figure shows different government cost rates, considering non-contributory pension benefits indexed by both wage and inflation and calculated for retirement ages of 60 and 65.Source: Clay et al., forthcoming.87PHILIPPINESAmong the ASEAN Member States, the Philippines’ pension system design comes the closest to a standard European model, which is also reflective of ILO Convention No. 102 or World Bank recommendations. However, it has yet to achieve the same degree of the effective and close-touniversal coverage that Europe and some countries in Asia (Republic of Korea and Japan) have. Also, government support to extend coverage has traditionally been limited, leaving the main responsibility to the social security administration. Only 0.5 percent of GDP is spent on old-age income security policies currently. There are two reasons for this: The first is that the Philippine population is still young on average. The population pyramid has a broad basis of children and youth, whereas the pyramid’s peak is small and pointed. Thus, there are relatively few older persons in the country to be covered. The second reason is that, of all older persons in 2015, only 39.8 percent received a small pension at all (ILO, 2017d).The legal retirement ages for the private sector social insurance scheme and the public scheme depend on the eligibility criteria, as reflected in the respective pension formulae, but are biased towards 60 years, with 65 optional in the public service (miners retire at age 55). Life expectancy at age 60 was estimated to be 17 years in 2017 and will be around 19 years in 2050 (DESA Population Division, 2017b). Progress is slow because of the detrimental health conditions that were assumed to continue prevailing. The contribution rate is 11 percent in the social insurance scheme, including maternity and sickness, and 21 percent in the public sector scheme. If the defined-benefit pensions were financed on a pay-as-you-go or partial-funding basis, the contribution rate would likely be sufficient for the time being. Coverage extension, together with possible grandfathering of provisions, would of course require contribution rate increases over time, which would need to be boosted further to levels of around 25 percent and possibly even more, if the pension income replacement rates are to be meaningful.The Philippine pension system has been characterized as regressive. An indication of this is the fact that of all families receiving a pension, considerably more than 50 percent is paid to the households in the top-three income deciles. Reforms are inevitable for coverage extension (box 14) and highly advisable in relation to benefit design. Also, the high fragmentation of the system must be reduced.Box 14 Social pensions: Estimated costs of increasing old-age coverage in the PhilippinesThe Social Pension for Indigent Senior Citizens targets persons aged 60 or older who receive no other public or private old-age pension benefits, have no regular financial family support and are “frail, sickly or with disability” (DSWD, 2010). Since 2011, a monthly pension of 500 Filipino pesos (PHP), or less than US$10, is given to the poorest older people to help improve their access to subsistence and medical daily needs (World Bank, 2018b). Thus, the scheme is of a small-subsidy nature rather than a full-income guarantee or security to beneficiaries.In 2016, less than 19 percent of the older population was covered by contributory oldage pensions, and less than 17 percent of the older population were beneficiaries of the social pension. In total, roughly 36 percent of older persons were covered by any old-age pension. Population projections for the country indicate that the share of older people in the population will surpass 14 percent of the total by 2050, which will correspond to roughly 22 million older persons – essentially double the 7.3 percent, or 7.5 million older persons, in 2015 (DESA Population Division, 2017b). In the face of this rapid growth in the number of older persons, the Filipino congress has shown political will to increase coverage and benefit88adequacy of the small current pension amount. In fact, senate bill 1865 was proposed in 2018 to get congressional approval for doubling the social pension to PHP1,000 and targeting not only the indigent poor persons aged 60 or older (PIA, 2018). Because coverage extension of contributory systems struggles with some of the common obstacles among ASEAN Member States, extending coverage of the social pension for all people aged 60 or older at the level of an adequately defined national poverty line is strongly recommended. Within an assessment-based national dialogue, the ILO has estimated various cost scenarios of universal, means-tested or pension-tested demogrants (figure 15). The estimates show that the provision of PHP1,000 per month for all persons aged 60 or older would cost around 0.54 percent of GDP at inception. With benefits indexed to the CPI and adjusted every five years, the 2028 costs would correspond to 0.4 percent of GDP (ILO, forthcoming a). Based on such estimates and given the structure of the country’s population pyramid, reforming the social pension to increase old-age pension coverage would not be costly in the short run. In the future, it would require further allocation of fiscal space, and coverage extension would be dependent upon sustained Government will and support (ILO, forthcoming a). Figure 15. Scenarios of costs projections of the social pension in the Philippines, 2017– 28 (% of GDP)Source: ILO, forthcoming a. With other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach. This is even more pressing because public spending amounts currently to between 1.5 percent and 2 percent. More spending on health is as urgent as it is on pensions because balancing the mutual interests of the social security stakeholders (contributors, taxpayers, beneficiaries, the health industry and others) is only possible by way of careful statistical monitoring.89As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy.Box 15 Expanding coverage to self-employed workers in the PhilippinesTo increase coverage of old-age pensions among self-employed workers, one suggestion under consideration is that the Government should match contributions made by workers to the Social Security System. The basic assumption is that by providing a 50 percent subsidy to the total contributions, more workers in informal employment could afford the regular contributions and would be incentivized to join the scheme. This approach, however, has been questioned in terms of its likely effectiveness and cost by various agencies, including HelpAge International and the World Bank. They suggest that a more effective alternative would be to provide non-contributory pensions (COSE and HAI, 2017; World Bank, 2016).Recent analysis by the World Bank concluded that subsidizing contributions would not be the most effective instrument to expand coverage. There are two fundamental concerns: The first being that poor and vulnerable workers are unlikely to be able to afford even half the contribution, and the second being that fiscal costs could be substantial, depending on the benefit level, the subsidy and the targeting mechanism. An initial assessment suggests that the annual cost of a pension benefit at the current individual poverty threshold and targeted at the bottom 40 percent of the labour force could cost as much as 0.6 percent of GDP. This would be similar to the scale of financing a universal social pension (figure 15). A universal pension would also benefit all older people, whereas a matching subsidy would mainly benefit a small portion of relatively better-off self-employed workers. For these reasons, the World Bank analysis finds that broadening the scope of beneficiaries eligible for the existing social pension could increase effective coverage in an immediate and more cost-efficient manner.Source: COSE and HAI, 2017; World Bank, 2016.SINGAPOREThe minimum statutory retirement age in Singapore is 62, and employers are required to offer reemployment to eligible employees up to the statutory re-employment age of 67. The minimum legal age to receive the monthly retirement pay-outs from the Central Provident Fund is 65. However, beginning at age 55, workers who have met specific balance thresholds can withdraw amounts from their Central Provident Fund savings that are above the specified threshold. Considering that life expectancy at age 65 was 21.1 years in 2018 (Singapore Department of Statistics, 2019), there is a risk that the average allocations from the Special Account even after merger with the Ordinary Account and the Retirement Account at age 55, may be insufficient to meet pre-retirement living standards throughout the post-retirement life. It is interesting that of all the active Central Provident Fund members who turned 55 in 2018, more than six in ten had at least the Basic Retirement Sum.Building on the steps of recent policy decisions, a further increase in the allocation to the Retirement Account would most likely contribute towards ensuring that a larger share of members would have access to a more adequate level of monthly income protection. Such an increase, however, would most probably also require further reducing the use of accumulated funds before retirement. A balance must also be ensured, so that increasing allocations for retirement do not negatively affect the funds available to meet other social risks and contingencies. A more precise definition of those solutions would naturally depend on detailed modeling (which was not undertaken in the context of this report).A closer alignment between the minimum withdrawal age and retirement age could also be an important step in this direction. This could be achieved by a closer alignment between the90minimum statutory retirement age and withdrawal age, which are not linked today. In view of the long life expectancy in the country and the healthy life expectancy at the time of retirement, as well as the perspective of social partners by having discussions with them (employer and worker representatives), the retirement age can be increased. ILO Convention No. 102 establishes that “the prescribed (retirement) age shall be not more than 65 years or such higher age as may be fixed by the competent authority with due regard to the working ability of older persons in the country concerned”. An increase in retirement age would not go against the principles stated under the Convention. Alternatively, or as a complementary policy, the Government could consider introducing a flat demogrant, such as found in Brunei Darussalam, for all citizens and residents aged 65 or older and at the level of an adequately defined national poverty line or the median income level. This universal benefit could be designed as an element of the Silver Support Scheme. In addition to supporting persons in the lowest quintile of income distribution, it would contribute towards increasing the level of adequacy of persons who, despite benefiting from monthly pay-outs, might still face challenges with their pension adequacy. The eligibility age for the demogrant could be indexed with the statutory retirement age and growing life expectancies of men and women. Introducing a flat, universal, tax-financed pension might require a tax reform that aims to increase the Government’s fiscal space. Figure 16. Tentative cost estimates of a non-contributory pension in SingaporeSource: Asher, 2013. Figure 16 reflects the tentative cost estimates of such a non-contributory, budget-financed pension for people older than 65. The benefit level is taken between 10 percent and 50 percent of the median gross monthly income of employed residents, excluding the Central Provident Fund employer contributions. Two scenarios are projected: (i) 3 percent wage growth and 3 percent GDP growth; and (ii) 1 percent wage growth and 2 percent GDP growth, in tune with real wage growth in the early 2010s. The tentative projections indicate that with real wages and real GDP assumed to grow at 3 percent, the monetary cost of a non-contributory pension would be between 0.13 percent and 0.67 percent of GDP in 2030 (Asher, 2013). The extent to which it would be supportive to such measures to increase the ordinary wage ceiling, which is relatively low, at slightly above the national average wage level (Trading Economics, 2018)and to change the rules for calculating the additional wage ceiling must be left to technical analysis.91Singapore might also want to consider the option of going a step further and assessing the feasibility and potential impact of transforming at least a component of members’ individual accounts into a fully social insurance mechanism – following what countries like India or Indonesia have done.With other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by monitoring social spending and its financing through a social budgeting approach.As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy and quickly adapt it to regional and international demands. The ageing of the labour force could be countered by a continued active policy of inviting commuters72 to work in Singapore’s economy – currently more than 40 percent of Singapore’s total labour force consists of non-citizens (table 13).Table 13. Singaporean labour force, by citizenship status, 2010–17 (thousands)Total labour forceCitizens and permanent residentsRatio of total: citizens and residents201020112012201320142015201620173 135.9 3 237.1 3 361.8 3 443.7 3 530.8 3 610.6 3 672.8 3 657.02 047.3 2 080.1 2 119.6 2 138.8 2 185.2 2 232.3 2 257.6 2 269.71.51.61.61.61.61.61.61.6Source: ILO, 2018d; Ministry of Manpower (Singapore), 2019.THAILANDThe legal pensionable age is 55 under the social insurance fund and 60 under the voluntary savings fund and the old-age allowance, which is a universal social pension,or demogrant. At age 55, the remaining life expectancy is around 27 years, which is expected to grow to 30 years by 2050. At age 60, remaining life expectancy was estimated to be 22 years in 2017 and will become 25 years by 2050 (DESA Population Division, 2017b).For such a long life expectancy after retirement age, the legal contribution rate of 7 percent for social insurance, with3 percent from employees and employers and 1 percent from the Government, is far too low. This holds also despite the recent modest increase of the contribution collection ceiling, from 15,000 Thai baht (THB) to THB20,000. The increase aimed to guarantee a decent minimum replacement rate of the pension pay-out. In addition to the adequacy problem, the coverage provided under the contributory scheme is still limited to persons working in the formal sector.In the past, the Government tried to improve the situation of older people by introducing an additional voluntary tier and supporting it through a matching contribution system – but with no significant effects recorded. Only after implementing the nationwide and unconditional social pension did the coverage situation improve substantially. The general level of this pension, however, is too low. The pension system is characterized by its subsidy nature to (assumed) other sources of old-age income (such as through the extended family) than by its ability to provide sufficient levels of income security.72 “Commuters” are foreigners who come to work in Singapore temporarily.92Given Thailand’s imminent ageing process (which is similar to the situation in Singapore) and its high coverage with low benefits, future increases in the legal retirement age will be inevitable to enable increases in benefits structurally. More specifically, the retirement age for social insurance will need to be increased in a well-organized, step-by-step process, first to 60 years and later even to age 65. As in other countries, scheme members, generally belonging to the poorer strata of the population, might claim that their group-specific life expectancy is considerably lower than it is for the overall population. After an increase of the retirement age, many scheme members might pay contributions but not reach pensionable age. This and related arguments must be taken seriously; they may indeed be justified and can be countered only by pronounced workplace safety policies, with visible positive effects on workers’ labour conditions and on the number and intensity of accidents and workplace-related diseases.73Well-organized increases of the retirement age can solve only half the problem of future pension financing; relying on a comeback of higher rates of return on funds may turn out to be in vain. To achieve higher rates of return by improving the reserve funds’ investments is advisable but will not be sufficient to make major contributions to solving the financing issues of Thailand’s overall pension system (box 16).Box 16 Improving financial sustainability in Thailand’s old-age pension systemExtensive parameter reform is needed in Thailand’s Social Security Office scheme, which is a mandatory contribution-based old-age pension scheme for workers in the private sector. The Social Security Office reserves run the risk of being depleted due to the low statutory retirement age (60), the low contribution rates of the scheme and its rules of benefit calculation, which are only on the basis of years of contribution and level of earned salary before retirement. The relatively large shares of older people in the population and the relatively drastic Thai population ageing trends put real pressure on the current rules of the scheme, according to actuarial and other technical analyses (ILO, 2016 c and e). According to the current rules, a “standard member” of the Thai Social Security Office scheme may receive:after 15 years of contribution payment, a pension at 20 percent of eligible salary;after 20 years of contribution payment, a pension at 27.5 percent of eligible salary;after 30 years of contribution payment, a pension at 42.5 percent of eligible salary; andafter 40 years of contribution payment, a pension at 57.5 percent of eligible salary.Although the scheme’s formula for benefit calculation is in line with ILO Convention No. 102, the parameter rules allow for contribution periods to guarantee income for retirement periods that are equally long or even longer. Life expectancy at age 55 in Thailand is 26 years and, by 2050, it will grow to 30 years (DESA Population Division, 2017b). For instance, a standard worker might contribute for around 30 years and receive a pension of about 40 percent of (the last five years’) salary over another period of about 30 years. The current parameters will potentially allow for a large number of workers to contribute insufficiently to retire at a relatively young age and receive a pension for periods that are equally long to contribution years, which will lead to reserves abruptly declining between 2034 and 2054, as figure 17 illustrates.73 The retirement age increases will be necessary in private (savings) systems as long as annuities are to be paid at decent levels.93Figure 17. Projection of Thailand’s Social Security Office reserves levels, 2014–54 (baht)Source: ILO,2016c.Improving financial sustainability of the Social Security Office scheme is essential to guarantee future income security for contributors in the private sector and to ensure its solvency. Hence, increasing (i) the statutory retirement age, (ii) the contribution rates and (iii) the contribution collection ceiling and adjusting the pension formula to possibly include grandfathering of pensions as well as more government engagement in the old-age social security agenda are basic yet fundamental steps for ensuring sustainability of the Social Security Office scheme in the near future (ILO, 2016c).As a complementary policy, the Government should consider increasing the existing demogrant for all citizens aged 60 or older at a level of a national poverty line to be determined. The retirement age (60) could be indexed with the increasing life expectancies of men and women. Significantly increasing the flat tax-financed social pension would require tax reform, and the establishment of indexation mechanisms is essential. As for the old-age allowance and Social Security Office scheme, the lack of their automatic indexation drastically contributes to the reduction of beneficiaries’ purchasing power. The Government recently announced the establishment of a mandatory provident fund. Although the intention is to increase old-age income security by increasing private savings, the key problem with this measure is that it was taken without fully considering its interdependency with the sustainability and role of the social insurance pension. The Social Security Office was closely monitored in the past through a regular sequence of actuarial evaluations, which all stipulate future substantial contribution rate increases. Yet, the new provident fund is planned to soon reach a contribution rate of 10 percent. The co-existence of these two schemes is endangered because cumulated contribution rates might turn out prohibitive for many people. To manage the policy more coherently, a National Pension Policy Committee was set up, which is a reasonable step towards improving governance. The lack of harmonization of the two schemes, in terms of their respective legislation but also relative to their administrative cooperation in practice,is a clear challenge that is compounded by the coordination requirements of the old-age allowance and possibly the civil 94servants’ scheme. An assessment of the options of creating a more harmonized system would help solve some of the persistent adequacy problems of Thailand’s system.With other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach.Before benefits can be increased to levels of international standards, the Government will have no choice but to continue relying on co-support of intra-family transfers. However, with low fertility rates, old-age pension reforms will be increasingly critical because co-residency and similar arrangements will no longer be a solution in practice as more and more single households emerge.As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy. Careful consideration should be given to initiating pro-labour-productivity policies.VIET NAMIn 2018, Viet Nam’s Communist Party approved an important policy tool to advance the social security agenda. The Master Plan on Social Insurance Reform is a framework to transition the oldage pension system into one with universal coverage, achieved through a multi-tier system (box 17).The legal retirement age is 55 for women and 60 for men, at which point remaining average life expectancy is around 20 years for men and 25 years for women. The life expectancy is estimated to increase to 24 years and 29 years for men and women, respectively, by 2050 (DESA Population Division, 2017b). For such long life expectancies,the legal total contribution rate of 22 percent is insufficient. But the Government is obliged by legislation to cover any emerging deficit. The accrual rate for pensions had been 3 percent per annum for women and 2.25 percent for men during the first 20 years of insured work, falling to 2 percent thereafter for both.Box 17 Multi-tier old-age pension system in Viet NamThe Vietnamese Government approved the Master Plan on Social Insurance Reform in 2018 (Resolution No. 28),signaling its commitment to reform the national old-age pension system. The Master Plan opened up valuable space for addressing parameters, design and implementation-related issues in the current and possibly revised pension schemes. In line with international standards and based on bilateral technical cooperation programmes with the Ministry of Labour, Invalids and Social Affairs, the resolution proposes a multi-tiered old-age system, including non-contributory and contributory elements. It consists of three tiers: a taxfunded demogrant in the form of a pension-tested scheme, a mandatory contribution-based defined benefit (covering workers in formal and informal employment) and a complementary voluntary scheme for the formal sector (table 14).95Table 14. Proposal for a multi-tier old-age pension system in Viet NamTier1: Tax-funded pensionTarget groupCharacteristicsAll residents who are not benefiting from a contributory pensionPension-tested social pensionTo ensure at least a minimum level of protectionNature of participationPension testedFinancing method General revenues2: Basic pension3: Supplementary voluntary pensionMandatory for formal sector and voluntary for informal sectorFormal and informal sectorsPublic pension plans, publicly managed defined benefitMandatoryOccupational or private pension plans for those willing to complement the Pillar 2 pensionsVoluntaryDefined benefit or defined contributionContributions, partially funded; funding levels to be determinedFully funded;individual contributionsSource: ILO, forthcomingc.Tier 1 is strongly aligned with the ILO Social Protection Floors Recommendations, 2012 (No. 202). Such a guarantee is particularly relevant for people who had no or only limited sufficient capacities to contribute to the pension schemes during their working lives, such as low- and irregular-income earners, workers in informal employment and difficult-to-reach groups, like agricultural workers.While tax-funded pensions are important for providing a basic level of income security, they may only provide modest benefit levels. Tax-funded schemes address the poverty and redistribution objectives of pension systems but lack the income-smoothing function. That is why mandatory public pension schemes financed by contributions are key to providing higher levels of benefits to ensure income security for large groups of the population. The second tier is a contributory tier, targeted largely at formal sector workers but should be designed in such a way that it can be extended to workers in informal employment. This tier is managed by the Viet Nam Social Security administration.Occupational pensions or other voluntary private pensions constitute to the voluntary tier 3. Such individual pension plans provide a supplementary pension on top of the mandatory public pension for groups seeking higher replacement rates at retirement (ILO, forthcoming c).In its small 2014 parameter reform, the Government reduced the women’s annual accrual rate to 2 percent, equalizing it with the male rate. With legal retirement ages unchanged, this implies a reduction for female pensions at the official retirement age. The universal pension-tested social pension paid by age 80 was complemented with a means-tested scheme, which pays a pension by age 60 to citizens with no family support or other caregiver. With this reform, the Government aimed to close the coverage gap for people aged between 60 and 80 years and not covered by social insurance. The monthly benefit amount paid was doubled but, of course, administrative input increased.96As in many other countries, the pension system in Viet Nam is complemented with a voluntary social insurance scheme for the self-employed and other groups of independent workers. Depending on their household incomes and based on the poverty line in rural areas, participants are entitled to state support in the form of subsidies to the monthly social insurance premiums. The subsidies equal 30 percent for participants from poor households (=46,200 Vietnamese dong (VND) per month), 25 percent for participants from households living just above the poverty line (=VND38,500 per month) and 10 percent for others (=VND15,400 per month).Overall, coverage in terms of numbers of contributors and beneficiaries is modest, while social insurance benefits are high – 60 percent or more after a full working life. The share of households with at least one pension recipient is largest in the top-income quintile, although the share of such households in this quintile is slightly less than 20 percent (see the country brief ).Given its demographic ageing and economic trends, an increase of the legal retirement age to 65 for women and men will be inevitable unless Viet Nam undergoes a productivity explosion due to technological progress. An important element of this reform is to equalize the retirement age for men and women to reduce the gender imbalances in the system and in the labour market.An essential element of any further reform must be to reduce the options for scheme members to withdraw their funds to a minimum. The percentage of members withdrawing their contributions is extremely high, which has a significant impact on the current and potential number of pensioners or coverage, the adequacy of benefits received and the financial situation of older persons who had decided to withdraw their contributions.And the low levels of active contributors, at 20.6 percent of the working-age population in 2015 (ILO, 2017d), indicate substantial space for increasing coverage by increasing compliance with legislation, where, when applicable, the law might need to be overhauled, especially in relation to SMEs. For this to materialize, it is important to foster the enforcement capacity of the Viet Nam Social Security agency, including through administrative measures like internal process optimization and investment in modern equipment, including IT hardware and software. These improvements would allow for following-up on employers’social security-related duties by effectuating better coordination with the labour inspectorate and by improving the Viet Nam Social Security administration’s external outreach to communicate its messages to its members and to enterprises. Special focus might have to be devoted to SMEs. Innovations in the cooperation between SMEs and the Viet Nam Social Security administration might have to be developed, for example, through the development of simplified but still effective mechanisms for the payment of taxes and contributions.A major step towards solving the coverage problem can be for the Government to consider extending its social pension system for persons aged 80 or older to all persons once they turn 65. Relying on positive coverage effects of the voluntary system might prove to be in vain, as has been the case in many other countries. Public monies should not be wasted on test-trying a matching contribution system. Also, tax incentives to support the extension of voluntary schemes would only directly benefit persons with higher incomes. Generally speaking, voluntary pension systems can be regarded as the “cream on top” of affluent societies with a solid public layer – they cannot solve the poverty problem of the poor.The ILO estimated the cost of a pension-tested benefit for all older people in Viet Nam (box 18). For a pension amounting to a quarter of the minimum salary of the public service, the cost was estimated to reach around 0.4 percent of GDP by 2030. Because much time will pass before the contributory scheme matures, introducing a tax-funded pension-tested scheme is an appropriate measure, at least as an effective interim solution to address the old-age income problem.97Unless the universal pension is made applicable to persons aged 65or older (at least), the Government will have no choice but to continue relying on co-support of intra-family transfers for a possibly long transition period. Due to the ageing population, lower fertility rates and urbanization, these transfers will be increasingly critical, because the co-residency and similar arrangements will no longer be available in practice as more and more single households emerge.Box 18 Simulated impacts and costs of reforming the social pension in Viet Nam An ongoing ILO study (forthcoming c) simulated the impacts and costs of two reform options for the social pension and found that they made a significant difference to the poverty and inequality status of older people. * Option 1: reduction in the eligibility age for the social pension to 60 years for women and 62 years for men, with the pension benefit of VND350,000 per month. * Option 2: reduction in the eligibility age to 67 years for women and men to reach the Master Plan on Social Insurance Reform target of 60 percent of the older population covered, with the pension benefit of VND350,000 per month. Option 1 would have the greatest impact on the national poverty rate; it would reduce the national near-poverty rate by 7.6 percent, compared with a reduction of 5.3 percent that would result from option 2.The estimated impact on the poverty rate among social pension recipients are significant and would make a meaningful difference to the lives of recipients as well as their households. Option 1 would reduce the poverty rate among recipients by 49.7 percent, while option 2 would reduce the poverty rate among recipients by 44.1 percent. The near-poverty rate would be reduced by 35.5 percent under option 1 and by 34.9 percent under option 2 (figure 18).Figure 18. Reduction in poverty rates at the national level and among social pension recipientsIf option 1 were implemented, it would provide old-age income security to a majority of older people, with a coverage gap of 1 million, or 6 percent, of older people. Its estimated cost would not exceed 0.55 percent of GDP by 2030. If option 2 were implemented, 6.5 million, or 40 percent, of people older than the statutory retirement age would have no pension. The cost is estimated to be 0.3 percent of GDP in 2030. After 2030, the costs will decrease,as GDP growth is projected to outpace the size of growth of the older population and more people are expected to receive a social security pension over time.Source: ILO, forthcoming cWith an extremely quickly ageing population, the question is not if Viet Nam can afford the cost of a universal pension but if the society and economy can afford to have such a considerable amount of98its population without income security. In 2030, the number of people aged 65 or older is projected to reach 12.2 million. The number of recipients of contributory pensions is projected to reach fewer than 2 million persons, implying that more than 10 million older persons would have no access to a pension.As for other required social programmes, such as public health or health insurance and elderly care, the Government would be well advised to enlarge its information base by substantiating its statistical systems and monitoring social spending and its financing through a social budgeting approach.As for the socioeconomic framework for the pension policy, the Government must continue diversifying the economy.4.3 Widening the fiscal space for social security, and the demands for fiscal space of other programmesASEAN is committed to extending social protection to all. In 2013, not the least triggered by the effects of the 1997–98 financial and economic crises as well as the 2008 Great Financial Crisis, the ASEAN Declaration on Strengthening Social Protection was adopted, pledging the completion of social protection floors as a priority to achieve growth with equity. In 2015, the ASEAN Member States agreed on a Regional Framework and Plan of Action for implementing the Declaration.Increased social security is also a core priority of the ASEAN 2016–20 Senior Labour Officials Meeting’s Work Programme. Member States are defining a monitoring framework to measure progress in extending social protection, using relevant SDG targets and indicators. This instrument will be used to gauge compliance of the Member States with the 2013 Declaration.To fulfill this commitment, multiple measures must be taken; among them is the extension of fiscal space. Today, at a time of fragile global recovery, fiscal consolidation and slow growth, the need for and the challenges to creating fiscal space have never been greater.Given the significance of social protection for human development and for achieving the SDGs, it is indeed imperative that governments explore all possible means of expanding fiscal space to promote the SDGs and national development through jobs and social protection. Options to expand fiscal space and generate resources for social protection are manifold, even in the poorest countries, like Cambodia or the Lao People’s Democratic Republic.74 The following highlights options of relevance for the region.Reallocating public expendituresThis includes assessing ongoing budget allocations through public expenditure reviews, social budgeting and other types of budget analyses; replacing high-cost, low-impact investments with those with larger socioeconomic impacts; eliminating spending inefficiencies; and/or tackling corruption. For example, Thailand has reallocated military expenditures to deliver universal health care; and Indonesia and many other developing countries have reduced or eliminated fuel subsidies and used the proceeds to extend social protection programmes (ILO, 2016d; Duran-Valverde and Pacheco, 2012).74 O rtiz, Cummins and Karunanethy (2017) described in some detail eight options to expand fiscal space, which are all supported by policy statements of the international financial institutions and the United Nations. In this report, we take advantage of their work but do not consider all options equally relevant for the countries included. Although the options are more or less correctly described, the how to of implementation remains unresolved.99Increasing tax revenues relative to GDP75This is achieved by altering different types of tax rates – on corporate profits and labour income, financial activities, property, consumption, imports, exports and natural resources – or by strengthening the efficiency of tax collection methods and overall compliance. Many countries are increasing taxes for social protection: for example, Mongolia is financing universal pensions, child benefits and other schemes from mining and gas taxes (Ortiz, Cummins and Karunanethy, 2017). Many countries use the revenue from lotteries or other earmarked taxes (such as a tax on alcohol or tobacco) to supplement the budget for welfare services.Expanding social security coverage and contributory revenuesIncreasing coverage and therefore collection of contributions is a reliable way to finance social protection, freeing fiscal space for other social expenditures (Cichon et al., 2004). Social protection benefits linked to employment-based contributions also encourage the formalization of the informal economy. Enforcing existing regulations and adapting administrative mechanisms to facilitate registrations are important elements of those strategies.Eliminating illicit financial flowsEstimated at more than 10 times the size of all official development assistance received, a colossal amount of resources illegally escapes developing countries each year. Policy-makers should devote greater attention to cracking down on money laundering, bribery, tax evasion, trade mis-pricing and other financial crimes that are both illegal and deprive governments of revenue needed for social protection and achieving the SDGs.Adopting a more accommodating macroeconomic frameworkThis entails permitting higher budget deficit paths to stimulate public investment and, thus, employment and/or higher levels of inflation without jeopardizing macroeconomic stability.Each ASEAN Member State is unique, and all options should be (i) carefully examined, including the potential risks and trade-offs; and (ii) considered in national dialogue, including with government, employers and workers as well as with civil society, academics, international actors and others. National social dialogue is best to articulate optimal solutions in macroeconomic and fiscal policy, the need for social protection and investments for achieving the SDGs.76Pension cost estimationTo give additional concreteness to the discussion, some rough estimations of long-term costs for pension systems in ASEAN Member States were developed (table 15). The calculations were of a macroeconomic approach and should not be mixed up with actuarial studies. A core simplifying assumption was to calculate, for each country in local currency, current (2017) nominal per capita productivity as the fundamental economic parameter reflecting the economic status of an economy. The calculation of per capita productivity was done on the basis of the current demographic structure of each country and by dividing current (2017) nominal GDP by the number of persons in the current (2017) age group of 20–64.This approach can be debated because it may lead to underestimating or overestimating effective per capita productivity. We then assumed that the same (2017) per75 For example, in India, the tax or GDP ratio is not only low but has fallen in recent years (Srivastava, 2013). The same applies for Sri Lanka.76 In recent years and with the facilitation of the United Nations, the dialogue was organized in some countries in the form of a social protection assessment-based national dialogue.100capita productivity would apply to the future, to 2050. This approach was used to roughly calculate the magnitude of pension expenses that must be expected under current economic conditions (reflected by productivity) and under future (2050) demographic structures as well as full pension coverage and a 25 percent-of-productivity replacement rate of pensions.The calculation results must be taken with some grains of salt but nevertheless reflect, to some extent, the magnitude of pension policy reforms required in the future, including the potential magnitude of the fiscal space needed for such policy. The volume of fiscal space required clearly depends on the significance of state involvement in future pension policies and pension finance.It should also be emphasized that the results, by approach and as reflected in table 15, reflect all retirement incomes – not only those paid out of public coffers but also from other sources, be they occupational pensions (paid by enterprises) or private savings, (virtual) rents paid by owneroccupied property, the rest of the world or others.With this approach, we reflected on aspects of a broad discussion taking place in the international welfare state literature, including on “productivist welfare states”. In stipulating a 25 percent pension replacement of productivity, we implicitly calculated the costs of a pension package that guarantees freedom from want in each country (table 15). Due to lack of data, however, we were not in a position to double-check whether the applied 25 percent norm would effectively abolish poverty in each of the listed countries.Table 15. Estimates for a comprehensive pension package under full coverage of older persons in 2050 (as % of GDP), and current fiscal space allocated on old-age pensions in ASEAN Member States (latest available year)CountryBrunei Darussalam Cambodia Indonesia Lao PDR Malaysia Myanmar Philippines Singapore Thailand Viet NamA – Estimated pension costs un- B – Current level of public spending der full coverage and demograph- on pensions (latest year available) ic structure of 205011n/a80.2 (2010)70.5 (2010)70.1 (2010)90.9 (2012)70.7 (FY2015)70.6 (2012)150.7 (2011)112.2 (2015)105.5 (2015)Note: Spending as share of GPD was estimated for each country on the assumption that in 2050,all persons aged 65 or older will be covered and the average pension will equal 25 percent of productivity, where productivity is calculated as nominal GDP divided by the population aged 20–64(see text).Data source: For population 2017: DESA: PP2017_POP_F07_1_POPULATION_BY_AGE_BOTH_SEXES-1, accessed 3 June 2019; for GDP: https://data.worldbank.org/indicator/NY.GDP.MKTP.CN, accessed 3 June 2019; authors’ calculations; ILO, 2017d.101The difference between columns A and B in table 15 cannot be used for calculating the additional fiscal space needed by executing expansionary pension policy because it would depend on the intensity with which governments would transfer additional public monies to pensions. If, theoretically, all (100 percent) pension expenditure listed in column A is provided out of public coffers, then the difference between column A and column B would tentatively indicate the additional tax and mandated contributions required under such policies.Irrespective of these considerations, the findings in table 15 strongly indicate that, if Asian societies aim to be old-age decent, they must prepare for a pension policy that spends around 10 percent of GDP, with a variance of around, plus or minus, 3 percentage points.77 If countries fail to achieve the targeted spending levels, either full coverage will not be reached or pensions will probably not be decent, or both. All uncovered remaining burden will be left on the shoulders of families, who, by nature of the ageing process, will no longer be able to fulfill this task physically. And old-age poverty will continue.Table 15 reflects potential pension spending of the countries when they will have reached,or have come closer to,demographic maturity, in 2050.78 Estimates are independent of the economic sources financing the pension paid-outs, which, in principle, may be enterprises, government, private households and the rest of the world.Assuming that governments, in resorting to their general revenues, would fund the costs with a universal pension targeting poverty alleviation (rather than poverty abolishment), table 16B presents estimates of costs of a pension scheme covering all persons aged 65 or older with the value equivalent to the poverty line for the first year of implementation. These estimations disregard all pensions that might stem from social insurance (defined benefit) or from individual accounts (defined contribution); any other pensions, such as personal pensions or pensions drawn from abroad, are not included.Table 16. Estimated cost of the first full year of implementation of universal tax-financed old-age pension in ASEAN Member States (% of GDP)CountryCambodia Indonesia Lao PDR Malaysia Myanmar Philippines Thailand Viet Nam Total (average)Initial costs of a universal pension for persons aged 65 or older (value of the pension equivalent to the poverty line)1.4 0.5 0.8 2.2 3.0 0.4 1.1 1.5 1.2Notes: Values were calculated using the ILO Social Protection Floor Calculator. Results are only indicative, especially because they exclude administrative and upfront investment costs,such as in government equipment and the like.77 Singapore is an outlier in this regard due to the higher percentage of aged population expected in 2050 (projectionbased on UN data). 78 H ere we define full demographic maturity as a situation in which countries’ populations stagnate over generationsas a result of stagnant mortality and net-reproduction rates equaling 1,000. Not all countries reach this status at the same time, nor is to be expected that any country ever reaches such situation of stagnation. The uniform choice of 2050 for that maturity status being reached is only of an indicative nature. Actually, for 2050, the population numbers from the latest UN demographic projections were used.102Source: ILO, 2019.Generally speaking, the results in table 16 confirm what was implicitly already said in section 3.2, that for all ASEAN Member States, the introduction of a flat, tax-funded demogrant is most probably a low-cost undertaking relative to GDP. However,considerations turn much more difficult from a finance minister’s point of view because GDP, as a theoretical statistical construct, is not taxable. From their point of view, any required expansion of fiscal space is still a major challenge and potential obstacle to finding the resources required for financing even low-cost demogrants, as demonstrated here.Pensions and the demands for fiscal space by other programmesClaiming an extension of pension coverage and payment of decent pension levels is well justified, and the topic is at the core of this report. At the same time, it must not be forgotten that there are other social needs to be addressed by social security. Among them are health care provisions, care for older people and groups of persons otherwise disadvantaged, employment injury protection, family benefits, unemployment and anti-poverty measures. For developed countries with mature economies and aged population structures, a “one third, one third, one third” rule appears to guide the allocation of available social security resources: One third goes to old-age income security, one third to health protection and one third to all other social programmes taken together. For a country that spends 18 percent of GDP on social protection, this would imply 6 percent, 6 percent, 6 percent on old age, health and the other programmes.If, for the moment, one accepts this rule while relating to the figures in table 16, then countries should be prepared to allocate in 2050 GDP resources of between 21 percent for the Lao People’s Democratic Republic, Myanmar and the Philippines and 40–45 percent for Singapore on overall social protection, while the others would broadly spend between 25 percent and 30 percent.Although these magnitudes are plausible, actual estimates intended to serve policy measures would need to be based on much more detailed statistics and model calculations – which are not within the scope of this report. Detailed information on national accounts as well as formal and behavioural rules governing the social security systems must be taken into account. The impact of the age structure, of the labour force, its education level, health status and life expectancy in good health conditions, including labour productivity, is paramount in this respect. In fact, working conditions and retirement ages can significantly change social spending levels relative to GDP.Ageing, by definition, implies fewer children and youth, either in absolute or in relative numbers,as a proportion of the total population or in relation to the active employed population. In other words, while more resources of the society will need to be allocated to older persons, the falling, absolute or relative numbers of children and youth will absorb less. This observation is, without doubt, correct: In ageing societies, societal resources can be shifted from children and youth, where they are less needed, to the older population for their increased needs.But that argument must be used with care because, as explained previously, part of the solution to the financing problems of ageing societies is maintaining and increasing older workers’ labour productivity. Labour productivity not only depends on exogenous technological progress but requires a well-educated or trained labour force that is able to adapt to exogenous technological changes and manages to forward technological progress endogenously. The well-being of the older generation thus depends on investment in young people. As has been shown (Stephens, 2017),the related policy decisions are of a long-term nature, with long-term implications. The results of103success or failure to invest in children and youth will be seen only with longtime lags. Societies, to be on the safe side and in respect of maintaining decent living standard for the older generations, are well advised to not neglect education and training for their children and youth but to focus on respective investments. To cut a long argument short, while children and youth-related resources could be freed to some extent in ageing societies and shifted to the aged generations, governments should be careful. To increase the productivity and well-being of the population, investments in young people’s education is imperative. To explore the resulting effects in more detail while extending the information base on the social security system, application of what has been called “social budgeting” (Scholz, Cichon and Hagemejer, 2000) is recommended. Basically, social budgeting consists of two core parts: One is detailed and methodologically consistent statistics of all revenue going into a country’s social security system (contributions, taxes (subsidies), capital income, others) and all social security expenses, such as old age, health, other programmes in terms of cash and in-kind and including administration costs, such that macroeconomic adequacy of the relative shares of all programmes can be evaluated at each point in time. Second, while incorporating detailed statistics on the stakeholders of social security (employment, contributors, unemployed, pensioners, health system clients as well as providers, demographic information, etc.), the social budgeting statistics is a basis for short- and long-term scenario calculations, aiming at exploring the future cost developments of all social security programmes. This is necessary for making early informed decisions regarding social and financial systems and programme adequacy and sustainability. Hence, all ASEAN Member States are advised to implement appropriate statistical systems, and ideally, consensus is to be found on the methodology to be applied. The system of national accounts provides a useful basis. The OECD statistical methodology or the one applied by EUROSTAT (ESSPROS, 2012) also could be used.1045. CONCLUSIONS AND RECOMMENDATIONS105The analysis on the prospects for future social old-age income policies in ASEAN Member States takes a broad macroeconomic and demographic perspective in this report. It shows that social policy, generally, is not alien to South-Eastern Asian societies – neither in its needs nor in relation to Asia’s prevailing modern development model that,three decades ago,was labelled “productivist” and which may need modification. While general preconditions are positive for the accelerationoractivationofrespectivepolicies,thefastageingofsocietiesofmost ASEANMemberStates poses special challenges – ageing being understood as demographic processes of simultaneously falling mortality and fertility rates, around which some ASEAN countries are more advanced while others lag.CONCLUSIONS The clock is ticking for most ASEAN Member States. The fast pace of ageing is going to pose serious challenges within societies and to their economies under no-policy-change conditions, with growing numbers of older people uncovered by pension systems or receiving only extremely low pensions. The traditional model, in which children are the central source of income for their aged parents, will not be sustainable in societies in which the old-age dependency ratios arequickly changing and people are generally living so long that increasing numbers of children will not only have to provide support to their parents but also to their grandparents. By contrast to these findings, it appears as if ASEAN Member States still need to prepare themselves to undertake the required pension policy reforms. For possibly long transition periods, they will most likely continue to rely on co-support of intra-family transfers. However, with further urbanization and low fertility rates, old-age pension reforms will be increasingly critical because co-residency and similar arrangements will no longer be a solution as more and more single households emerge. Given the decades-long prevailing of the productivist development model, societies in ASEAN have, in their overwhelming majority, shied away from the costs of establishing coherent old-age income security policies that would allow for paying decent pensions to “all”. Indeed, ensuring income security in old age for ageing societies is not at all cheap. Considering the large share of informal employment in total employment (at 68.2 per cent in Asia and the Pacific: ILO, 2018e) and the need to maximize coverage urgently, the main technical solutions to policy motions will need to rely on tax-funded benefits, such as already applied in Brunei Darussalam and Thailand, with benefit levels improved. Voluntary schemes, including those providing matching contributions, have not worked in most of the countries, despite their popularity as a measure to increase participation of workers in the informal economy. The contributions remain high for participants, especially those from poor or near-poor backgrounds who cannot achieve decent pensions. The frequency of scheme members resorting to lump-sum pay-out appears mainly a consequence of how they were not able to accumulate enough years of contributions due to irregular working careers, particularly when few years of work in the formal sector are complemented with many years in the informal sector. For countries with defined-benefit schemes, it is generally imperative to be conscious of the biggest enemy of social insurance schemes, which is delaying required reforms. Delaying the necessary decisions too long can have serious implications in terms of the long-term106sustainability of the schemes, but it also creates unfair situations between generations. In the ASEAN Member States, the dominant problem is not so much generosity of benefits but the general underfunding of schemes, represented by inadequately set revenue parameters (too-low contribution rates and, where applicable, too-low and non-indexed contribution collection ceilings). The appropriate definition of insurable earnings is one of the often disregarded but most important elements regarding schemes’ financial sustainability and benefit adequacy. Comprehensiveness of pension systems includes simultaneous implementation of a “care industry”. Despite the explicit and implicit costs involved, societies will have to bear these anyway. Asian societies will only continue to be decent societies if they also look after their aged compatriots who, although not sick, cannot look after themselves anymore and thus need personal help. The general public’s trust in their country’s social security systems is essential to ensure the short- and long-term participation of the schemes’ members. It is important to understand that not only the members must trust the scheme but the society at large must trust it. Creating and maintaining a“social security culture”is no easy task, and it requires permanent, institutionalized effort. More fundamentally, its creation requires changes in Asia’s stillprevailing productivist development model, which, without doubt, has been economically successful over decades, albeit at various degrees in the different countries of this report. Expansionary social policy will require additional fiscal space (more tax and/or contribution revenue). To some extent, that additional fiscal space will come quasi-automatically: popular social programmes have thus far found the general public’s support everywhere in the world. Extension of the fiscal space is, of course, much easier if internationally coordinated. If such policy orientation is agreed upon at the international level, ASEAN Member States have no reason to believe that they will be aged before rich. When accompanied by adequate international macroeconomic policy coordination, the countries of ASEAN can continue pursuing growth-oriented policies, although no longer mainly footed on international demand. Instead, they will be shaped predominantly through domestic investment and income requirements, including significantly expanded social security systems.RECOMMENDATIONS Governments should move towards setting common ASEAN targets in terms of pension coverage and relative benefits levels, which should be guided by the relevant ILO Conventions. This would enhance the design and implementation of inclusive old-age income security systems at the national level without impacting global and regional economic competitiveness. Pension policies and systems should be approached in a more comprehensive way. The combination of resources from taxation and contributions are essential to provide national pension systems with adequate instruments to reach objectives of universal coverage and adequacy. Contributory and non-contributory schemes should not be seen as detached mechanisms. Non-contributory pension schemes should be implemented by all ASEAN Member States, at first categorically targeted to certain age groups in the context of lower fiscal capacities but aiming to reach universal coverage in the (foreseeable) long term. Only universal, non-107contributory pension schemes can effectively reach the vast numbers of people in informal employment and also women, who are more like to have interrupted careers and in work the informal economy. It is important to further develop and continue efforts to extend existing and new contributory schemes to persons working in non-standard forms of work. This should cover the classic workers in the informal sector and informal workers in formal enterprises and also the new non-standard forms of work, such as workers in the platform economy. Efforts are needed to strengthen employers’ social sense of responsibility, encourage social dialogue, ensure compliance of enterprises with legislation and simplify administrative and registration rules for people who are self-employed and for SMEs. Countries should prioritize the improvement of the adequacy of pension benefits. The definition of adequate pension would profit from the guidance of the minimum international standardsstipulated in ILO Conventions No. 102 and No. 128 (for contributory pensions) and Recommendation No. 202 (for non-contributory pensions). Pension benefits must be indexed regularly to preserve the purchasing power of the benefits and keep the pensions meaningful. ASEAN countries need to assess what are the most adequate mechanisms and the best benchmarks to use as a basis for indexing their pensions in payment. To ensure long-term sustainability of defined-benefit schemes, countries should implement reforms, which can include scheduled increases in the retirement age andscheduled reductions in the income-replacement rates when benefit provisions are too high, among other measures. Reforms that are undertaken in time can allow people to gradually adjust and adapt to them over time. Contribution rates must be gradually and smoothly increased to an “old normal”. Pensions are expensive. If contribution rates of around 25 per cent are considered too high for the ASEAN Member States’ poor, then there is no choice of either subsidizing the contributions of the poor by way of tax-financed government transfers or, as suggested here, by way of introducing a flat demogrant-type pension for all. For countries that have opted for relying mainly on defined-contribution solutions, based on individual savings accounts under the management of provident funds, radical changes are not recommended. Provident funds should add elements of redistribution and risk-sharing into their schemes, including, for instance, the setting of mandatory conversion of individual savings into annuities and the provision of minimum guaranteed returns on investment. Early withdrawal from savings accounts should be prohibited. The function of retirement income should be separated from other functions, such as housing, education and emergency medical care. This applies to provident funds but also to some social insurance schemes, in which it is still possible to receive the benefit in the form of a lumpsum under certain conditions. Developing an overarching culture of transparency, including maximum possible alignment of different schemes (where they exist in parallel) is essential to gain the trust of the schemes’ members.108 Following good international practices in terms of disclosing financial information could be a way of contributing to members’ confidence in their systems. The role of social partners in the governance and decision-making processes of social security is another element with potential to contribute to the creation of trust in the systems. Each society’s social policy literacy must be increased. Investing in communication and creation of a social security culture is important and remains a permanent task under any scheme anywhere in the world, especially during a social security build-up phase, as is currently underway in Asia. Social security expansion, increasing coverage and adequacy of pensions could be part of a macroeconomic policy aiming at ameliorating the conflictual international trade dynamics: Increasing national pension benefits and other social security would increase national demand and imports and contribute to closing the trade surpluses, thus reducing capital exports. At its core, this switch would imply to dispose of the productivist model that is based on low labour costs and international demand. It would be replaced by a model that rests upon the productivity effects of high(er) labour costs while international demand would still be a trigger for production but would be compensated by imports resulting from national demand.109REFERENCESAkerlof, G.A. 1970. “The market for lemons: Quality uncertainty and the market mechanism”,in Quarterly Journal of Economics. Vol. 3, No. 84. Available at: https://academic.oup.com/qje [3 June2019].Allianz. 2013. International pensions. What’s happening in Malaysia: Getting prepared for retirement? Allianz International Pension Papers (Munich, Alliance. ). Available at: https://projectm-online.com/app/uploads/what-is-happening-in-malaysia.pdf [4 Apr. 2018].Asia-Pacific Migration Network. 2018. “ASEAN Labour Migration Statistics: The International Labour Migration Statistics Database (ILMS) in ASEAN”.Available at: http://apmigration.ilo.org/ASEANlabour-migration-statistics [16 Oct. 2018].Association of Southeast Asian Nations (ASEAN). 1967. The ASEAN Declaration (Bangkok). Available at: http://ASEAN.org/the-ASEAN-declaration-bangkok-declaration-bangkok-8-august-1967 [13Aug. 2019].—.2009. ASEAN Human Rights Declaration (Phnom Penh). Available at: https://ASEAN.org/ASEANhuman-rights-declaration/[13 Aug. 2019].—.2013. ASEAN Declaration on Strengthening Social Protection (Brunei Darussalam). Available at: https://ASEAN.org/?static_post=ASEAN-declaration-strengthening-social-protection-regionalframework-action-plan-implement-ASEAN-declaration-strengthening-social-protection[13 Aug. 2019].—.2017. Statistical yearbook 2016/2017 (Jakarta, ASEAN Secretariat). Available at: http://www. ASEANstats.org/wp-content/uploads/2018/01/ASYB_2017-rev.pdf [13 Aug. 2019].—.2018. “ASEAN key figures”. Available at: https://www.ASEANstats.org/wp-content/ uploads/2018/12/ASEAN-Key-Figures-2018.pdf [13 Aug. 2019].Asher. 2013. Singapore’s Central Provident Fund: A sustainable and fair solution to the country’s pension challenge? (Singapore, International Monetary Fund). Available at: www.imf.org/external/ np/seminars/eng/2013/oapfad/pdf/asher.pdf [13 Aug. 2019].Atkinson, E. 2013. Out of the Household: Master-Servant Relations and Employer Liability Law. Journal of Law & the Humanities, Vol. 25, No. 2. Harvard Law School, Yale.Barr, N.; Diamond, P. 2010. Pension reform in China: Issues, options and recommendations (London and Boston, MA, London School of Economics and Massachusetts Institute of Technology). Available at: https://economics.mit.edu/files/6310[13 Aug. 2019].Beveridge, W.H. 1942. Social insurance and allied services: Report (London, Her Majesty’s Stationery Office).Bleicken, J. 2015. Augustus: The biography(New York, Penguin).Brody, E.M. 1990. Women in the middle: Their parent care years(New York, Springer).Channel NewsAsia. 2017. “Abuse of the elderly: Why some are reluctant to speak up”. Available at: http://elder-abuse-spotlight.blogspot.com/2015/ [24 May 2018].110Chia, N.C.; Tsui, A.K.C. 2012. Adequacy of Singapore’s Central Provident Fund payouts: Income replacement rates of entrant workers (Singapore, National University, Department of Economics).Cichon, M.; Scholz, W.; van de Meerendonk, A.; Hagemejer, K.; Bertranou, F.; Plamondon, P. 2004. Financing social protection,Quantitative Methods in Social Protection Series (Geneva, International Labour Office and the International Social Security Association).Clay, T.; Idris, N.; Cunha, N.; Knox-Vydmanov, C. Forthcoming. A social protection floor for older people in Myanmar(Yangon, ILO).Coalition of Services of the Elderly (COSE) and HelpAge International (HAI). 2017. The feasibility of a universal social pension in the Philippines(Manila).Central Provident Fund. 2016. “Central Provident Fund contribution and allocation rates” (Singapore, Central Provident Fund Board). Available at: https://www.cpf.gov.sg/Employers/EmployerGuides/ employer-guides/paying-cpf-contributions/cpf-contribution-and-allocation-rates [4 Mar. 2018].Department of Social Welfare and Development (DSWD). 2010. Guidelines on the implementation of the social pension for indigent senior citizens, Administrative order No. 15 (Quezon City, Philippines, Government of the Philippines). Available at: www.dswd.gov.ph/issuances/AOs/AO_2010-015.pdf [16 Oct. 2018].De Soto, H. 2007. The mystery of capital: Why capitalism triumphs in the West and fails everywhere else (New York, Basic Books).Deutsche Bundesbank. 2017. “Die rolle von banken, nichtbanken und zentralbank im geldschöpfungsprozess”, inMonatsbericht,April. [“The role of banks, non-banks and the central bank in the process of money creation”.] Available at: www.bundesbank.de/Redaktion/DE/Downloads/ Veroeffentlichungen/Monatsberichtsaufsaetze/2017/2017_04_geldschoepfungsprozess.pdf?__ blob=publicationFile[7 Jan. 2019].Di Addio, A.C.; Whitehouse, E. 2012. Towards financial sustainability of pension systems: The role of automatic adjustment mechanisms in OECD and EU countries (Berne, Switzerland, Swiss Federal Office for Social Insurance).Dixon, John E. 1999. Social security in global perspective. Westport, CT, Praeger Publishers.Dorfman, M.C.; Wang, D.; O’Keefe, P.; Cheng, J. 2013. “China’s pension schemes for rural and urban residents”, inR. Hinz et al.:Matching contributions for pensions: A review of international experience (Washington, DC, International Bank for Reconstruction and Development, World Bank),pp. 217– 241.Duran-Valverde, F.; Pacheco, J.F. 2012. Fiscal space and the extension of social protection: Lessons from developing countries,Extension of Social Security Paper Series No. 33 (Geneva, ILO).Ebenstein, A. 2001. Friedrich Hayek: A biography (Chicago, Chicago University Press).Employees Provident Fund (EPF). 2015. Annual report (Kuala Lumpur, Government of Malaysia). Available at: www.kwsp.gov.my/portal/en/about-epf/investment-highlights/annual-report/annualreport-2014 [15 Dec. 2015].—. 2015b. “Social Security in Malaysia: Issues and challenges, presentation made at the International Social Security Conference, Kuala Lumpur.111“English Poor Laws”. Available at: https://en.wikipedia.org/wiki/English_Poor_Laws#New_Poor_Law [22 Nov. 2016].European Commission. 2009. The 2009 ageing report: Economic and budgetary projections for the EU-27 Member States (2008-2060). Joint report prepared by the European Commission and the Economic Policy Committee (Brussels, Directorate-General for Economic and Financial Affairs).European System of Accounts (ESA). 2013. “Regulation (EU) No 549/2013 of the European Parliament and of the Council on the European system of national and regional accounts in the European Union”, 21 May. Available at: http://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=OJ:L:2013:174:0001:0727:EN:PDF[13 Aug. 2019].European System of Integrated Social Protection Statistics (ESSPROS). 2012. ESSPROS manual (Luxembourg, Office for Official Publications of the European Communities).EurWork. 2001. “ILO submits study on pensions scheme”, in European Observatory of Working Life, Mar.Available at: www.eurofound.europa.eu/sr/observatories/eurwork/articles/ilo-submits-studyon-pensions-scheme [13 Aug. 2019].Flassbeck, H.; Spiecker, F.; Meinhardt, V.; Vesper, D. 2012. Irrweg grundeinkommen. Die große Umverteilung von unten nach oben muss beendet warden (Frankfurt, Westend).[German]. [GMI – The wrong path: The great re-dristribution from the top to the bottom must come to an end.]Giles, J.; Wang, D.; Park, A. 2013. Expanding social insurance coverage in urban China, Policy Research Working Paper 6497 (Washington, DC, World Bank, Development Research Group, Human Development and Public Services Team).Gillion, C. 2000. The development and reform of social security pensions: The approach of the International Labour Office (Geneva, ILO, Social Security Department). Available at: www.ilo.org/ wcmsp5/groups/public/---ed_protect/---soc_sec/documents/publication/wcms_secsoc_7776. pdf[13 Aug. 2019].Gillion, C.; Turner, J.; Bailey, C.; Latulippe, D. (eds). 2000. Social security pensions: Development and reform(Geneva, ILO).Gleeson–White, J. 2011. Double entry: How the merchants of Venice shaped the modern world – and how their invention could make or break the planet (London, Allen Lane).Graunt, J. 1662. Natural and political observations mentioned in a following index, and made upon the bills of mortality (London). Available at: https://en.wikipedia.org/wiki/John_Graunt [4 Mar. 2017].Halley, E. 1693. “An estimate of the degrees of the mortality of mankind, drawn from curious tables of the births and funerals at the city of Breslaw: With an attempt to ascertain the price of annuities upon lives, Philosophical Transactions. Available at: http://rstl.royalsocietypublishing.org/ content/17/192-206/596.full.pdf+html [4 Mar. 2017].HelpAge India. 2015. The state of elderly in India 2014 (New Delhi). Available at: https://www. helpageindia.org/wp-content/themes/helpageindia/pdf/state-elderly-india-2014.pdf [x Mon. year].HelpAge International (HAI). 2013. The changing well-being of Thai elderly: An update from the 2011 survey of older persons in Thailand (Chiang Mai). Available at: www.helpage.org/silo/files/ the-changing-wellbeing-of-thai-elderly-an-update-from-the-2011-survey-of-older-persons-inthailand.pdf [15 July 2018].112—. 2016. Work, family and social protection: Old age income security in Bangladesh, Nepal, the Philippines, Thailand and Viet Nam (Chiang Mai, Thailand).Hinz, R; Holzmann, R.; Tuesta, D.; Takayama, N. (eds). 2013. Matching contributions for pensions: A review of international experience (Washington, DC,International Bank for Reconstruction and Development, World Bank), p. 6.Holliday, I. 2000. “Productivist welfare capitalism: Social policy in East Asia”, in Sage Journals, Political Studies, Sep. Available at: http://journals.sagepub.com/doi/pdf/10.1111/1467-9248.00279 [13 Aug. 2019].Holzmann, R.;Hinz, R.; von Gersdorff, H.; Gill, I.; Impavido, G.; Musalem, A.R.; Palacios, R.; Robolino, D.; Rutkowski, M.; Schwarz, A.; Sin, Y.; Subbarao, K. 2005. “Old-age income support in the 21st century: An international perspective on pension systems and reform (Washington, DC, World Bank).Holzmann, R.; Hinz, R.; Tuesta, D. 2013. “Early lessons from country experience with matching contribution schemes”,in R. Hinz et al.: Matching contributions for pensions: A review of international experience (Washington, DC,International Bank for Reconstruction and Development, World Bank), pp. 3–26.Holzmann, R. 2014. Old-age financial protection in Malaysia: Challenges and options, Social Protection Discussion Paper No. 1425 (Washington, DC, World Bank Group).Hudson, J. 2016. From welfare state to competition state? The evolving relationship between social security and the economy in the United Kingdom (York, UK, University of York).Huzaime, A.H. 2012. The World Bank’s 5 pillars pension and its refinements for application in Malaysia,paper presented at Dynamic Evolution of the Pensions World, Kuala Lumpur,April.International Conference of Labour Statisticians (ICLS). 1993. The 15th International Conference of Labour Statisticians(Geneva).—. 2003. The 17th International Conference of Labour Statisticians (Geneva).International Labour Office. 1952. “Convention 102, Social Security (Minimum Standards) Convention, 1952 (No. 102)” (Geneva). Available at: www.ilo.org/dyn/normlex/en/f?p =NORMLEXPUB:12100:0::NO::P12100_INSTRUMENT_ID:312247 [10 Dec. 2016].—. 2004. Aruba: Report to the Government: Actuarial review of the General Old-age Pension Scheme and the General Widows and Orphans Insurance Scheme as of 30 April 2003(Geneva). Available at: www.social-protection.org/gimi/gess/RessourcePDF. action;jsessionid=35TwZ2dLHL1LYhGdDwv1TVFH0zSqMYvSMv8P1w5HhG0vL97dGCgz! 726592978?ressource.ressourceId=8290 [13 Aug. 2019].—. 2005. Aruba: Report to the Government: Alternative policy options for the long-term financial consolidation of the Social Insurance Bank (Geneva). Available at: www.social-protection.org/gimi/gess/RessourcePDF.do?ressource.ressourceId=8302[13 Aug. 2019].—. 2008. Social security in Sri Lanka: A review of the social security system for the Ministry of Labour Relations and Manpower(Geneva, Social Security Department). Available at: http://www.social-protection.org/gimi/gess/RessourcePDF.action?ressource.ressourceId=7055113[13 Aug. 2019].—. 2012. “Recommendation 202: Social Protection Floors Recommendation, 2012 (No. 202) Recommendation concerning National Floors of Social Protection”, NORMLEX Information System on International Labour Standards. Available at:http://www.ilo.org/dyn/normlex/en/f?p=NORMLE XPUB:12100:0::NO::P12100_ILO_CODE:R202 [29 Nov. 2017].—. 2013. Measuring informality: A statistical manual on the informal sector and informal employment(Geneva). Available at: www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/--publ/documents/publication/wcms_222979.pdf[13 Aug. 2019].—. 2014a. Actuarial study of the proposed modifications to the social security system for the employment injury, old-age and death benefits (Jakarta). Available at: www.ilo.org/wcmsp5/ groups/public/---asia/---ro-bangkok/---ilo-jakarta/documents/publication/wcms_571382.pdf[13 Aug. 2019].—. 2014b. “R204 – Transition from the Informal to the Formal Economy Recommendation, 2015 (No. 204)”. Available at: www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_ CODE:R204 [13 Aug. 2019].—. 2014cSocial protection for older persons: Key policy trends and statistics, Social protection policy paper No. 11(Geneva, Social Protection Department). Available at: www.ilo.org/wcmsp5/groups/ public/---dgreports/---dcomm/documents/publication/wcms_310211.pdf [13 Aug. 2019].—. 2014d. World social protection report 2014/15: Building economic recovery, inclusive development and social justice (Geneva). Available at: www.ilo.org/wcmsp5/groups/public/--dgreports/---dcomm/documents/publication/wcms_245201.pdf [13 Aug. 2019].—. 2016a. Building an inclusive future with decent work: Towards sustainable development in Asia and the Pacific, Report of the General-Director, 16th Asia-Pacific Regional Meeting, Bali, 6–9 Dec. 2016. Available at: www.ilo.org/wcmsp5/groups/public/---ed_norm/---relconf/documents/ meetingdocument/wcms_531409.pdf [8 Nov. 2016].—. 2016b. Indonesia: Financing social protection through contributions and the removal of fuel subsidy,Building Social Protection Floors Country Note Series (Geneva).—. 2016c. Report to the Government: Actuarial valuation of Thailand social security administered by the Social Security Office as of 31 December 2013 (Geneva, Social Protection Department).—. 2016d. Social protection for the elderly in ASEAN+3: A vision of key policy trends, challenges and opportunities, Concept note (Bangkok).—. 2016e. Systemic review of Thailand’s pension system: Exploring options for reform and improvement, Report to the Government (Bangkok,ILO ROAP). [internal]—. 2016f. Thailand: Improving the lives of older persons through universal social pensions,Building Social Protection Floors Country Note Series (Geneva).—. 2016g. Zambia: Financing social protection through taxation of natural resources, Social Protection in Action Country Brief (Geneva).—. 2017a. India labour market update(New Delhi). Available at: www.ilo.org/newdelhi/whatwedo/ publications/WCMS_568701/lang--en/index.htm [13 Aug. 2019].114—. 2017b. Study on actuarial valuation of a private sector social security scheme in Cambodia (Phnom Penh).—. 2017c. Summary report assessment-based national dialogue on social protection: Lao People’s Democratic Republic(Vientiane).—. 2017d. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. An adequate and sustainable pension scheme for the private sector, Technical note (Cambodia).—. 2018b. Global social security and economic development: Retrospect and prospect, ILO working paper in cooperation with Renmin University (Beijing) and Friedrich-Ebert-Stiftung Foundation (Beijing).—. 2018c. ILO multi-pillar pension model: Building equitable and sustainable pension systems(Geneva). Available at: www.ilo.org/wcmsp5/groups/public/---ed_protect/---soc_sec/ documents/publication/wcms_645751.pdf[13 Aug. 2019].—. 2018d.ILOSTAT (Geneva). Available at: www.ilo.org/ilostat[12August 2018].—. 2018e.Women and men in the informal economy: A statistical picture (Geneva). Available at: www. ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/publication/wcms_626831. pdf[13 Aug. 2019].—. forthcoming a. Assessment-based national dialogue in the Philippines: Establishing a social protection floor for all(Manila).—. forthcoming b. Expansion of social security to workers in informal employment in ASEAN(Bangkok).—. forthcoming c. Report to the Ministry of Labour, Invalids and Social Affairs of Viet Nam: Exploring options for a multi-tier pension system in Viet Nam(Hanoi).—. 2019. “ILO social protection floor cost calculator”. Available at: www.social-protection.org/gimi/ gess/ShowWiki.action?wiki.wikiId=3051[13 Aug. 2019].—.Agence Francaise de Developpement(AFD). 2018. National Social Security Fund: An adequate and sustainable pension scheme for the private sector(Phnom Penh, ILO).International Monetary Fund (IMF). 2017. Regional economic outlook: Asia and Pacific. Preparing for Choppy Seas. World Economic and Financial Surveys (Washington, DC).Jackson, R.; Peter, T. 2015. From challenge to opportunity: Wave 2 of the East-Asia retirement survey (Alexandria, VA,Global Aging Institute). Available at: www.globalaginginstitute.org/assets/clientassets/gapi/downloads/publications/EARR_Wave2_Report_DL_LR-EIINC.pdf[13 Aug. 2019].Jones, C. 1993. “The Pacific challenge”, in: C. Jones: New perspectives on the welfare state in Europe (London, Routledge).Jones, R.S.; Urasawa, S.2014. Reducing the high rate of poverty among the elderly in Korea, OECD Economics Department Working Papers (Paris, OECD Publishing).Karunanethy, K. 2017. A Review of the provident fund approach to providing old-age income115security in four Asia-Pacific countries (Bangkok, United Nations Economic and Social Commission for Asia and the Pacific).Kim, Y-M. 2017. Limit of economic growth without welfare: The experience of South Korea, Paper commissioned by Renmin University, Beijing, China, in collaboration with ILO Beijing(Seoul, ChungAng University).Knodel, J. 2012. The situation of older persons in Myanmar: Results from the 2012 survey of older persons(Chiang Mai,HelpAge International).—. Teerawichitchainan, B.; Prachuabmoh, V.; Pothisiri, W. 2015. The Situation of Thailand’s Older Population: An update based on the 2014 Survey of Older Persons in Thailand. Population Research Center, Research Reports, Report 15-847. Ann Arbor: University of Michigan, Institute for Social Research. Available at: www.psc.isr.umich.edu/pubs/pdf/rr15-847.pdf [15 July 2018].Konfuzius. 2017. Gespräche (Lun-yu). Aus dem Chinesischen von Ralf Moritz. Reclams Universal Bibliothek Nr. 11110 (Stuttgart, 1982 [2017]). [The analects,translated from Chinese by Ralf Moritz.]Leisering, L. 2003. “Die Entstehung „Neuer Wohlfahrtsstaaten“ in Ostasien – Stichworte aus der neueren soziologischen Debatte mit perspektivischen Anwendungen auf Behindertenpolitik in Asien”,in B. von Maydell, Rainer Pitschas and Bernd Schulte (eds):Behinderung in Asien und Europa im Politik- und Rechtsvergleich(Baden-Baden, Germany, Nomos). [in German]. [“The emergence of ‘new welfare states’in East Asia – Terms of the late sociological debate with perspective of application to disability policy in Asia.”]Lim, L. 1989. “Social welfare”, in K. Sandhu, K. Singh and P. Wheatly: Management of success: The moulding of modern Singapore (Singapore, Institute of Southeast Asian Studies)..Lindert, P.H. 2010. Growing public: Social spending and economic growth since the 18th century (Los Angeles, CA, University of California Los Angeles).Maddison, A. n.d. The Maddison project(Groningen, Netherlands, Groningen Growth and Development Centre, University of Groningen). Available at: www.ggdc.net/maddison/maddisonproject/home.htm[13Aug. 2019]. [Estimates on long-term historical time series of population and economic growth]Manthe, U. 2011. Geschichte des Römischen Rechts (München, Germany, Beck). [History of the Roman law.]Mercer. 2016. “Melbourne Mercer Global Pension Index” (Melbourne, Australia, Australian Centre for Financial Studies).Mesa-Lago, V.; Castillo, M. 2011. Pensions in the Philippines: Challenges and ways forward(Manila, Friedrich-Ebert-Stiftung Foundation).Midgley, J.; Sherraden, M.W. (eds). 1997. Alternatives to social security: An international inquiry(Westport, CT, Auburn House, Greenwood Publishing Group).Miller, D. 1981. “The ‘sandwich’ generation: Adult children of the aging”, inSocial Work, Vol. 26.Ministry of Manpower (Singapore). 2016. “Factsheet on the Silver Support Scheme”. Available at: www.mom.gov.sg/~/media/mom/documents/budget2016/factsheet-on-silver-support-scheme. pdf?la=en [30 Sep. 2018].116—. 2018. Annual report. Singapore. Available at: www.cpf.gov.sg/Assets/common/Documents/ ViewAnnualReport2018inPDF_Part1.pdf [18 Nov. 2019].—. 2019. “Report: Labour Force in Singapore 2018”. Available at: https://stats.mom.gov.sg/Pages/ Labour-Force-In-Singapore-2018.aspx[18Nov. 2019].Ministry of Strategy and Finance (Republic of Korea). 2015. Long-term economic outlook 2060 (Seoul, Government of Republic of Korea).Mink, R. 2008. “General government pension obligations in Europe”, in IFC Bulletin, No. 28 (Bank for International Settlement).Mu, R. 2015. Pension coverage for parents and educational investment in children: Evidence from urban China, Policy Research Working Paper 7457 (Washington, DC, World Bank, Development Economics Vice Presidency, Operations and Strategy Team)Ng, K-H. 2015. “Can Singapore’s Central Provident Fund still meet retirement income needs?”,in Midgley, J.; Sherraden, M.W. (eds). 1997. Alternatives to social security: An international inquiry (Westport, CT, Auburn House, Greenwood Publishing Group), pp. 104–122.NTA Project. 2018.“National transfer accounts: Understanding the generational economy”. Available at: www.ntaccounts.org/web/nta/show/ [11 Mar. 2018].Nyayieka, F. 2017. “Pension reforms in Rwanda”,The New Times, 8 Sep. (Kigali). Available at: www. newtimes.co.rw/section/read/219511/ [8 Mar. 2018].Ong, C.B.; Peyron-Bista, C. 2015. The state of social protection in ASEAN at the dawn of integration (Bangkok, ILO Regional Office for Asia and the Pacific). Available at: http://socialprotection.org/ discover/publications/state-social-protection-ASEAN-dawn-integration[13 Aug. 2019].Ono, T. 2016. Japan – Historical experience of the interaction between economic development and social security; current core problems in reaction to national and global developments; in search of solutions for the future, Paper commissioned by Renmin University, Beijing, China, in collaboration with ILO-Beijing (Beijing, Renmin University, Friedrich-Ebert-Stiftung, ILO).Organisation for Economic Co-operation and Development (OECD). 2012a. Economic policy reforms (Paris). Available at: http://dx.doi.org/10.1787/growth-2012-en [19Nov. 2019].—. 2012b. Pensions at a glance Asia/Pacific 2011 (Paris).—. 2013. Pensions at a glance Asia/Pacific 2013(Paris).—. 2016. “OECD factbook 2015-2016: Economic, environmental and social statistics”. Available at: www.oecd-ilibrary.org/economics/oecd-factbook-2015-2016_factbook-2015-en [19Nov. 2019].—. 2019.Under Pressure: The Squeezed Middle Class (Paris, OECD Publishing). Available at:https:// doi.org/10.1787/689afed1-en[19Jan. 2020].Ortiz, I.; Cummins, M.; Karunanethy, K. 2017. Fiscal space for social protection and the SDGs: Options to expand social investments in 187 countries, ESS Working Paper No. 48 (Geneva and New York, ILO, UNICEF and UNWomen). Available at: http://www.social-protection.org/gimi/gess/RessourcePDF. action?ressource.ressourceId=51537[13 Aug. 2019].Pareto, W. n.d. “Pareto efficiency”. Wikipedia entry (English) [27 Feb. 2017].117Pensionfundsonline. n.d.“Singapore”. Available at: www.pensionfundsonline.co.uk/content/countryprofiles/singapore/101 [13 Aug. 2019].Petersen, K. 2017. Growth for welfare or welfare for growth? The changing relationship between economic development and social security in the Nordic countries(Geneva, ILO).Philippines Information Agency (PIA). 2018. “DSWS supports the bill doubling the social pension for the indigent citizens.” News Center, 18 Sep. Available at: https://pia.gov.ph/news/articles/1012868 [16 Oct. 2018].Philippine Statistics Authority (PSA). 2015. Labour force survey(Quezon City).Piggott, J. 2007. Pension reform and the development of pension systems: An evaluation of World Bank assistance, background paper (Washington, DC, World Bank).Piketty, T. 2014. Capital in the twenty-first century(Cambridge, MA, The Belknap Press of Harvard University Press) [Translated from original French by Arthur Goldhammer.]Plamondon, P.; Drouin, A.; Binet, G.; Cichon, M.; McGillivray, W.R.; Bédard, M.; Perez-Montaz, H. 2002. Actuarial practice in social security, Quantitative Methods in Social Protection Series (Geneva, ILO).Polanyi, K. 1945. Origins of our time: The great transformation (London, Gollancz).Porsche-Ludwig, M.; Bellers, J.; Gieler, W. (eds). 2013. Sozialpolitik in Asien: Ein Handbuch der Staaten Asiens von A-Z.(Berlin-Münster-Wien-Zürich-London, Lit Verlag). [German]. [Social policy in Asia: A handbook of the Asian nations from A to Z].Private Pension Administrator (PPA). 2017. “Malaysia”. Available at: www.ppa.my/wp-content/ uploads/2018/03/Private-Retirement-Schemes-Hits-Quarter-Million-Member-Milestone-pressrelease-31-July-2017.pdf[13 Aug. 2019].Department of Population (DOP) (Myanmar). 2016. The 2014 Myanmar Population and Housing Census: The union report,Census Report Vol. 2 (Nay Pyi Taw).Rieger, E.; Leibfried, S. 1999. “Wohlfahrtsstaat und Sozialpolitik in Ostasien. Der Einfluß von Religion im Kulturvergleich”, in G. Schmidt and Rainer Trinczek (eds):Globalisierung. Ökonomische und soziale Herausforderungen am Ende des zwanzigsten Jahrhunderts (Baden-Baden, Nomos, Soziale Welt, Sonderband 13). [in German][Welfare state and social policy in EastAsia: The impact of religion in cultural comparison].Ritter, G.A. 2014. “The creation of the German pension insurance and its major characteristics”, Revision of a lecture, “Thanks, Otto! 125 Years of Pensions and New Global Perspectives”, Berlin, 28– 29 October.Kapsos, S.; Bourmpoula, E. 2013.Employment and economic class in the developing world, ILO Research Paper No. 6 (Geneva, ILO). Available at: www.ilo.org/wcmsp5/groups/public/---dgreports/--inst/documents/publication/wcms_216451.pdf [8 Nov. 2016].Samuelson, P.A. 1958. “An exact consumption-loan model of interest with or without the social contrivance of money”,Journal of Political Economy, Vol. 66, No. 4.Sandhu, K.S.; Wheatly, P. 1989. Management of success: The moulding of modern Singapore (Singapore, Institute of Southeast Asian Studies).118Scholz, W. 2007. “Alterssicherung und Kapitaldeckung”, in Betriebliche Altersversorgung 7(Heidelberg), pp. 507–515. [Securing old-age income through capital funding].—. 2015. “Financing social security out of contributions: About origins, present discussions and prospects of a success story”,in International Social Security Review, Vol. 68, 4.—. 2017. “Global social security in political and economic contexts: Historical experiences and current trends”,in Chinese Social Security Review, Vol. 1, No. 1, pp. 135–152. [In Chinese, with abstract in English]—; Cichon, M.; Hagemejer, K. 2000. Social budgeting, Quantitative Methods in Social Protection Series(Geneva, ILO and International Social Security Association).Sherraden, M. 1997. “Provident funds and social protection: The case of Singapore”,in J. Midgley; M.W. Sherraden(eds):Alternatives to social security: An international inquiry (Westport, CT, Auburn House, Greenwood Publishing Group).—; L.Z.; Ku, H.; Deng, S.;Wang, S. (eds). 2015. Asset-building policies and innovations in Asia: Routledge research on social work, social policy and social development in greater China(London, Routledge/Routledge).Social Security Administration (SSA). 2015. Social security programs throughout the world: Asia and the Pacific, 2014: Research, statistics andpolicy analysis(Bangkok, Social Security Administration). Available at: www.ssa.gov/policy/docs/progdesc/ssptw/2014-2015/asia/index.html[13 Aug. 2019].Srivastava, R.S. 2013. A social protection floor for India (New Delhi, ILO Decent Work Team for South Asia and ILO Country Office for India).State Council (China). 2018. “5 major changes set for pension insurance system”, Chinanews, 18 Apr. Available at: http://english.gov.cn/policies/policy_watch/2018/04/02/content_281476099079092. htm[13 Aug. 2019].Stephens, J.D. 2017. The welfare state and economic development: The United States in comparative perspective, Background paper for the 2017-FES/ILO/Renmin University project on Global Social Security and Economic Development: Retrospect and Prospect (Chapel Hill, NC, University of North Carolina).Supavanich, P. 2016. “ASEAN cooperation on social protection”, Presentation during the ILO–China– ASEAN High Level Seminar to achieve the SDGs on Universal Social Protection through South–South and Triangular Cooperation, Beijing, 6–8 Sep. Available at: http://www.social-protection.org/gimi/ gess/RessourcePDF.action?ressource.ressourceId=53904[13 Aug. 2019].Trading Economics. 2018. “Singapore average monthly wages 1989–2018”. Available at: https:// tradingeconomics.com/singapore/wages) [16 May 2018].United Nations. 2002. Second World Assembly on Ageing, Madrid, Spain: Political Declaration and Madrid International Plan of Action on Ageing(New York, United Nations).United Nations Department on Economic and Social Affairs (DESA) Population Division. 2015. World population ageing 2015(New York). Available at: www.un.org/en/development/desa/population/ publications/pdf/ageing/WPA2015_Report.pdf[13 Aug. 2019].—. 2017a. World population ageing 2017(New York). ST/ESA/SER.A/397. Available at: www.un.org/119en/development/desa/population/publications/pdf/ageing/WPA2017_Highlights.pdf. [15 July 2018].—. 2017b. World population prospects: The 2017 revision(New York). Available at:https://esa.un.org/ unpd/wpp/DataQuery/ [15 July 2018].United Nations Development Programme (UNDP). 2018. Human development indices and indicators:2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/files/2018_ human_development_statistical_update.pdf [13 Aug. 2019]Unknown. 2016. The development of China’s pension reforms from 1951 to 2015 (anonymous note).Vaughan-Whitehead, D. 2015. The European social model in crisis: Is Europe losing its soul?(Geneva, ILO). Available at: www.ilo.org/global/publications/ilo-bookstore/order-online/books/ WCMS_314465/lang--de/index.htm[13 Aug. 2019].Venne, R. 2005. The Framework for Monitoring, Review and Appraisal of the Madrid International Plan of Action on Ageing, Programme on Ageing (New York, DESA).Available at: http://www.un.org/esa/socdev/ageing/documents/MIPAA_frmwrk.pdf[13 Aug. 2019].Wagner, A. 1876. “Grundlegung der Politischen Ökonomie”, in Teil I: Grundlagen der Volkswirtschaft [in German] [“Basic economics”, in Political Economy,Vol. I].Wentao, W. 2017.“Thoughts on social security within the framework of the‘ideal of great harmony’”,in Chinese Social Security Review, Vol. 1, No. 1, pp. 114–124 [in Chinese, with abstract in English].Werner, R.A. 2014. “How do banks create money, and why can other firms not do the same? An explanation for the coexistence of lending and deposit-taking”, in International Review of Financial Analysis, Vol. 36 (Southamption, UK, Centre for Banking, Finance and Sustainable Development, Southampton Business School, University of Southampton, UK). Available at: https://ac.els-cdn.com/S1057521914001434/1-s2.0-S1057521914001434-main.pdf?_tid=spdf59a089d1-7366-4f82-b572-508934d36ed9&acdnat=1519486355_6fa1399f92fb0d080734e5dd27c cc05e[19Nov. 2019].World Bank. 1993. The East Asian miracle: Economic growth and public policy (Oxford, UK, Oxford University Press).—. 1994. Averting the old age crisis: Policies to protect the old and promote growth, World Bank Policy Research Report (Oxford, UK, Oxford University Press). Available at:http://documents. worldbank.org/curated/en/973571468174557899/pdf/multi-page.pdf [19Nov. 2019].—. 2013. India - Country partnership strategy for the period FY13-FY17 (Washington, DC). Available at: http://documents.worldbank.org/curated/en/207621468268202774/India-Country-partnershipstrategy-for-the-period-FY13-FY17[13 Aug. 2019].—. 2014. Old-age financial protection in Malaysia: Challenges and options, Social Protection Discussion Paper No. 1425 (Washington, DC).—. 2016a. Live Long and Prosper: Aging in East Asia and Pacific(Washington, DC, World Bank East Asia and Pacific).Available at: https://openknowledge.worldbank.org/handle/10986/23133 License: CC BY 3.0 IGO[20Nov. 2019].—. 2016b. Review of the social security system: Considerations for strengthening sustainability120and coverage (Washington, DC). Available at: https://openknowledge.worldbank.org/ handle/10986/24671 [13 Aug. 2019]. —. 2018a. “Data bank: World development indicators and Gini Index”. Available at: h t t p : / / d a t a b a n k . w o r l d b a n k . o r g / d a t a / r e p o r t s . a s p x ? s o u r c e = 2 & s e r i e s = S I . P O V. G I N I # Excel file downloaded on 15.02.2018 [13 Aug. 2019]. —. 2018b. The Socpen and its role in closing the coverage gap among poor elderly Filipinos, World Bank Social Protection Policy Note No. 14 (Washington, DC). Available at: https://openknowledge. worldbank.org/bitstream/handle/10986/29874/127251-PN-PUBLIC-ADD-SERIES-SPL-Policy-Note14-SocPen.pdf?sequence=1&isAllowed=y [16 Oct. 2018]. —. n.d.An eye on East Asia: Labor markets in East Asia during crisis: Developments and policy implications,East Asia and Pacific Economic Management and Poverty Reduction (Washington, DC). Available at: http://documents.worldbank.org/curated/en/515711468245426038/ pdf/678790BRI00PUB067903B0EYE0on0EA0no2.pdf [1 June 2019]. —. n.d. bThe World Bank pension conceptual framework (Washington, DC). Available at: http:// siteresources.worldbank.org/INTPENSIONS/Resources/395443-1121194657824/PRPNoteConcept_ Sept2008.pdf[13 Aug. 2019]. Yan, W. n.d. “China’s pension reform: Implicit pension debt and financing options”, PowerPoint presentation(Washington, DC). Yew, S-Y. 1997. “Malaysia’s Employees Provident Fund and social security”, in J. Midgley; M.W. Sherraden (eds):Alternatives to social security: An international inquiry (Westport, CT, Auburn House, Greenwood Publishing Group).pp. 38–57. Young, T.T.; Chong, H.K. 2015.“The evolution of Singapore’s social security system”, inJ. Midgley; M.W. Sherraden (eds): 1997. Alternatives to social security: An international inquiry (Westport, CT, Auburn House, Greenwood Publishing Group).pp. 89–103. Zhang, Y.; Goza, F. 2006. Who will care for the elderly in China? A review of the problems caused by china’s one child policy and their potential solutions, Working Paper Series 05-07(Bowling Green, OH, Bowling Green State University). Available at:http://citeseerx.ist.psu.edu/viewdoc/ download?doi=10.1.1.500.5109&rep=rep1&type=pdf[13 Aug. 2019]. Zheng, G. 2013. “Top-level design to deepen the reform of the pension system in China”, in Teaching and Research, Vol. 12. [in Chinese] —. 2014. “The historical logic of the evolution of China’s social security”,in Journal of Renmin University of China, Vol. 1. [in Chinese]121ANNEX I: STATISTICAL TABLESTable A1. Scope of social security legislation in the ASEAN Member StatesCountryNumber Scope of policy oflegal areas coverage covered*Existence of programmes anchored in national legislationSick- Maternity Old ness (cash) age (cash)Employment Invalidity Survivors FamilyUnemploymentinjuryallowancesBrunei5 Intermedi- ▲oooDarus-atesalamCambodia 3 Limitedoo●oooNoneNone●●None▲Indonesia5 Intermedi- ▲▲ooooo▲ateLao PDR6 Intermedi- ooooooNoneoateMalaysia4 Limited▲▲ooooNoneoMyanmar4 Limitedoo●oPhilip-6 Intermedi- oooopinesateSingapore 7 Nearlyoooocompre-hensiveThailand8 Compre-oooohensiveViet Nam8 Compre-oooohensive●●o●ooNone▲oooNoneooooooooNote: *Covered by at least one programme. The scope of coverage is measured by the number of social security policy areas provided for by law. This indicator can take the value 0 to 8 according to the total number of social security policy areas (or branches) with a programme anchored in national legislation.Symbols:O =At least one programme anchored in national legislation, including employer-liability programmes based on mandatory risk pooling.● =Legislation not yet entered into force.▲=Limited provision (labour code only).Source: Table B.2. in ILO, 2017d.122Table A2. Features of old-age pensions programmes and legal coverageCountryBrunei DarussalamNoteDate of first lawType of programmeStat. pension age (M/W)Contribution ratesEstimated legal coverage ofand government con- working-age populationtributionInsured / employer %Govern- Total Contributory Non-mentcontrib.% of working-age population1955Provident 55 / 55 fund8,5 / 8,5 Deficit 100.0 coverage59.440.6Cambodia2002 Social insur- 55 / 55 ance……………Indonesia Lao PDR1977Provident fund with social insur-ance1999 Social insurance55 / 55 60 / 602 / 4_ 42.94,5 / 5 Admin. 9.5 Cost10.532.49.5_Malaysia1951 Social insur- 55 / 550,5 / 0,5_ 45.045.0_ance1provident 55 / 5511 / 13_fundMyanmar Philippines………………1954 Social insur- 60 / 60 3.63 / 7.37 Deficit 53.2ancecoverage……53.2_2011Social as- 65 / 65 sistance_ / _ Full cost…coverage……Singapore Thailand1953Provident 65 / 65 fund1990 Social insur- 55 / 55 ance1993Social as- 60 / 60 sistance20 / 17 Supple- 76.4 ment to low-income workers3 / 3 Partial cost coverage_ / _ Full cost…coverage76.4_……Viet Nam2 1961 Social insur- 60 / 557 / 14 Deficit 65.665.6_ancecoverage2004Social as- 60 / 60 sistance_ / _ Full cost…coverage……Note: 1 The Government subsidizes the contribution payment of self-employed persons with a maximum of MYR60 per annum. 2 Total contribution coverage (65.6 per cent) is estimated to be allocated, with 26.4 per cent of contributions mandatory and 39.2 per cent voluntary.Source: Table B.3. in ILO,2017.123Table A3. Overview of pension-related features in ASEANNote: Brunei Darussalam has no official poverty rate. These statistics were released in 2012 by the Community Development Department and Brunei Darussalam Islamic Religious Council and cited in www.thepovertyline.net/brunei.Latest available data is from 2018 (population 60+ years) and 2015 (old-age dependency ratio and life expectancy). Cambodia and the Lao People’s Democratic Republic do not have non-contributory pension allowances. Although some countries have provided national statistical data, in most places, international sources were taken into consideration to allow for consistent representation across countries. n.a. = not available. In Malaysia, the average income replacement rate of the social pension benefit is 31.3 per cent of the minimum wage. In the Philippines, the poverty headcount ratio is calculated at the national poverty line of approximatelyUS$6.20 a day in 2011 PPP.124ANNEX II: COUNTRY BRIEFSCOUNTRY BRIEF: BRUNEI DARUSSALAMElements of the old-age pension system Non-contributory social assistance • Old-Age Pension:The non-contributory demogrant provides a minimum living allowance to allcitizens and permanent residents aged 60 orolder. Non-contributory social insurance • Government Pension Scheme:A non-contributory, defined-benefit scheme for governmentemployees financed entirely from public revenues that provides a full pension, a gratuity or a combination of reduced pensions and a gratuity at retirement. Contributory social insurance • Employees Trust Fund:A defined-contribution scheme with mandatory coverage of public and private sector employees who are citizens or permanent residents of Brunei Darussalam and voluntary coverage of self-employed persons that allows lump-sum withdrawals for retirement, death, disability and housing support. • Supplementary Contributory Pension:A defined-contribution scheme with mandatory coverage of public and private sector employees and voluntary coverage of self-employed persons that provides an additional annuitized pension for 20 years. Voluntary and other provisions • Private insurance policies: provided by most large private sector companies for their employees. • Zakat: A religious obligation contribution administered by the Brunei Islamic Council under the Ministry of Religious Affairs. Zakat contributions under Islamic principles are collected to provide old-age benefits to poor Muslim citizens. Legal framework • 1954: Old-Age and Disability Pensions Act, amended in 1984 • 1992: Employees Trust Fund Act • 2009: Supplementary Contributory Pensions Order1. Development of old-age income security policies The Government Pension Scheme was established in 1959 and served as the main pension system until 1993. A non-contributory, defined-benefit scheme financed entirely through public revenues, it still provides pension benefits to public servants who were hired prior to 1993. The retirement age is 60 years, with some exceptions where early retirement may be taken, at 50 for men and 45 for women. Military and police personnel and prison wardens are covered by separate pension schemes.125The Government Pension Scheme is equal to“months of service”multiplied by“last drawn salary”and divided by 480.The replacement rate is at least 75 per cent of the last drawn salary. Government employees are either pensionable or not pensionable. Those who are eligible for pensions can choose between a full pension and a combination of a reduced pension (usually 75 per cent of the full pension for 15 years, following which the full pension is paid) and a gratuity (usually 25 per cent of 15 years of the full pension). Those who are not eligible for pensions can receive a gratuity. The gratuity benefit varies according to the type of appointment (temporary or permanent), length of service and either the average salary for the last three years of service or the last drawn salary (Sainah and Saim, 2010). In general, the replacement rates are generous (Asher and McArthur, n.d.).On 1 January 1993, under the Employees Trust Fund Act, all new and non-pensionable government employees started contributing to the Fund, while all private sector employees began contributing one year later (Asher and McArthur, n.d.). The Fund is a national, publicly administered provident fund that provides retirement and some social security benefits as well as housing support. On a mandatory basis, it covers all private sector employees and public servants who began working in 1993 or later. It also covers self-employed workers on a voluntary basis. The mandatory contribution rate is 10 per cent of the wage, divided equally between employers and employees. Additional voluntary contributions are allowed. There is no wage ceiling used to calculate contributions.The Fund’s members receive lump-sum payments constituting the employee and employer contributions and dividends earned. The dividend rate is set annually according to the financial health of the fund, interest rates on savings accounts and inflation rates (Pension and Development Network, n.d.). The dividend rate has been positive in real terms since 1996. Six types of withdrawals are possible: (i) retirement withdrawal at 55 years; (ii) pre-retirement withdrawal up to 25 per cent of accumulated savings at 50 years; (iii) survivors’ withdrawal; (iv) emigration withdrawal equal to the full account balance; (v) housing purchase withdrawal up to 45 per cent of the accumulated balance; and (vii) incapacitation withdrawal equal to the full account balance.The Supplementary Contributory Pension was introduced in 2010 to provide an additional annuitized benefit to retirees older than 60 and survivors. It was designed due to concerns about insufficient accumulation of retirement savings in the Employees Trust Fund and to provide a replacement rate of about 50 per cent of pre-retirement income (Sainah and Saim, 2010). Coverage is limited to current contributors to the Employees Trust Fund. The contribution rate is 7 per cent of the monthly wage, shared equally by the employer and the employee. The minimum wage for calculating contributions is 500 Bruneian dollars (BND) and the maximum wage is BND2,800.Self-employed persons make voluntary contributions to the Supplementary Contributory Pension without any minimum or maximum wages for calculating contributions. The Government makes matching contributions (BND42.50 per month in 2008 for workers with reported incomes of BND500 or less) to support low-income workers and to provide incentives for them to contribute. The Supplementary Contributory Pension is paid from age 60, provided that the insured person has at least 35 years of continuous contributions. It is equal to at least BND150 per month, payable for up to 20 years. If the insured person does not meet the requirements, a lumpsum of the accumulated balances is paid (Pension and Development Network, n.d.).A demogrant, named as the Old-Age Pension, was established in 1989 to provide a minimum living allowance to all citizens and permanent residents aged 60 and older. To be eligible, persons born in Brunei Darussalam must have lived there in the 10 years immediately before retirement, and those born outside Brunei Darussalam must have lived there in the 30 years immediately before retirement. The benefit is BND250 per month, or about 17.8 per cent of the estimated average insured wage for Employees Trust Fund members. The Old-Age Pension benefit is adjusted to inflation on an ad hoc basis. It effectively covers 90.9 per cent of the population older than 60 (31,696 persons in March 2017).1262. Socioeconomic backgroundBrunei Darussalam is a small sultanate (with a Malay Islamic monarchy) of 400,000 people that has long maintained its unique balance of tradition and modernity. It has had a stable government since its independence in 1984.The country has large deposits of crude oil and natural gas, which have been the main source of its wealth since their discovery in 1929. With a GDP per capita of US$28,291 in 2017, Brunei Darussalam ranks among the richest countries in Asia. GDP growth stood at 1.3 per cent in 2017. In some months of 2018, the country experienced negative growth as a result of falling oil and gas prices, which account for more than 90 per cent of export revenues.Brunei Darussalam has launched a number of reforms in recent years aimed at attracting new foreign direct investment; diversifying its economy by developing the high-tech, agricultural and manufacturing industries; and fostering private sector growth (Oxford Business Group, 2016). The GDP growth is expected to pick up to 1.5 per cent on average per year, on the back of a modest recovery in global oil and gas prices (IMF, 2017).The Government has an important role in providing income security to its population. The development of the country is guided by its five-year national development plans. The National Vision aims, by 2035, for the country to be recognized for the accomplishments of its well-educated and highly skilled people, their quality of life and its dynamic and sustainable economy. “Ensuring that, as the nation prospers, all citizens are properly cared for” is one of the pillars of the strategy (Sainah and Saim, 2010). Accordingly, social protection has been expanded greatly in the past few decades, and the socioeconomic situation of its citizens has improved dramatically. Large shares of the Government budget are spent on infrastructure, health, education and social protection. Today, about 25–30 per cent of the national development budget continues to be allocated to socioeconomic development.Table CB1. Key socioeconomic and demographic datafor Brunei DarussalamSocioeconomic parametersYearGDP per capita (US$) GDP growth, real (%) Inflation (%) Poverty rate (%)* Gini coefficient Human Development Index (ranking)Demographic parametersTotal population (million) Share of urban population (%) Share of population aged 60 or older (%) Old-age dependency ratio Life expectancy at birth (years) Life expectancy at age 60 (years)Latest available year**20200.4 77.68.5 5.7 76.7 20.528290.60 1.3 -0.2 5.0 n.a. 3920500.4 78.39.6 7.7 77.5 21.02017 2017 2017 201220180.5 85.7 28.7 35.1 82.0 24.4Note: Demographic data are based on UN population projections. *Brunei Darussalam has no official poverty rate. These statistics were released in 2012 by the Community Development Department and Brunei Darussalam Islamic Religious Council and cited in www.thepovertyline.net/brunei [accessed 28Feb. 2019]. Old-age dependency ratio defined as the number or persons age 65 or older over the number of persons aged 15–64. n.a. not available. **=Population, share of urban population, share of older persons: 2018. Dependency and life expectancy: 2015.Data source: DESA Population Division, 2018; IMF, 2018; World Bank, 2018; UNDP, 2018. 127Figure CB1. Brunei Darussalam GDP, 1965–2015 (current US$)Source: World Bank, 2018. Table CB2. Labour market and pension indicatorsfor Brunei Darussalam, latest available yearLabour market indicatorsLabour force participation rate: aged 15 and older (%) Men Women Own-account and contributing family workers (% of labour force) Average monthly wage (US$) Pension indicators Active pension contributors (% of working-age population) Old-age pension recipients (% of population older than 60) Social pension Public social protection expenditure for older persons (% of GDP) Social pension Benefit level, social pension (% of GDP per capita)Year 62.768.9 56.54.3 2978Year n.a. 90.9 90.9 n.a. 0.4 5.72017 2017 2017 2017 20182014 20142014 2015Note: n.a. = not available.Source: ILO, 2017; ILO, 2018a; ILO, 2018b;HAI, 2018.Figure CB2. Brunei Darussalam population pyramid, 2015 and 2050Source: DESA Population Division, 2018 (estimates and medium variant). 128The baby boom after World War II accounts for the bulk of the current population aged 50 or older who will fall into the senior citizens category in a few years. With the rapid development in local health delivery systems, the life expectancy at birth is now 76.6 years for men and 79.8 years for women. The overall dependency ratio is high due to a large proportion of children younger than 15. The old-age dependency ratio was 5.7 in 2015 and is expected to increase to 35.1 by 2050.Although women have longer average life expectancy than men, they have lower labour force participation, especially in the formal sector. Furthermore, until 2010, married women without a first university degree had only month-to-month contracts with employers in the public sector. Thus, women earn lower average wages than men and thereby have lower pension contributions. This makes them more vulnerable to old-age poverty (Sainah and Saim, 2010).Although official statistics on the number of migrant workers in Brunei Darussalam were not available, the United Nations Department of Economic and Social Affairs estimatedthere are about 2,000 net migrants in the country (WPP, 2018). On the other hand, Asher and McArthur (n.d.) estimated that foreign workers constituted more than 40 per cent of the labour force in the mid-1990s.3. Old-age pension challengesBrunei Darussalam is one of the few countries among the ASEAN Member States that has achieved 90.9 per cent, or near universal, pension coverage. This is largely attributable to the Government’s commitment to provide income security to its population, effectively realized through the universal Old-Age Pension as well as favourable economic and fiscal conditions that were utilized for social protection. There are still some gaps in coverage, however. Foreign workers who amount to a sizeable number in the labour force are legally exempt from the Employees Trust Fund. While many private sector employers provide their own pension plan for employees, those workers who fall into the lower-income brackets are likely to not be part of any pension provision.Family and social structures are still quite traditional, and an extended family system is practised. Older people live with and are cared for by their children. However, in recent years, there has been a gradual deterioration of family structures (Sainah and Saim, 2010). Over time and with ageing of the population, this can put pressure on older people who do not receive adequate pension benefits.For the Employees Trust Fund alone, the accumulation savings at retirement were too low to support a meaningful replacement of pre-retirement income. This has improved considerably with the introduction of the Supplementary Contributory Pension. As for the Old-Age Pension, the benefit level is equivalent to 5.7 per cent of per capita GDP in Brunei Darussalam (Pension Watch, 2018), which may not appear as an adequate level of benefit. However, the benefit measured as a percentage of the international poverty line is equal to 657 per cent, which shows that it is far more generous than in other countries that have introduced social pensions.The Government Pension Scheme, which is governed by the Old Age and Disability Pensions Act (1954), is not considered adequate for older people with disabilities. If an older person suffers from a disability, they require additional support in the form of medication, appliances, nursing care, etc. However, only one type of pension is provided, either for people with disabilities or older people. If the pension is insufficient to cover all the necessities, a written application can be made for additional allowances, and all applications are dealt with on a case-by-case basis, although there is no guarantee that an application will be successful (Sainah and Saim, 2010).Guaranteeing adequacy of benefits under the contributory system constitutes the greatest challenge for the national pension system. To provide adequate benefits and thereby guarantee129old-age income security, the contribution rates need to be increased. Furthermore, in light of the ageing population and increasing life expectancy (table CB1), financing for the Old-Age Pension will need to be increased in the future to ensure adequate benefit levels.To achieve higher labour mobility, there needs to be convergence in basic parameters between the different pension schemes, such as the retirement age (World Bank, 2007). Although Brunei Darussalam is ageing, the demographic situation is not dramatic. With high per capita GDP and growing life expectancy, the pension financing issues should be manageable mainly by way of increasing retirement ages to 60 and 65 and making more use of the Old-Age Pension as a policy instrument.Projected costs of military pensions are expected to rise in the years to come. With 90 per cent of government funding for old-age income security provisions being derived from oil and gas revenues, diversification measures are necessary to make the system less vulnerable to price shocks (World Bank, 2007). The Employees Trust Fund and zakat funds are invested both locally and in foreign markets to gain maximum returns. However, this is not done in a fully transparent way, with the sources of funding and places where Employees Trust Fund monies are invested are not revealed to members (Sainah and Saim, 2010).4. Policy directionsThe main policy directions consist in continually reforming the pension system to stay abreast of demographic, social and economic changes. Regular, periodic adjustments of the Old-Age Pension benefit with inflation to preserve the value of the pension should be done. The Employees Trust Fund savings, which are withdrawn as a lumpsum at retirement do not provide income security in old age because retirees can spend the money in a short period of time. The Supplementary Contributory Pension provides a pension for 20 years, although life expectancy at age 60 is currently slightly more than 20 years and expected to increase further in the coming years. To provide adequate income security in old age, the pension should be provided throughout each person’s life.Old-age income security in Brunei Darussalam: Policy messages• Adjust the Old-Age Pension benefit to inflation on a periodic basis.• Consider harmonizing the retirement age between different pension schemes and increasing it gradually to 60 and then 65.• Consider increasing the Employees Trust Fund and Supplementary Contributory Pension contribution rate for adequate pensions.• Extend the Supplementary Contributory Pension benefit throughout the pensioner’s life.At the current legal retirement age of 60, the average life expectancy will increase from 21 years in 2015 to about 24 years in 2050. For such long life expectancies, the Employees Trust Fund contribution rate of 10 per cent is very low and not sufficient, even after considering the Supplementary Contributory Pension contributions of 7 per cent. Well-organized increase of the retirement age to 65 can solve half the problem. A further increase of the contribution rate of at least 8 per cent (or provision of an equivalent tax-financed subsidy) would be required for achieving relatively decent pension levels upon retirement. Any options to withdraw the money early should be reduced to a minimum; at retirement, funds should be annuitized as a rule.130Table CB3. Pension scheme parameters for Brunei DarussalamEmployees Trust Fund Supplementary Contributory PensionPrivate and public sector employeesQualifying Contribution years conditions Pensionable ageBasic formulaPension formulaMinimum benefit Maximum benefit Base salary55 Lump sum + accrued interest---35 60 Monthly benefit paid for up to 20 yearsBND150 -Benefit indexation--% of gross earningsEmployeeat least 5% BND17.50at least 3.5% or BND17.50ContributionsSelf-employed5%EmployerGovernment-Earnings base MeasureMinimum-3.5% or BND17.50Matching contributions for those earning less than BND500; BND17.50 for self-employedBasic salaryBND500Basic salaryMaximum-BND2,800Old-age pensionQualifying conditionsBenefit amountLocal currency% of GDP per capitaAge 60 and resident for previous 10 years (30 years if non-citizen) BND2505.7%Note: The Supplementary Pension Fund was implemented in 2010. The Government retroactively pays contributions into the Fund for any periods before 2010 where the worker participated in the provident fund. The retroactive contributions count towards the minimum contribution period of 35 years.Source: HAI, 2018; Pension and Development Network, n.d.REFERENCESAsian Development Bank.2017. “Older people in Brunei Darussalam”, presentation to the 3rd ASEANJapan Active Ageing Regional Conference, 26–27 June, Manila. Available at: http://aging-asia.info/ sites/default/files/presentations/s4p4-brunei.pdf[13Aug. 2019].Asher, M.; McArthur, I. n.d. Country profile on Brunei Darussalam. Available at: http://citeseerx.ist.psu.edu/viewdoc/ download;jsessionid=7549AA20CCB36F59953BAEC9015DA767?doi=10.1.1.198.2948&rep=rep1 &type=pdf[13 Aug. 2019].HelpAge International (HAI). 2018.“Pension watch: Country profile on BruneiDarussalam”.Available at:www.pension-watch.net/pensions/country-fact-file/brunei-darussalam[13 Aug. 2019].131International Monetary Fund (IMF). 2018. World economic outlookupdate, July(Washington, DC).Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[13 Aug. 2019]. International Labour Organization. 2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva). —. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu [13 Aug. 2019]. —. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[13 Aug. 2019]. Ong, C.; Peyron-Bista, C. 2015. The state of social protection in ASEAN at the dawn of integration (Bangkok, ILO Regional Office for Asia and the Pacific). Oxford Business Group. 2015. The report: Brunei Darussalam 2016.Available at: www.oxfordbusinessgroup.com/brunei-darussalam-2016/economy[13 Aug. 2019]. Pension and Development Network. n.d. “Country profile on Brunei”. Available at: www.pensiondevelopment.org/51/brunei.htm[13 Aug. 2019]. Sainah, H.; Saim, H. 2010. Social protection in Brunei Darussalam – Current challenges. Available at: www.eria.org/publications/research_project_reports/images/pdf/y2009/no9/ CH-05_Brunei_pp.124-156.pdf[28Feb. 2019]. Tant, E.M. 2014. Equity in access to healthcare in Brunei Darussalam: Results from the Brunei Darussalam Health System Survey (HSS). Available at: https://dukespace.lib.duke.edu/dspace/handle/10161/8817[13 Aug. 2019]. United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/ wpp/DataQuery/ [13Aug. 2019]. United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/ sites/default/files/2018_human_development_statistical_update.pdf [13 Aug. 2019] Unknown. n.d. “Country report:Brunei Darussalam”, presentation at the11th ASEAN & Japan High Level officials Meeting on Caring Societies: Active Aging, 3-5 December 2013, Tokyo. Available at: www.mhlw.go.jp/bunya/kokusaigyomu/ASEAN/2013/dl/ Brunei_CountryReport.pdf[13 Aug. 2019]. World Bank. 2007. Brunei pension report(Brunei Darussalam, Centre for Strategic and Policy Studies). —. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [13 Aug. 2019]. —. n.d. ASPIRE, Pensions database (Washington, DC).132COUNTRY BRIEF: CAMBODIAElements of the old-age pension systemSocial insurance National Social Security Fund for Civil Servants: A public scheme that covers all civil servants.Benefits provided include retirement, survivor, death, invalidity and work injury. It is financed through a pure pay-as-you-go strategy and funded from state revenues (non-contributory). National Fund for Veterans: The Fundcovers war veterans, members of the armed forces and the national police force. Benefits provided include sickness cash, maternity, marriage allowance, work injury, retirement, disability, death and survivor benefits. It is also a tax-financed, pure pay-as-you-go scheme.Legal frameworkIn 2002, national legislation included a provision for the launch of mandatory old-age pensions for the private sector. So far, this provision has not been implemented. 1994: Law on Common Statute of Civil Servants Law on Pensions and Invalidity Pensions for the Soldiers of the Cambodian Armed Forces, as amended by the Royal Decree NS/RKT 0406/008 of 2006 and Royal Decree NS/RKT 0710/595 of 2010 2002: Law on Social Security Schemes for Persons Defined by the Provisions of the Labour Law 2008: Royal Decree No. 0108/039 (conversion of civil servant benefits to contributory)1. Development of old-age income security policiesCambodia is an outlier in the region, together with Myanmar, in terms of the relatively recent start of a pension system for private sector workers. Social security entitlements were first introduced in Cambodia in 1955 with the adoption of Royal Decrees No. 55 and No. 306 (ILO, 2012). These provisions fell victim to the Khmer Rouge regime (1975–79). A Department of Social Security was established again only in 1993, under the Cambodian Ministry of Social Affairs, Veterans, and Youth Rehabilitation.The right to social security in Cambodia is recognized by the 1993 Constitution. Article 36 stipulates that “Every Khmer citizen shall have the right to obtain social security and other social benefits as determined by law” (ILO, 1999, p. 6), whileArticle 75 states: “The State shall establish a social security system for workers and employees” (ILO, 1999, p. 11).By 1994, civil servants gained comprehensive social security coverage under the National Social Security Fund for Civil Servants (NSSF-CS),1 including retirement pensions, with benefits fully funded from general state revenue. In addition, the National Fund for Veteranswas established in 2010 to cover armed forces and is a similar system to the NSSF-CS in terms of financing strategy, benefits provided and scheme parameters.1 Law on Common Statute of Civil Servants of the Kingdom of Cambodia.133Cambodia’s first social security law2 was adopted in 2002, setting insurance provisions for work injury, health care (medical care, sickness and maternity leave), old age, invalidity and survivorship for the private sector and for the launch of a pension branch in the public administration; most of the provisions are yet to be implemented.The National Social Security Fund (NSSF) began operating in 2008 and introduced work injury insurance for private sector workers, with compulsory membership for enterprises with at least eight employees. Provisions of medical and maternity insurance followed in 2014. By late 2018, the NSSF implemented the Social Health Insurance scheme for persons defined by the provisions of the labour law, amounting to around 1.55 million members. From January until December 2018, the NSSF covered 207,510 civil servants, 55,222 former civil servants (pensioners) and 54,771 veterans with the Social Health Insurance.Cambodia’s National Social Protection Policy Framework 2016–2025 was launched in 2017, following the advice of a multisector working group. For the first time in Cambodia, a policy framework brings social security and social assistance under one umbrella and includes indicators on reforms related to the old-age pension system. Features of the framework include:• development and implementation of an old-age pension scheme for the private sector;• implementation of an old-age pension scheme for non-poor workers in the informal economy;• implementation of a voluntary old-age pension scheme;• unification of the pension schemes under one management system, expansion of coverage of public employees and the voluntary exit of service for public employees;• review of the contribution rate for the pension scheme of all public employeesand a change from non-contributory to a contributory system; and• development of a legal framework for the management, monitoring and implementation of all public pension schemes to ensure consistency of pensions.2. Socioeconomic backgroundWhile overcoming the legacy of severe, decades-long political instability and civil war, Cambodia’s economic performance in recent decades has been among the top ones in the world, with average annual GDP growth of more than 7.5 per cent during 1994–2015, and GDP per capita at nearly US$1, 400 in 2017 (table CB1), up from around US$200 in 1992. As a result of such robust economic growth, Cambodia is, since 2016, no longer classified as a low-income but as a lower-middle-income economy, according to the World Bank income classification, which is primarily based on countries’ gross national income per capita levels (World Bank, 2018). As of 2018, Cambodia remained one of the least developed countries in the world (UNPD, 2018).The economy is dominated by the agriculture sector and the export-led garment, textile and footwear industry, which represented 78 per cent of total merchandise exports in 2016 (ILO, 2017d), and two-thirds of total manufacturing employment in 2012 (ILO, 2017c). Tourism in Cambodia continues to expand. Cambodia also has offshore oil and natural gas deposits, but these are still to be commercially extracted and could potentially attract foreign investment if managed properly (ADB, 2014). In 2013, most jobs were either in low-or middle-skill occupations, and agriculture was the largest sector of the economy, responsible for 64 per cent of total employment (ILOSTAT, 2018).2 Law on Social Security Schemes for Persons Defined by the Provisions of the Labour Law.134Some of the central issues regarding labour market conditions in the country include high levels of informality and general lowquality of employment. In 2012, an estimated 90.2 percent of total employment outside the agriculture sector was informal (ILOSTAT, 2018). In 2017, more than half of all workers, or 54 percent, were in vulnerable employment, as own-account workers or contributing family workers (ILO, 2017b). Workers in the garment industry have been the target of significant increases in minimum wages recently (ILO, 2018a). However, they constitute a small share of the employed in relation to the rural economy.Table CB1. Socioeconomic and demographic datafor CambodiaSocioeconomic parameters GDP per capita (US$) GDP growth rate, real (%) Poverty rate (%) Gini coefficient Human Development Index (ranking)1389.6 (2017) 6.8 (2017)13.5 (2014) 30.8 (2012) 143/188 (2016)Demographic parametersLatest available data*20202050Population (million)16.216.722.0Share of urban population (%) Share of population aged 60 or older (%)Old-age dependency ratioLife expectancy at birth** (years) Life expectancy at age 60** (years)23.424.241.26.47.616.96.47.618.467.669.577.517.117.521.5Note: *Population and share of urban population: 2018.Share of older persons: 2008 census data.Dependency and life expectancy: 2015. **2020 projections correspond to 2015–20 period.Data source: DESA Population Division, 2018; IMF, 2018; UNDP, 2018; World Bank, 2018.Current incentives to release labour from agriculture are low because farmers hold on to land as a safety net. In addition, with a still relatively young population, thousands of young people join the workforce each year but face issues that include skills mismatch, poor education and lack of training. There is a growing migratory outflow of Cambodian workers to neighbouring countries, especially Thailand. An estimated one million workers have left the country, but they provide valuable income support to households in Cambodia through their remittances (ADB and ILO, 2015).Figure CB1. Macroeconomic performance of Cambodia and Eastern Asia and the PacificPanel A: GDP growth, by decade average (%) Panel B. Stability of growth, by decade Data source: University of Groningen, 2013; World Bank, 2018. 135Table CB2. Labour market and pension indicatorsfor Cambodia, latest available yearLabour market indicators Labour force participation rate: aged 15–59 (%)Men WomenOwn-account and contributing family workers (% of labour force) Average monthly wage (US$)Year81.0201788.5201577.2201547.320151212013Pension indicatorsActive pension contributors (% of working-age population: aged15–64)0.0Pension recipients (% of population older than statutory pension age: 55)3.2Public social protection expenditure for older persons (% of GDP)0.2Civil servant and veteran pensions0.2Average benefit, civil servant pensions (% of GDP per capita)80.1Year20172015 2010 2010 2009Data source: ILO, 2017e; ILO, 2018b; ILO, 2018c.Figure CB2. Cambodia’s population pyramid, 2020 and 2050Note: UN population projections.Data source: DESA Population Division, 2018.Cambodia’s economic achievements in the past two decades have also resulted in unprecedented reduction of the poverty rate, which declined to 13.5 per cent in 2014, from 50.2 per cent in 2004, as measured by the national poverty line.3 Yet, because poverty dynamics remain a delicate balance, as those who are just above or below the threshold could face (further) poverty spells in case of external shocks. For example, slightly higher poverty thresholds, such as the US$2 per day defined around 40 per cent of the population as poor in 2011, while at the US$3 per day line, the share, rose to 72 per cent (ADB and ILO, 2015).3 Per capita income less than 3,871 Cambodian riels (KHR) per day.136In spite of significant relative poverty reduction, income inequality remains a grave concern in Cambodia, and there are substantial welfare disparities between the urban and rural populations, and even higher within urban areas (ADB and ILO, 2015). Gender-based inequalities are rife, too, with women being disadvantaged in pursuing secondary and higher education, high-skill jobs and decision-making roles. Within countries in ASEAN, however, Cambodia’s income inequality is considered to be relatively low.Cambodia has a relatively young population, with roughly 60 per centyounger than 30 in 2018 (DESA Division Population, 2018). In 2013, youth, defined as persons aged 15–24, represented 24 per cent of the labour force (ILOSTAT, 2018). In roughly 30 years, the share of the older population among the total population is projected to rise 1.4 times (table CB3). This rapid ageing process until 2050 can be primarily attributed to declining trends of fertility and improvements in life expectancy, as well as negative net migration (ILO, 2017a).3. Old-age pension challengesIn relative terms, Cambodia is experiencing one of the least severe demographic pressures at present, with 7.6 per cent of the population aged 60 or older in 2017 (table CB3), which was a smaller share than Myanmar (at 9.4 per cent), Viet Nam (at 11.1 per cent) and Thailand (at 17.1 per cent).Demographic trends are expected to remain relatively unchallenging in the coming decades (figure CB2), with the old-age dependency ratio reaching 18.4 per cent by 2050 (while it is projected to be 52.5 per cent in neighbouring Thailand, for instance). Thus, financing old-age pensions should not pose major challenges from a demographic perspective for the foreseeable future in Cambodia.Yet, given the high rate of informality in the labour force and low-wage levels (table CB2), establishing a pension scheme with a contribution-based financing system constitutes a major challenge, particularly if based on the traditional social insurance model. Cambodia’s labour market characteristics show that, even in the formal sector, there is a high presence of short-term contracts, which would potentially generate low density of contributions.As one of the least economically developed countries among the ASEAN Member States, as well as youngest in demographic terms, Cambodia’s two main current challenges with regards to the development of old-age income security policies can be summarized as: (i) to design and implement a well-functioning pension scheme that covers persons in the formal economyin the private sector; and (ii) to identify mechanisms to provide pension coverage to workers in the informal economy. Oldage pension models being discussed include an integrated system that would include compulsory contributory pensions for workers in the formal sector, with a tax-funded scheme based on a flat pension for those with less contributory capacity.With regards to the current funding strategy for the NSSF-CS, which covers only civil servants without contributions paid from the workers covered, reform is under discussion. The challenge would be to start levying contributions from civil servants, as recommended by the ILO (2017) and gradually approximate the conditions of the civil servants’ scheme with those for the scheme to be established for private sector workers.The NSSF-CS provides benefits of 80 per cent of the final salary, although the pension formula does not apply to the total salary but to the basic salary only (table CB3). According to available data (ILO, 2017a), basic salary accounts for about 58 per cent of total remuneration paid to workers,4 and the allowances for the position, location and risks associated with the job account for 42 per cent of total earnings. This means the effective accrual rate is 3 per cent per year of service for the first 204 Excluding the allowances for spouses (KHR6,000 per month) and children (KHR5,000 per month and child).137years and 2 per cent thereafter, in terms of the total salary. Hence, the maximum pension is 455 per cent of the salary, which is close to the minimum standards established under ILO Conventions No. 102 and No. 128 (ILO, 2017a).Moreover, the NSSF-CS pay-as-you-go rate6 is expected to rise from 21.3 per cent in 2016 to 86.5 per cent in 2115, mainly due to increase in the dependency ratio (table CB3). Because this estimation considers only the basic salary, the pay-as-you-go rate is currently 12.4 per cent. To avoid increasing costs for the treasury, it is important to reform the parameters of the current scheme.4. Policy directionsA law on social security schemes is planned to be drafted and implemented by the end of 2019. The launch of a mandatory pension scheme for the private sector is in line with action plans of the NSSF and ILO Convention No. 102.In addition, there are ongoing policy discussions on transforming the current scheme for public employees from a tax-financed to a contribution-financed system. In this context, an assessment on the possibility of providing the same contribution-based old-age pension benefits for the public and private sector workers is needed. The draft law on social security schemes is expected to cover persons in the public sector and persons defined by the provisions of the labour law, including personnel in air and maritime transportation, household workers and self-employed persons. Portability (public, private and self-employed) is also a strong objective of the ongoing reform. The NSSF is expected to work on a memorandum of understanding in relation to migrant workers with other countries to ensure portability and exportability of benefits.OLD-AGE INCOME SECURITY IN CAMBODIA: POLICY MESSAGES• Establish a defined-benefit, contribution-based, old-age pension scheme for workers in the private sector.• Establish a tax-based, universal, flat-rate, old-age pension (demogrant).• Revise or reform the current tax-based scheme for the public sector.• Develop a voluntary defined-contribution old-age pension scheme.Considering the features of the current labour market, as well as the demographic and socioeconomic trends, a comprehensive and inclusive Cambodian national pension system could consist of a universal, flat-rate, old-age pension coupled with a social insurance-based defined-benefit scheme and complemented by voluntary, privately run defined-contribution schemes.Taking into consideration the high level of informality in the economy and the time that a contributory system requires to mature, the implementation of a tax-funded social pension will be essential to ensure that a significant share of the older population in Cambodia can benefit from oldage income security in the near and mid-term future. Without that component, the percentage of older persons receiving a pension risks being low for a relatively long time, meaning that individuals and households will need to cope alone with the financial burden of ageing.5 58 per cent x 80 per cent. 6 Total expenditures as a percentage of insurable earnings (which is called the pay-as-you-go rate).138Table CB3. Pension scheme parameters for CambodiaCivil servants and veterans Qualifying conditions Contribution years (CY)Pensionable agePension formulaBasic formula Minimum benefit Maximum benefit Base salary (BS)ContributionsBenefit indexation Government20 55 (60%+2%*(CY-20))*BS Depending on salary grade 80%*BS Final basic salary plus allowances, excluding position allowanceOnly on adhoc basis Fully funded from general revenueREFERENCESSource: ILO, 2017a.Asian Development Bank (ADB). 2014. Cambodia: Diversifying beyond garments and tourism: Country diagnostic study(Manila). Available at: www.adb.org/sites/default/files/publication/149852/ cambodia-diversifying-country-diagnostic-study.pdf[13 Aug. 2019].—;International Labour Organization (ILO). 2015. Cambodia: Addressing the skills gap (Manila, ADB). Available at: www.ilo.org/wcmsp5/groups/public/---asia/---ro-bangkok/documents/publication/ wcms_425375.pdf [13 Aug. 2019].HelpAge International (HAI). 2018. “Pension watch: Country profile on Cambodia”.Available at:www. pension-watch.net/country-fact-file/cambodia/[13 Aug. 2019].International Monetary Fund (IMF). 2018. World economic outlook update, July 2018 (Washington, DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[13 Aug. 2019].International Labour Organization. 1999. Constitution of the Kingdom of Cambodia (Bangkok). Available at: www.ilo.org/dyn/natlex/docs/ELECTRONIC/35789/72242/F1994980854/KHM35789. pdf[13 Aug. 2019].—. 2012. Cambodia: Social security for the formal economy: Outlook and challenges ahead(Phnom Penh). Available at: www.ilo.org/wcmsp5/groups/public/---europe/---ro-geneva/---ilo-brussels/ documents/publication/wcms_196011.pdf[13 Aug. 2019].—. 2017a. Actuarial valuation of the National Social Security Fund for the Civil Servants and actuarial study on the implementation of a new pension scheme in the private sector (Phnom Penh).—. 2017b. “Cambodia: Employment and environmental sustainability fact sheets 2017” (Phnom Penh). Available at: www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/ publication/wcms_627803.pdf[13 Aug. 2019].—. 2017c. Developing Asia’s garment and footwear industry: Recent employment and wage trends (Bangkok). Available at: http://www.ilo.org/wcmsp5/groups/public/---asia/---ro-bangkok/ documents/publication/wcms_581466.pdf—. 2017d. “Garment sector in Cambodia: Declining employment despite rising exports”. Available at: www.ilo.org/asia/media-centre/news/WCMS_555470/lang--en/index.htm[13 Aug. 2019].139—. 2017e. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva). —. 2018a. How has garment workers’ pay changed in recent years? Available at: www.ilo.org/ wcmsp5/groups/public/---asia/---ro-bangkok/documents/publication/wcms_631686.pdf[13 Aug. 2019]. —. 2018b. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[13 Aug. 2019]. —. 2018c. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[13 Aug. 2019]. United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [13Aug. 2019]. United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[13Aug. 2019]. University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[13 Aug. 2019]. World Bank. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [13 Aug. 2019]. —. n.d. ASPIRE, Pensions database (Washington, DC).140COUNTRY BRIEF: INDONESIAElements of the old-age pension system Non-contributory social assistance • Social Assistance for Older Persons (Asistensi Sosial Usia Lanjut – ASLUT, formerly JSLU, JaminanSosial Lanjut Usia. tax-funded, means-tested, monthly social pension for persons aged 60 and older and without other income sources. Contributory social insurance • Taspen:A retirement savings and pay-as-you-godefined-benefit monthly pension scheme for civil servants. • Asabri:A scheme for armed forces and military personnel, similar to Taspen. • Jaminan Pensiun: A new scheme for private sector workers under administration of BJPS Employment (Ketenagakerjaan). It is adefined-benefit scheme with mandatory matching contributions from employers and employees in formal employment and contributions only from workers in self-employment. Provident fund Jaminan Hari Tua: A defined-contribution scheme with mandatory participation for all workers in the private sector, with matching contributions to be paid from employers in case of wage employees.Voluntary savings schemes • Dana Pensiun Pencari Kerja: Employer-sponsored pension funds are defined-benefit ordefined-contribution schemes that can be established by one or multiple employers for their employees. Defined-benefit plans offer options for lump-sum or monthly pension payments, and contributions can be made by employers only or by employers and employees. Definedcontribution schemes offer pension pay-outs that correspond to the workers’ accumulated contributions and related investment earnings.• Dana Pensiun Lembaga Keuangan: Financial institution pension funds are defined-contribution schemes that can be established by banks or life insurance companies for persons with or without links to the labour market.Legal framework • 1992: Act No. 3 on Employees’ Social Security • 1998: Law No. 13 on Welfare for Elderly People • 2004: Government Regulation No. 43 on Efforts to Improve the Social Welfare of Elderly People • 2004: Law No. 40 on the National Social Security System • 2011: Law No. 24 on establishing Social Security Administrator for Health Agency, BPJS • 2015: Government Regulations: No. 45 on the social insurance; No. 46 on the providentfund; No. 60 on the implementation of the provident fund1411. Development of old-age income security policiesAround the 1970s, Indonesia launched three social security schemes for different categories of public and private sector employees. Astek, or Employees’ Social Insurance, was a provident fund covering formal private sector and state-owned enterprise employees. It consisted of a savings scheme financed from employees’ own contributions, which provided lump-sum cash benefits at retirement age, among provisions for other social contingencies during contribution periods. Taspen and Asabri, which exist even today, are defined-benefit,pay-as-you-go schemes covering civil servants, military and police personnel (ILO, 2012; Tambunan and Purwoko, 2002).By 1992, with the Act on Employee’s Social Security, the country started to reform the system for private sector workers. The new system, Jamsostek, replaced Astek and was mandatory for all companies with 10 or more employees or a pay-roll greater than 1 million Indonesian rupiah (IDR), with a few exceptions. Under Jamsostek, employers and employees contributed at a higher rate than before to the provident fund, and the retirement age was set at 55 (Tambunan and Purwoko, 2002). Only in 2002, after a constitutional amendment, were Indonesians granted constitutional rights to social security, and the responsibility of the State in developing a social security system was recognized. In 2004, the national social security law was enacted. It is considered the most important legal instrument to base the development of the social security system in Indonesia (ILO, 2012).The 2004 law extended mandatory social security coverage to all citizens and set a multi-tier coverage approach to be progressively implemented to achieve that end. It includes the adoption of non-contributory and contributory social insurance schemes for different categories of workers and individuals without a link to the labour market. Nevertheless, until 2015, old-age income security of the private sector was based only on a provident fund (Jamsostek) which, in practice, not only had low old-age income security capacity due to the high levels of withdrawals before retirement age, low levels of accumulated savings and the absence of an automatic conversion of savings into annuities but it also excluded the entire informal economy from participating (ILO, 2012).Only after the BJPS Employment (Ketenagakerjaan) was established as the new social security provider in 2011 and with the adoption of regulations in 2015 that were needed to support the national social security law, was participation in the provident fund, now called Jaminan Hari Tua, extended to all companies and individuals in Indonesia. In addition, a new defined-benefit, old-age pension scheme was introduced, the Jaminam Pensiun, with mandatory coverage of all workers in the private sector, both in formal and informal employment (ISSA, 2018; ILO, 2012).Other than mandatory schemes, Indonesia’s current old-age pension system includes privately managed voluntary savings schemes for all persons, which can be established by either employers or financial institutions, such as banks or insurance companies (ISSA, 2018; ILO, 2012).Indonesia’s current first pension pillar is based on the Asistensi Sosial Usia Lanjut (ASLUT). This is a means-tested and targeted scheme that aims to cover the most vulnerable older people, with monthly pensions set at the level of the national poverty line (HAI, 2018). It was first introduced in a few provinces in 2006, under administration of the Ministry of Social Affairs. In 2013, the coverage was 26,500 persons, or 0.1 per cent of people older than the eligibility age of 60 years, due to poor targeting and limited budget (ILO, 2017c).2. Socioeconomic backgroundIndonesia is an archipelago of more than 17,000 islands, the world’s fourth most populous country as well as the biggest economy in South-Eastern Asia. Its GDP per capita is one of the highest among ASEAN Member States (table CB2). After the 1997–98 Asian financial and economic crises, substantive social, economic and political progress was made due to major reforms in and across all these spheres. From a military-led to a more democratic and decentralized public administration142and to an increasingly open and market-based economy (IMF, 2018), the economy grew at an average of roughly 5 per cent annually. The country also developed a relatively comprehensive social protection system, which enabled health and educational outcomes of the population to improve dramatically.Table CB1. Socioeconomic and demographic data for IndonesiaSocioeconomic parameters GDP per capita (US$) GDP annual growth rate, real (%) Poverty rate* (%) Gini coefficient Human Development Index (ranking)Year 38475.1 5.7 39.5 1162017 2017 2017 2013 2018Demographic parametersPopulation (million) Urban population share (%) Share of population aged 60 or older (%)Old-age dependency ratio Life expectancy at birth (years) Life expectancy at age 60 (years)Latest available data**266.8 55.3 8.9 7.6 68.62020272.2 56.6 9.5 8.5 69.4 16.72050321.6 72.8 19.2 20.7 73.5 18.7Note: *Poverty defined as the population living below the US$1.90 per day (2011 PPP) threshold. **Population, share of urban population, share of older population: 2018. Dependency ratio and life expectancy: 2015.Data sources: DESA Population Division, 2018;HAI, 2018; UNDP, 2018; World Bank, 2018.Extreme income poverty, as defined by the US$1.90 per day international poverty line, has decreased steeply since the 2000s, from 39.3 per cent to 5.7 per cent in 2017 (World Bank, 2018). Indonesia’s human development, defined according to health, education and national income indicators, ranked 116 in 2018 on the Human Development Index, which places the country among others with “medium human development”, and ahead of only four ASEAN Member States (Cambodia, the Lao People’s Democratic Republic, Myanmar and Viet Nam, despite being the largest ASEAN Member States economy (UNDP, 2018).Figure CB1. Macroeconomic performance of Indonesia and Eastern Asia and the PacificPanel A: GDP growth, by decade average (%)Panel B. Stability of growth, by decadeData source: University of Groningen, 2013, World Bank, 2018.143Table CB2. Labour market and pension indicatorsfor Indonesia, latest available yearLabour market indicators Labour force participation rate (aged 15–59)Men Women Own-account and contributing family workers (% of labour force) Average monthly wage (US$)Year201786.1201753.3201757.020171552015Pension indicators Active pension contributors (% of working-age population:aged 15–64)Private sector provident fund (JHT) Social insurance pension (BPJS) Voluntary occupational pensions Old-age pension recipients (% of population older than statutory pension age: 55) Public social protection expenditure for older persons (% of GDP)Year7.720157.620153.820151.8201014.020150.52010Data source: HAI, 2018; ILO 2017b; ILO, 2018a; ILO, 2018b.Figure CB2. Indonesia’s population pyramid, 2020 and 2050Note: UN population projections.Data source: DESA Population Division, 2018.3. Old-age pension challengesIndonesia is still at a relatively early stage in the demographic ageing transition, and will experience a lower-than-average increase in number of older people and dependency ratio by 2050 (table CB3).Nevertheless, while the population is expected to grow 1.2 times by 2050, the share of the population aged 60 years or older will more than double in the same period (table CB3 and figure CB2), and the old-age dependency ratio is expected to nearly triple. Life expectancy at age 60, still comparatively 144low among ASEAN Member States, is expected to increase only slightly by 2050 (table CB3). Internal migration is an additional challenge. The share of urban population is projected to increase 1.3 times by 2050, which, if not properly managed, can create even larger informal markets, with a class of workers in vulnerable employment and lacking formal protection against old-age and other social contingencies.With roughly a fifth of the population projected to be at pensionable age in 30 years from now, the limited coverage of Indonesia’s old-age pension system is an urgent issue that can potentially harm the significant socioeconomic progress achieved in the past couple of decades. As seen in table CB2, only 14 per cent of older persons received formal pensions in 2015, either contributory or funded by the State. Moreover, until 2015, the informal economy was entirely excluded from the existing publicly managed mandatory savings or pension schemes. As a result, only 108,000 workers participated in privately managed voluntary schemes in 2017, which corresponds to a rate that is close to zero per cent of workers in informal employment. Only a small proportion of older persons was covered by the social pension, ASLUT.The majority of the labour force is made up of non-salaried workers in vulnerable forms of employment, namely own-account or contributing family work, accounting for 57 per cent of all workers (table CB2). In this context, designing an appropriate mechanism to increase participation of this large group in pension savings or social insurance pension schemes will remain a challenge.The new social pension scheme, Jaminan Pensiun, covers only wage-earning employees who were required to register under the BJPS Employment by 2015. The remaining population is required to register by 2019 (ISSA, 2018). Even so, old-age pension coverage in the formal sector remains low, primarily due to limited compliance with mandatory participation requirements. In 2016, the BJPS had only 160 inspectors responsible for supervising compliance of employer and employee registration in the new social security system and for reporting to labour inspectors. Improvement is needed in the coordination between the BJPS Employment with labour inspection bodies and provincial and district administration levels (ILO, 2017b).The two old-age schemes reached 24.5 million actively paying members in 2017– with 14.3 million members in Jaminan Hari Tuaand 10.2 million members in Jaminan Pensiun7 – representing roughly 20 per cent of an estimated labour force of 127 million workers (ILO, 2017a). Most of the population covered in these schemes is in the highest income distribution quintiles. The current system mainly covers workers in formal employment, which can be attributed to the lack of better and more redistributive, non-contributory old-age pension or social pension schemes.As for the main issues in the current pension scheme parameters, the contribution rate of the savings fund (Jaminan Hari Tua) is relatively low,at a combined rate of 5.7 per cent (table CB3), which can lead to potentially low lump-sum or monthly pay-outs upon retirement. Data from the previous provident fund scheme, Jamsostek, indicates that in 2013, 75 per cent of all the funds were withdrawn before reaching retirement age (Cunha, 2017). The accrual rate of the Jaminan Pensiunscheme is low, at just 1 per cent per qualifying year, resulting in a very low replacement rate of 30 per cent after the minimum contribution period of 30 years (table CB3). Indonesia is an outlier in the region in this regard, and instituting new accrual rates, as applied in many countries, while making sure that contribution rates are adequate to finance such a measure, would ensure more adequate minimum income security in old age.7 Participation in the new defined-benefit scheme (Jaminan Pensiun) is expected to equalize with the provident fund scheme as mandatory participation is gradually phased in, starting with large and medium-sized companies.145The public social protection expenditure for older people was estimated at 0.5 per cent in 2010 (table CB2), a figure that is substantially below the average in Asia and the Pacific, which is 2 per cent. This low level of fiscal investment reflects the system design, which places pension financing under the responsibility of employees and employers, through the provident fund and the new social insurance scheme, and lack of tax-financed pensions for older people without contributory pension entitlements. Meanwhile, the nominally contributions-financed pension schemes covering civil servants, military and police are subsidized from general tax revenues because the aggregate benefit payments have outstripped collected contributions. 4. Policy directions The current contribution rate for the Jaminan Pensiun scheme is 3 per cent, which will not be enough to cover the scheme costs in the future. A comprehensive reform could include an increase in the contribution levels as well as transfer of part of the contribution channelled to the provident fund, which does not provide effective protection in terms of old-age income security.Old-age income security in Indonesia: Policy messages • Improve the flat-rate, tax-based social pensions: universal pensions to increase old-age incomesecurity system coverage. • Reform the parameters in the private sector pension scheme (Jaminan Pensiun) to ensureadequacy of benefits: higher contribution rates. • Reform the parameters in the public sector pension scheme: higher contribution rates to ensurefinancial sustainability.To guarantee a more adequate level of protection, other parametric reforms can be done, such as an increase in the annual accrual rate to ensure that the pension provides at least the minimum levels of protection proposed under ILO Conventions No. 102 and No. 128: 40 per cent and 45 per cent respectively, for a contributor with 30 years of service. In addition, a reform in the current ASLUT or a new flat-rate, tax-funded, universal pension could be a solution to reduce the large number of older people who do not benefit from a pension in Indonesia. With the changes in family structure and the migration of workers from rural to urban areas, older-people protection through a national pension system gains increasing relevance. This solution would implicate an increase in public expenditure in old-age pensions but seems to be the most feasible solution if the goal is to guarantee the protection of those who are and will be uncovered in a status quo scenario.146Table CB3. Pension scheme parameters for IndonesiaSavings fund Social insurance (Jaminan Pensiun)(Jaminan Hari Tua)Contribution years (CY)-Qualifyingconditions Pensionable age561556 increasing to 65 by 2043Basic formulaLump sum1%*CY*ICAMinimum benefit-300,000 Indonesian rupiah (IDR, or US$23)P e n s i o n Maximum benefit formulaIndex career average-IDR3.6 million (US$270)-Average of individual indexed (by CPI) wage over contribution periodBenefit indexation-CPIEmployee2%1%Self-em% of gross ployedat least 2%-earningsEmployer3.7%2%ContributionsGovernment MeasureE a r n i n g s Minimum baseMaximumWage and allow-ances ---Wage and allowances -IDR7.3 million indexed to inflationData source: ILO, 2017a.REFERENCESBPJS Ketenagakerjan. 2017. “BPJS Ketenagakerjan: Program Jaminan Hari Tua (JHT).” Available at: www.bpjsketenagakerjaan.go.id/page/Program-Jaminan-Hari-Tua-(JHT).html[18 Oct. 2018].—. 2017a.“BPJS Ketenagakerjan: Program Jaminan Pensiun.”Available at: www.bpjsketenagakerjaan. go.id/page/Program-Jaminan-Pensiun.html[18 Oct. 2018].Cunha, N.C. 2017. Actuarial study on the new pension scheme, BPJS Ketenagakerjaan, presentation made at Bogor, 7 March 2017. Available at: www.ilo.org/wcmsp5/groups/public/---asia/---robangkok/---ilo-jakarta/documents/presentation/wcms_548382.pdf[18 Oct. 2018].HelpAge International (HAI). 2018. “Pension watch: Country profile on Indonesia”.Available at:www. pension-watch.net/country-fact-file/indonesia/[14 Aug. 2019].International Labour Organization. 2012. Social protection assessment-based national dialogue: Towards a nationally defined social protection floor in Indonesia. Jakarta: ILO.—. 2017c. Actuarial study on the reform of the Indonesian BPJS Ketenagakerjaan (Jakarta). Available at: www.ilo.org/wcmsp5/groups/public/---asia/---ro-bangkok/---ilo-jakarta/documents/ publication/wcms_571382.pdf[18 Oct. 2018].—. 2017a. “Extending social protection coverage in Indonesia: Improving social security inspection”, 147workshop tools. Available at: www.ilo.org/jakarta/whatwedo/eventsandmeetings/WCMS_546542/ lang--en/index.htm[18 Oct. 2018]. —. 2017b. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva). —. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019]. —. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019]. International Monetary Fund (IMF). 2018. World economic outlook update, July 2018 (Washington, DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[14 Aug. 2019]. International Social Security Association (ISSA). 2018. Social security country profiles: Indonesia. Available at: www.issa.int/en/ [18 Oct. 2018]. Tambunam, Tulus T. H.;Purwoko,Bambang. 2002. “Social Protection in Indonesia.” In E. Adam; M. von Hauff; M. John (eds):Social protection in Southeast Asia and East Asia(Singapore, Friedrich-EbertStiftung). Available at: http://library.fes.de/pdf-files/iez/01443003.pdf[18 Oct. 2018]. United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019]. United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[14 Aug. 2019]. University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[18Oct. 2018]. World Bank. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019]. —. n.d. ASPIRE, Pensions database (Washington, DC).148COUNTRY BRIEF: LAO PEOPLE’S DEMOCRATIC REPUBLICElements of the old-age pension systemContributory social insurance• Public sector National Social Security Fund: A pay-as-you-godefined-benefit scheme for civil servants, military and police personnel, with matching contributions from employer (government) and employees.• Private sector Social Security Fund: A contributory defined-benefit scheme for employees of privatesector and state-owned enterprises with 10 or more employees.Voluntary schemesSelf-employed Social Security Fund: A contribution-funded defined-benefit scheme for self-employed persons, with workers contributing at a rate of 5 per cent of declared earnings.Legal framework • 1975: Temporary Party Regulations • 1986: Party Regulations No. 53 and No. 54 • 1991: Constitution, article 26 • 1993: Decree No. 178 on Social Security for Civil Servants • 1999: Employees in Enterprises Decree No. 207, implemented in 2001 • 2008: Ministerial Decree No. 70/PM• 2013: Social Security Law, implemented in 20141. Development of old-age income security policiesRegulations on social security for persons serving the Lao People’s Revolutionary Party, including military and the police and other civil servants, have existed since the country’s independence in 1975. A temporary regulation by the Revolutionary Party, replaced only in 1986, stipulated the payment of benefits for death, war-related invalidity, disability and old age, to be fully funded by the State. In 1986, two new sets of rules, No. 53 and No. 54, were issued. These regulated the extant taxfunded social security benefits and were to be implemented by the Committee for Social Welfare and War Veterans. They also included provisions for other social contingencies: sickness, maternity and loss of child (Sonthany, 2008). In 1991, a new Constitution was adopted, which established the right to assistance in old age.From independence until 1993, old-age pension benefits in the Lao People’s Democratic Republic were funded entirely by the State and only paid to government employees. In 1993, the first contribution-based social security fund for public employees was established, with matching taxfunded contributions from the State as the employer. In 2001, the implementation of a similar, mandatory contribution-based scheme for workers in the formal private sector followed, also requiring both employers and employees to contribute to a pension fund (Sonthany, 2008).149The Social Security Law was passed in 2013 and implemented in the following year. It led to the formation of the National Social Security Organization (NSSO) through a merger of the Social Security Organization and the State Authority for Social Security. The Social Security Fund under management of the NSSO is considered as the second and third pillars of the country’s old-age pension system. The second pension pillar corresponds to the mandatory, contribution-based defined-benefit schemes for private and public employees in the formal economy, and the third pillar refers to the voluntary savings, contribution-based defined-benefit schemes for the selfemployed or, more broadly, for those without formal ties to the labour market. The social security system of the country lacks a first pillar, which often refers to universal or means-tested, tax-funded social pension schemes that are able to address income security in old age, particularly for persons with low contributory capacities.The mandatory matching contributions rate for the Social Security Fund public sector scheme are at 6 per cent of gross monthly insurable wages, while private sector employees and employers each contribute 2.5 per cent of insurable earnings; self-employed workers can contribute 5 per cent of declared earnings. The floor for insurable earnings used in the formula for the contribution and benefit calculation for workers in the private sector is defined as equal to the minimum wage, while the ceiling is defined as five times the minimum wage.Benefit adjustment of pensions, or indexation of pension benefits, is made every year in line with annual changes in average wages of all persons insured under the system (ISSA, 2018). According to the legislation, old-age pension benefits to private sector workers must be adjusted to meet the minimum specified in the Social Security Law and should also not exceed 75 per cent of average wages of the total persons insured in the system (ILO, 2017a).Despite participation in the Social Security Fund being mandated by law for all workers in formal employment, public and private, compliance with the law is low in the private sector, with only 100,000 workers being insured in 2017 (World Bank, 2018b). This corresponds to less than 2 per cent of the total employed and around 10 per cent of total wage employees (ILO, 2018c). Moreover, coverage rate of persons in pensionable age by the private sector Social Security Fund scheme was close to zero, at 0.26 per cent, in 2015. In contrast, the Social Security Fund for public sector covered nearly 5per cent of older persons in the same year (table CB1).2. Socioeconomic backgroundIn the mid-1980s, the Government started working on an economic reform process, known as the New Economic Mechanism, which aimed at gradual decentralization to move towards a market economy. Under this reform, the growth of the private sector was encouraged with the objective of attracting foreign direct investment (Rosser, 2009). Mining of copper and gold, the hydropower industry, coffee and rice agricultural exports, logging activities and growing ecotourism, have been the basis of the steady economic development of the landlocked country in the past decades (figure CB1). The country’s growth has been among the fastest in Asia, with an average in the 2000s and 2010s of around 7 per cent (World Bank, 2018b).In 2011, the classification of the country changed from “low income” to “lower-middle-income” (World Bank, 2011). In 2018 and for the first time, the Lao People’s Democratic Republic fulfilled the required eligibility criteria and is no longer one of the least developed countries in the world, as defined by the United Nations in its country classification based on gross national income, human assets and economic vulnerability indices.150Despite the general improvements in the past decades, the country remains with poor infrastructure, with 8 per cent of the rural population still living in areas without road access and dependent on low-productivity agricultural activities that have, on average, seven times lower productivity levels than non-agriculture sectors (ILO, 2017a). Even though the share of agriculture in GDP declined from 41 per cent to 27 percent during 2008–13, the share of employment in agriculture has not shrunk accordingly and still accounts for 60 per cent of the economy’s working hours (ILO, 2017b), with the number of workers in the sector having increased from 1.7 million in 1991 to 2.7 millionin 2014 (ILO, 2018c).Table CB1. Socioeconomic and demographic data for the Lao People’s Democratic RepublicSocioeconomic parametersYearGDP per capita (current US$) GDP growth rate, real (%) Inflation rate (%) Poverty rate* (%) Gini coefficient Human Development Index (ranking)2457.4 2017 6.9 2017 0.8 201716.7 2015 36.4 2012 139/188 2018Demographic parametersLatest available data** 20202050Population (million) Urban population share (%)7.07.29.235.036.372.8Share of population aged 60 or older (%) Old-age dependency ratio6.716.16.26.715.3Life expectancy at birth (years)65.467.274.2Life expectancy at age 60 (years)16.617.019.2Note: *Poverty defined as the population living below the US$1.90 per day (2011 PPP) threshold. **Population and share of urban population: 2018. Dependency and life expectancy: 2015.Data source: DESA Population Division, 2018; IMF, 2018;UNDP, 2018; World Bank, 2018.Overall, the Lao People’s Democratic Republic’s public administration, currently carrying high public debt and still in the early stages of development of social dialogue mechanisms, has limited capacity for delivering public services, including social protection and labour regulation. Four of every five Laotian workers were in vulnerable employment in 2015, as either own-account or contributing family workers (table CB2), with women being overrepresented in these forms of work. Women are likely to be employed in the informal economy and have lower wages, job and income insecurity as well as little formal protection in the event of occupational hazards, social contingencies or economic or climate-related shocks. The country has been increasingly affected by natural disasters, such as floods, tropical cyclones or draughts. Damages or losses are mainly due to the country’s limited institutional capacity rather than its exposure to such events, which is relatively low when compared with other countries in the region (ILO, 2018c).In terms of human development progress, life expectancy increased by 13 years and average years of schooling doubled from 1991 to 2015 (UNDP, 2016). Gender parity remains a significant151challenge, although progress on many fronts show its in a better position in relation to other ASEAN Member States, namely in the share of female seats in the Government and the female labour force participation rate (UNDP, 2016). Among the ASEAN Member States, it ranks second, only behind the Philippines, in terms of women’s economic participation and opportunity, but lags far behind in terms of girls’ completion rates in secondary school, with rates of child marriage being the highest among ASEAN Member States (WEF, 2017).Figure CB1. Macroeconomic performance of the Lao People’s Democratic Republic and Eastern Asia and the PacificPanel A: GDP growth, by decade average (%)Panel B. Stability of growth, by decadeData source: University of Groningen, 2013; World Bank, 2018.TableCB2. Labour market and pension indicatorsforthe Lao People’s Democratic Republic, latest available yearLabour market indicatorsLabour force participation rate: aged 15–59 (%) Men WomenOwn-account and contributing family workers (% of labour force) Average monthly wage (US$)Year81.3201582.8201579.8201581.020151092010Pension indicators Active pension contributors (% of working-age population:aged 15–64) Old-age pension recipients (% of population older than statutory pension age: 60)Private sector pensions (Social Security Fund) Public sector pensions (Social Security Fund) Public social protection expenditure for older persons (% of GDP)Year1.320105.620100.220155.020150.12010Data source: ILO, 2017c; ILO, 2018a; ILO, 2018b. 152Figure CB2. Lao People’s Democratic Republic’s population pyramid, 2020 and 2050Note: UN population projections.Data source: DESA Population Division, 2018.Income poverty, as defined by the national poverty line, has halved from the early 2000s to 2015. In 2015, the share of population living on less than US$1.90 a day was 16.7 percent, but more than a third of the population had low levels of human development with regards to health, education and living standards (UNDP, 2016). While income inequality improved in the same period in rural areas, with the Gini coefficient increasing from 35.0 in 2007 to 36.4 in 2012 (table CB3), large income discrepancies remain within urban areas (UNDP, 2016) and are likely to grow in face of drastic urbanization trends projected by 2050, as seen in table CB3.3. Old-age pension challengesThe population is the youngest among ASEAN Member States, with children aged younger than 14 years projected to represent 31.7 per cent of the total population in 2020, and the working-age population is projected to be 64 per cent of the total (figure CB2). In 2050, these shares are projected to increase to 20.1 per cent and 69.2 percent of the total population, respectively. The number of persons aged 60 or older will increase slowly in the next decade, followed by more rapid growth thereafter. The percentage of people aged 60 or older is expected to increase 2.4 times, from 6.7 per cent in 2020 to 16.1 per cent in 2050 (table CB3) (DESA Population Division, 2018).Only a small proportion of the current and future senior citizens are covered by the NSSO old-age pension schemes. Active contributors to the pension scheme amounted to 1.3 per cent of the working-age population in 2010. In the same year, the proportion of women and men older than the statutory pensionable age receiving an old-age pension was 5.6 per cent (table CB2).Legal coverage for old age as a percentage of the working-age population was estimated at 9.5 percent. Public social protection expenditure for older persons in 2013 was 0.2 per cent of GDP (ILO, 2017c).The public sector pension scheme is a pay-as-you-go scheme without reserves. It is inherently fragile in terms of financial stability due to limited coverage, low contributions and a relatively high replacement rate. If current replacement and contribution rates are to be sustained, budget subsidies from the general government budget or subsidization from the private sector pension might be required, jeopardizing the sustainability of the whole old-age social security system.1534. Policy directions The biggest challenge for old-age income security in the Lao People’s Democratic Republic is the fact that more than 94 per cent of persons older than 60 do not benefit from a pension. Due to the small number of contributors to the Social Security Fund scheme, this is not expected to change without the adoption of a tax-funded scheme. The most effective way of closing this gap would be the creation of a first pillar, preferably in the form of a universal social pension scheme.Old-age income security in the Lao People’s Democratic Republic: Policy messages • Adopt a flat-rate, tax-based universal old-age pension scheme (social pension). • Reform the parameters in the National Social Security Fund scheme for private sector: retirementages, contribution rates. • Gradually reduce retirement benefits for the National Social Security Fund public sector.In terms of the reforms of the second pillar, actuarial valuation shows that benefits for the private sector scheme might not be financially sustainable in the long run. A pension reform is deemed necessary. While the pension branch is currently still in surplus and accumulating reserves, these reserves will not be sufficient to cover the future cost of pension liabilities. Particularly, the current scheme for the public sector presents a serious risk in terms of its financial sustainability. The deficit projected as of 2018, at the current contribution rate, is 16.5 per cent. In a pay-as-you-go mode, the contribution rate is projected to increase from 16 per cent to more than 40 per cent by 2040. One of the reasons is the extremely generous benefit levels. A parameter reform of the pension branch could include the following measures or a combination thereof to reduce the future cost of old-age income security. In line with the proposals for the private sector fund, the following can be considered for the public sector (ILO, 2017a): Gradual reduction of pensions (public sector). Set retirement age for all at age 60, and gradually increase it further, in line with demographic trends. Increase gradually the contribution rate of the long-term benefits’ branch to increase the pace of accumulation of technical reserves for future pension payments. Undertake a review of benefit provisions for non-work-related disability (public sector).154TableCB3. Pension schemes parameters forthe Lao People’s Democratic RepublicQualifying conditionsPension formulaContribution years Pensionable age Basic formulaPrivate sector defined benefit (National Social Security Fund)Public sector defined benefit (National Social Security Fund)5 (minimum)2560 (men), 55 (women)1.5%*BS*pension pointsPoints earned per year=individual wage per average wage of all in-sured70% to 100%*BS, depending on year of joining civil service or revo-lutionary movement and length of serviceIf less than 30 contribution years, each year missing to 30 is countedas half a pointContributionsMinimum benefitMaximum benefitBase salary (BS)Benefit indexation% of gross earningsEmployees Self-employed EmployersGovernmentEarn- Measure ings base MinimumMaximum75%*BS Average wage of all insured Average wage of all insured5.5% 9% 6% -Official minimum monthly wage Five times the official minimumwageFinal salary Average wage of all insured 8% 8.5%Minimum wage (900000 kip) 4500000 kipData source: ILO, 2017a.REFERENCESInternational Labour Organization (ILO). 2017a. Assessment-based national dialogue on social protection: Lao People’s Democratic Republic(Vientiane).—. 2017b. Decent Work Country Programme: The Lao People’s Democratic Republic 2017–2021(Vientiane). Available at: www.ilo.org/asia/publications/WCMS_630313/lang--en/index. htm [3 Aug. 2018].—. 2017c. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].—. 2018c. “Lao People’s Democratic Republic: Employment and environmental sustainability factsheets 2017”. Available at: www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/ documents/publication/wcms_625980.pdf[3 Aug. 2018].155International Monetary Fund (IMF). 2018. World economic outlook update, July 2018 (Washington, DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[14 Aug. 2019].International Social Security Association (ISSA). 2018. “Social security country profiles: Lao People’s Democratic Republic.” Available at: https://www.issa.int/en/country-details?countryId=LA®ionI d=ASI&filtered=false[accessed 3 Oct. 2018].Rosser, A. 2009.“Lao People’s Democratic Republic”, inIDS Bulletin, vol. 37, pp. 27–39. 10.1111/j.1759436.2006.tb00259.x.Sonthany, P. 2008. Lao People’s Democratic Republic: Social security(Geneva, ILO). Available at: https://www.ilo.org/secsoc/information-resources/publications-and-tools/Workingpapers/WCMS_ SECSOC_6601/lang--en/index.htm[3 Oct. 2018].United Nations. 2018. Country profile: Lao People’s Democratic Republic (New York). Available at: https://www.un.org/development/desa/dpad/wp-content/uploads/sites/45/LDC_Profile_LaoPDR. pdf. [3 Oct. 2018].United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019].United Nations Development Programme (UNDP). 2016. Human development report 2016: Human development for everyone(New York). Available at: www.hdr.undp.org/en/content/humandevelopment-report-2016-human-development-everyone[3 Oct. 2018].—. 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/files/2018_human_development_statistical_update.pdf[14 Aug. 2019].University of Groningen. 2013. “Maddison Project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[14 Aug. 2019].World Economic Forum (WEF). 2017. The global gender gap report 2017 (Geneva). Available at: www3.weforum.org/docs/WEF_GGGR_2017.pdf[3 Oct. 2018].World Bank. 2011. “Lao People’s Democratic Republic now a lower-middle income economy”, press release, 17 Aug. Available at: www.worldbank.org/en/news/press-release/2011/08/17/lao-pdr-nowlower-middle-income-economy [3 Oct. 2018].—. 2018a. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019].—. 2018b. Lao People’s Democratic Republiceconomic monitor, June 2018: Safeguarding stability an ongoing agenda (Washington, DC). Available at: https://openknowledge.worldbank. org/bitstream/handle/10986/29920/127222-WP-P164978-PUBLIC-LEMJuneFinalforweb. pdf?sequence=1&isAllowed=y[3 Oct. 2018].—. n.d. ASPIRE, Pensions database (Washington, DC).156COUNTRY BRIEF: MALAYSIAElements of the old-age pension systemNon-contributory social assistance• Bantuan Orang Tua: A means-tested social pension for persons aged 60 or older with no family or with family incapable of financial support.Non-contributory social insurance• Public Sector Pension Fund: Mandatory social insurance providing old-age, invalidity or deathpensions to civil servants or dependants; it is a partially funded, defined-benefit scheme. A minimum contribution of 17.5 percent of a civil servant’s salary, borne by the employer (government) goes into a pension trust fund (Employees Provident Fund) every month.Contributory social insurance• Employees Provident Fund: Contributions are compulsory for private sector and some public sector employees, voluntary for domestic workers, self-employed persons (see SP1M), and non-citizens. It is a fully funded, double-account defined-contribution scheme, with 70 percent of contributions going into a retirement account that can only be withdrawn when member reaches age 55 or in case of immigration, illness, disability or death, and 30 percent goes into a savings account that can be withdrawn at any time for defined purposes.8• Armed Forces Fund (Lembaga Tabung Angkatan Tentera: A 10 percent contribution rate is compulsory for serving members, with the Government contributing 15 per cent. For other officers, participation is voluntary, and contributions can range from MYR25 to MYR2,000 per month. A lump-sum payment, composed of a worker’s contributions plus dividends, is made in case of death, retirement, discharge from service or from age 50. The Government’s contributions go to a fund and are paid out as monthly pensions (LTAT, 2016).• Voluntary Private Retirement Schemes: Contributions can reach the MYR3,000 ceiling a year from the employee side or 19 percent of the salary on the employer side. Lump-sum payments can be withdrawn in case of invalidity, illness, at age 55, emigration or death (PPA, 2017).• 1Malaysia Retirement Scheme (SP1M). As an arm of the Employees Provident Fund, it covers self-employed persons who do not earn regular income or are not covered in mandatory schemes. Benefits include death and invalidity. Annual contributions can range from MYR50 to MYR60,000. The Government contributes 15 percent of the amount contributed, up to a ceiling of MYR250 a year (KWSP, 2018).8 Assets from the retirement account cannot be withdrawn until the employee is at least 55 years old and can be deferred up to age 70. There are more liberal withdrawal restrictions on the savings account. It is accessible for the purpose of housing, education and other needs such as in the case of health issues. The scheme’s participants can withdraw their entire savings from the second account at the age of 50. See http://databank.worldbank.org/data/ views/reports/reportwidget.aspx?Report_Name=CountryProfile&Id=b450fd57&tbar=y&dd=y&inf=n&zm=n&country=MYS [accessed 6 July 2018].157Other provisions:• Zakat: The Islamic community provides means-tested old-age benefits for Muslim citizens.• Social Security Organization (SOCSO) schemes: Employment Injury Insurance and Invalidity Pension Scheme.Legal framework• 1951: Employee Provident Fund Ordinance• 1969: Employees’ Social Security Act• 1973: Tabung Angkatan Tentera (Armed Forces Fund) Act (Act 101)• 1980: Statutory and Local Authorities Pension Act• 1991: Employees Provident Fund Act• 2012: Minimum Retirement Age Act1. Development of old-age income security policiesSocial protection programmes for older people in Malaysia date back to the 1800s, with the establishment of the pension system in 1871 and the introduction of the Penang Municipal Provident Fund. Other pension funds were subsequently established, mainly for workers in plantations and mining industries. These were combined in 1951 into the general Employees Provident Fund and later amended into the 1991 Employees Provident Fund Act.Malaysia’s second old-age pension pillar, the Employees Provident Fund, is the oldest provident fund in the world. In 2017, active contributors represented 47 per cent of the labour force (EPF, 2018). Their personal accounts are credited every year with a dividend rate subject to the Fund’s investment return rate, which reached 7.3 per cent in 2017. There is a guaranteed minimum return rate of 2.5 percent, and since 2013, the real dividend rate has averaged at 6.4 percent (EPF, 2018).Malaysia’s first pension pillar currently consists of an old-age social assistance programme under the administration of Malaysia’s Department of Social Welfare, which covers persons aged 60 or older who have no other income sources or no family financial support. Exclusion errors in the targeting of beneficiaries remain high as a consequence of inaccurate means-testing, which jeopardizes the receipt of pensions by the target group most in need. The pension level in 2017 amounted to 300 Malaysian ringgit (MYR) (JKM Malaysia, 2018), which corresponded to less than a third of the minimum wage in that year.In 2017, the mandated pension schemes for civil servants and those serving in the armed forces covered around 11 percent of the labour force (ILO, 2018b).9 Coverage of the Public Sector Pension Fund increased from 2006 to 2014 and is expected to continue rising, with 140,000 covered expected to retire by 2020 (LTAT, 2016).9 Share corresponds to total employed in the International Standard Industrial Classification (Rev 4, group O) corresponding to “Public administration and defence” in Malaysia in 2017 divided by total labour force in 2017.158In 2017, more than 2.6 million people employed in Malaysia were self-employed, which amounted to 18 per cent of total employment. At the same time, the private retirement schemes, which were introduced in 2012 to increase pension coverage among workers outside the mandatory schemes, had only 250,000 active members (PPA, 2017). Even though take-up in these schemes has been increasing, with the Government providing tax incentives and matching contributions for the first four years, the coverage of self-employed workers remains low.The old-age contributions of the Islamic community forzakat, a charitable mechanism, are sizeable. In 2013, they were equivalent to 20 percent of total social spending in the country, excluding health care (OECD, 2016b).In 2012, total social protection expenditure, including health care amounted to 3.8 per cent of GDP, with 0.9 per cent of GDP spent on social protection for the older population, excluding health care (table CB3).2. Socioeconomic and demographic contextMalaysia is an upper-middle income economy that has recorded decades of steady growth of 5 per cent or more for more than 30 years, surpassing the average growth by decade in Eastern Asia and the Pacific since the 1970s (figure CB1). This, coupled with reasonable political stability, allowed its GDP to rise from US$11.5 billion in 1960 to US$330 billion in 2016 (table CB2). A natural-resource rich country, Malaysia now has an open, diversified and industrialized economy consisting of services (at 55 per cent of GDP), agriculture (at 8.5 per cent of GDP) and manufacturing (at 22.8 per cent of GDP) (World Bank, 2014).Table CB1. Socioeconomic and demographic data for MalaysiaSocioeconomicsGDP per capita (US$) GDP growth rate, real (%) Poverty rate* (%) Gini coefficient Human Development Index (ranking)9813 4.2 0.646.3 57/188Year2017 2016 2016 2009 2017DemographicsLatest available year**20202050Population (million) Share of urban population (%)32.032.941.776.077.287.3Share of population aged 60 or older (%) Old-age dependency ratio Life expectancy at birth (years)Life expectancy at age 60 (years)1010.723.18.510.124.474.775.580.919.820.023.8Note: *Poverty defined as the population living below the US$1.90 2011 PPP per day threshold. **Population, share of urban population, share of the older population: 2018. Dependency and life expectancy: 2015.Data source: DESA Population Division,2018;IMF, 2018; UNDP, 2018; World Bank, 2018.159Although Malaysia used to be one of the most unequal countries in South-Eastern Asia, contrary to the regional trend it has been moving towards a more equal distribution of income. Over the period 2002–04, the Gini index was sustained at around 0.46, down to 0.44 in 2007–09, then dropped again to 0.43 in 2012 and 0.401 in 2014, the lowest level ever for the country.Figure CB1. Macroeconomic performance of Malaysia and Eastern Asia and the PacificPanel A: GDP growth, by decade average (%)Panel B. Stability of growth, by decadeData source:University of Groningen, 2013; World Bank, 2018.Table CB2. Labour market and pension indicatorsfor Malaysia, latest available yearLabour market indicatorsLabour force participation rate (%) Male FemaleOwn-account and contributing family workers (% of labour force)Informal employment rate (%) Average monthly wage (US$) Median monthly wage (US$)Year68.0201780.1201754.7201721.8201711.4201571320175342017Pension indicatorsActive pension contributors (% of working-age population:aged 15–64) Male FemalePension recipients (% of population older than statutory pension age: 55)Year28.1201032.4201023.6201019.82010Public social protection expenditure for older persons (% of GDP)0.92012Benefit amount, social assistance pension (% of minimum wages)31.32016Data source:HAI, 2018; ILO, 2017; ILO, 2018a; ILO, 2018b.160Figure CB2. Malaysia’s population pyramid, 2020 and 2050Note: UN population projections.Data source: DESA Population Division, 2018.In 2018, Malaysia’s population was more than 32 million, with an estimated share of old-age population of 10 per cent. Total population is expected to increase from 31 million in 2015 to just below 42 million in 2050. The percentage of the population aged 60 or older is expected to increase from 9 per cent to almost 23 per cent over this period. The female share of the population aged 60 or older will only change from 53 per cent to 56 per cent.The total fertility rate steadily declined from 4.9children per woman in 1970 to 2.4 children in 2005 and below replacement rates in 2014 to 2 children (Karunanethy, forthcoming).Malaysia is projected to experience an almost three-fold increase in its old-age dependency ratio, from 7.4 in 2010 to 21.7 in 2040.As in many countries in the region, another factor posting challenges for Malaysia’s older population is the increasing urbanization, creating a vacuum in traditional, intra-family, old-age support. While only one of two Malaysians resided in urban areas in 1991, the share of urban inhabitants increased to nearly three out of four persons in 2015 and is expected to reach 80 per cent by 2020 (Karunanethy, forthcoming).Due to sustained economic growth and rapid urbanization, the vast majority of the employed,at 89.6 per cent,is in formal employment (table CB2). The total labour force participation rate was 68 per cent in 2017; 80.1 per cent for men and 54.7 for women. With this rate, women’s participation in the labour market is relatively low, at leastwhen compared with Thailand and Cambodia (Karunanethy, forthcoming).3. Old-age pension challengesMalaysia’s social security system suffers from fragmentations of different kinds, the most important of which are across economic sectors and across the political authorizing environment (World Bank, 2014).Due to the lack of a social security net, workforce participation is essential to build up retirement savings within the Employees’ Provident Fund. As in many other countries in the region, the low labour force participation rate of women leaves them particularly vulnerable to old-age poverty.161Around 37 per cent of the labour force is not covered by the fragmented Malaysian pension system. The main group without coverage is self-employed persons, who could contribute on a voluntary basis, but only about 1 per cent actually does so. Unpaid family workers are also largely uncovered, as well as labour migrants who are excluded from joining the Employees’ Provident Fund (World Bank, 2014).If one would make a conversion of the average amounts that members of the Employees’ Provident Fund have in their accounts at the moment of retirement, Malaysia would have one of the lowest gross replacement rates for both men and women (at 35.1 per cent and 31.9 per cent, respectively, in 2012) in the region (World Bank, 2014).The old-age support for private sector employees needs to improve in terms of the relatively low level of benefits and lack of financial awareness. The fact that the system allows several withdrawals before reaching retirement age significantly reduces the amount that members have in their account at the time of retirement. In addition and in the absence of regulation concerning compulsory annuitization, a large majority of the scheme members opts for a lump-sum payment.A study by Bank Negara Malaysia (BNM, 2015) found that only 6 per cent of employees were able to sustain themselves for more than six months after losing their job. At retirement, about 71 per cent of Employees’ Provident Fund members opt for the lump-sum withdrawal, with half of them exhausting their savings within five years.The rising share of government employees in the labour force will soon run the risk of affecting pension expenditure and fiscal sustainability (World Bank, 2014). Since 2007, the Public Sector Pension Fund net income is not enough to cover the pension expenditure, and the deficit has been consistently growing. It is expected to reachMYR32 billion by 2030 (Karunanethy, forthcoming).4. Policy directionsOld-age income security in Malaysia: Policy message• Improve provident fund (Employees’Provident Fund) regulation and general trust in the pension scheme (such as through compulsory annuitization of retirement funds).• Assess the feasibility of integrating a defined-benefit component under the Employees’Provident Fund scheme.• Increase old-age pension coverage with the adoption of a flat-rate, tax-funded, universal scheme (social pension).Because Malaysia is developing a Social Protection Blueprint, some options to improve in the system should be discussed:• The launch of a tax-funded pension to ensure that every older person in Malaysia benefits from a pension.• Converting the funds available in the Employees’ Provident Fund’s individual accounts into periodical payments.• Following the direction of Indonesia and assessing the feasibility of adding a social insurance component to the system to ensure at least a minimum level of replacement.• Ensuring that the tax-funded schemes and the contributory schemes are jointly assessed and designed to ensure the complementarity between the different pillars.162Table CB3. Pension schemes parameters for Malaysia(including sickness and maternity bene- Provident fund fits)Public sector pensionArmed forces schemeQualifying conditionsBenefitsContributionsMinimum withdrawal age60 (formerly 55)60 (formerly 58)Years of contributionBasic formulaMinimum benefit Maximum benefit Base salary Benefit indexationEmployee1021(1/600 X months of service X last drawn salary(excluding allowances)), but Lump sum capped at 360 months and 3/5th of last drawn salaryMYR950 per month for those who serveda minimum of 25 years2% increment per year 11%10%Self-employedMYR50 to MYR5000% of grossEmployerearningsGovernmentEarnings baseMeasure MinimumMaximum13% (salary MYR5000) 10% matching contributions up to MYR120 per year for self-employed and domesticworkersMYR10MYR300015% 17.5Sources: EPF, 2018;Karunanethy, forthcoming.REFERENCESAllianz International Pensions. 2013. What’s happening in Malaysia: Getting prepared for retirement?(Kuala Lumpur). Available at: https://projectm-online.com/app/uploads/what-ishappening-in-malaysia.pdf[6 July 2018].Armed Forces Fund Board (LTAT). 2016.Annual report 2016 (Kuala Lumpur). Available at:www.ltat. org.my/borg/LTATAnnualReport2016e.pdf[8 Aug. 2018].Bank Negara Malaysia (BNM). 2015. Financial capability and inclusion study 2015(unpublished) (KualaLumpur).Employees Provident Fund (EPF). 2018. 1Malaysia retirement savings scheme (Kuala Lumpur).163Available at: www.kwsp.gov.my/portal/en/general/1malaysia-retirement-savings-scheme[6 July 2018].HelpAge International (HAI). 2018. “Pension watch: Country profile on Cambodia”.Available at:www. pension-watch.net/country-fact-file/cambodia/[14 Aug. 2019].International Labour Organization. 2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].International Monetary Fund (IMF). 2018. World economic outlook update, July 2018.Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlook-updatejuly-2018[14 Aug. 2019].JKM Malaysia. 2018. “JKM Malaysia”. Available at: www.jkm.gov.my/jkm/index.php?r=portal/ left&id=VEpUUXV3THFURkZETmxWNjZpQ1BXdz09 [6 July 2018].Karunanethy, K. forthcoming. A review of the provident fund approach to providing old-age income security in four Asia-Pacific countries (Bangkok, ESCAP).Lee, A.H.; Khalid, M.A. 2016. “Is inequality in Malaysia really going down? A puzzle explored”,DECRG Kuala Lumpur Seminar Series. Available at: www.worldbank.org/en/events/2016/10/04/isinequality-in-malaysia-really-going-down[6 July 2018].=Organisation for Economic Co-operation and Development (OECD).2016a. Economic surveys: Malaysia economic assessment(Paris). Available at: www.oecd-ilibrary.org/economics/oecdeconomic-surveys-malaysia-2016_eco_surveys-mys-2016-en[6 July 2018].—. 2016b. Live long and prosper – Ageing in East Asia and Pacific(Bangkok).Private Pension Administrator Malaysia (PPA). 2017. Private retirement schemes hits quarter-millionmember milestone. Available at:www.ppa.my/wp-content/uploads/2018/03/Private-RetirementSchemes-Hits-Quarter-Million-Member-Milestone-press-release-31-July-2017.pdf[6 July 2018].United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019].United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[14 Aug. 2019].University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[14 Aug. 2019].World Bank. 2014. Old-age financial protection in Malaysia: Challenges and options, Social Protection Discussion Paper No. 1425 (Washington, DC).—. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019].—. n.d. ASPIRE, Pensions database (Washington, DC). 164COUNTRY BRIEF: MYANMARElements of the old-age pension systemNon-contributory social assistance• Universal Social Pension Scheme:A tax-financed pension for citizens aged 85 and older.Non-contributory social insurance• Old-age pensions for public servants:A non-contributory, unfunded, defined-benefit pension scheme for civil servants, military and political personnel and employees of state economic enterprises.Contributory social insurance• Social Security Board pension:A contributory pension scheme mandatory for private sector employees and voluntary for informal workers and farmers, stipulated in the 2012 Social Security Law but still to be implemented.Legal framework• 2012: Social Security Law (started implementation in 2014 according to Social Security Rules of 2012)• 2013: Civil Service Personal Laws, 2013 and Rules1. Development of old-age income security policiesMyanmar’s Social Security Boardwascreated in 1956,after adoption of the Social Security Act of 1954. Following decades of isolation and military rule, a new Social Security Law was passed in 2012. The Social Security Board is now responsible for implementing the provisions of the 2012 law, which provides for compulsory social security coverage for private sector companies with more than five employees and voluntary coverage for companies with fewer than five workers as well as students, self-employed persons and farmers. The law started implementation in 2014 in stages. In the first stage, and thus far, only the existing benefits from the previous scheme were implemented and increased.Insured members of the Social Security Board enjoy access to medical care, a funeral grant, sickness cash benefits, maternity and paternity cash benefits and work injury benefits. Provisions for the invalidity benefit, superannuation benefit and survivors’ benefit insurance system, unemployment benefit insurance system and other social security system (including housing), although stipulated in the law, have not been implemented yet. The Social Security Board is carrying out four reforms: IT reform, medical reform, administrative reform and legal reform, after which, the Invalidity, Superannuation and Survivor Benefits are expected to be gradually implemented.Civil servants, military and political personnel and employees of state economic enterprises enjoy coverage under an old-age pension scheme. The scheme is a non-contributory, unfunded, definedbenefit scheme financed by the state budget. A pension equal to 50 per cent of the final salary of a civil servant who has worked for 35 years is provided. It is not automatically indexed to inflation, but the Government has increased it periodically to preserve the purchasing power. In FY2017/18, there were 812,966 pensioners and total government expenditure of 0.77 per cent of GDP, or 3.58 per cent of165government expenditures (Pension Department, 2019). Because the scheme is entirely financed by the state budget, concerns about its future sustainability with the ageing process setting in have driven the Government to consider both parameter and systemic reform options (World Bank, 2016b).To deal with these challenges, Notification Order 27, issued by the President Office in February 2019, established a Central Provident Fund Implementation Committee, led by the Deputy Minister for Planning and Finance, to support public as well as private sector employees in providing housing, healthcare and monthly retirement pensions. According to this plan, a Central Provident Fund Law will be drawn up, instead of the initial plan of developing a pension scheme under the Social Security Board following the established under the Social Security Law of 2012.The Pension Department is undertaking both a systemic reform and a parameter reform. The systemic reform aims to change the current defined-benefit pension system into a funded, definedcontribution system to reduce the financial pressure for the Government. This is planned to be achieved through the establishment of a Central Provident Fund covering public sector employees. According to the plan, private sector employees are also expected to be covered in the long run. This is expected to reduce budget deficits and the high pension liabilities of the Government.The new pension system is expected to run in parallel with the existing pension system, at least in the beginning.The Social Security Board scheme, which does not implement the old-age provisions yet, had 1.1 million members as of December 2017, equivalent to 5 per cent of the labour force (Frontier, 2017). Considering that workers in vulnerable employment make up 57.2 per cent of the labour force (selfemployed at 33.3 per cent and contributing family workers at 23.9 per cent),10 achieving meaningful social security coverage will require designing appropriate mechanisms to reach such workers. The Social Security Law allows for progressive extension of social security provisions to reach informal workers, including farmers, by reaching out to small enterprises (smaller than the current threshold of five workers). Although the law allows them to enrol voluntarily, only the compulsory registration was opened in 2014, and so far, voluntary registration has not started for any of the benefits.Myanmar spent about 0.7 per cent of its GDP on social protection and other services in FY2014/15, the lowest in the ASEAN region. Most of this went to civil service pensions (Myanmar Times, 2015; World Bank, 2016b). Since 2012 though, the Government has prioritized social protection in its development agenda. The National Social Protection Strategic Plan was adopted in 2014, which advocates extending social protection to all, especially the vulnerable, based on a life-cycle approach. Two out of eight flagship social protection programmes in the Plan commenced implementation in 2017: the Maternal and Child Cash Transfer in certain states11 and the social pension.The social pension was piloted from June 2015 to May 2016 to test implementation arrangements and challenges. This was extended nationwide in June 2017. The Universal National Social Pension Scheme is entirely funded by the State and implemented by the Ministry of Social Welfare, Relief and Resettlement. A monthly cash transfer of 10,000 Myanmar kyat (MYK), which is equal to 6.8 per cent of GDP per capita and 61 per cent of the US$1.90 poverty line at PPP, is provided (HAI, 2018).The scheme categorically targets people aged 85 and older and has 168,578 beneficiaries, which indicates near universal coverage in the intended age group. It plans to cover 202,293 beneficiaries in FY2019/20 (U Kyaw Linn Htin, 2019). Although the absolute coverage is low, it speaks of the willingness10 In addition to those in vulnerable employment, many wage workers do not have legal contracts, paid leave and social security registration, effectively making them informal workers. In total, wage workers comprise 40.4 per cent of the labour force.11 Chin State, Rakhine State, Kayah State, Shan State, Ayarwaddy Region and Naga self-administered zone166of the Government to expand coverage to everyone when fiscal space increases and administrative capacity is improved. The categorical nature of the scheme is a step in the right direction because it ensures universality of protection and a rights-based approach to protection in old age.2. Socioeconomic backgroundMyanmar has experienced rapid economic growth in the past few years, at more than 6 per cent annually. Still, poverty and vulnerability are high, with two-thirds of people living below or just above the poverty line. More than half the population is engaged in agricultural activities and is affected by frequent natural disasters (ILO, 2015). In the absence of large-scale redistributive policies, the rapid economic growth is accompanied by rising inequalities (table CB1). Nevertheless, reforms are being rolled out to address issues facing the labour market, including laying down the basis for a comprehensive social protection system.Although still a relatively young country, the share of older people in Myanmar is increasing gradually but significantly. In 2015, only 8.9 per cent of the population was aged 60 or older, a proportion that is projected to double to 18.5 per cent by 2050 (DESA Population Division, 2018). With little access to pensions at present, many older people live with their children and receive care and financial support from them (HAI, 2013). Almost half of the people living in rural areas cope with vulnerabilities through intrinsic social structures or organizations, which have existed for a long time and are built on social and religious beliefs and structures (Griffiths, 2016).Older people in Myanmar typically live in low-income households. In 2014, more than 60 per cent reported a monthly household income of less than US$3 per day. Less than 20 per cent of older persons reported accumulating savings over their working lives. Widespread illiteracy places this population group at considerable disadvantage and exacerbates their dependence on younger family members. And 59 per cent of the older population cited their children as the main source of material support, while 24 per cent cited work (Knodel, 2014). In the absence of old-age income security and adequate healthcare, many households are vulnerable to falling into poverty due to health-related shocks. Out-of-pocket health expenditure as a percentage of current health expenditure was as high as 73.9 per cent in 2015 (World Bank, 2018).Table CB1. Socioeconomic and demographic data for MyanmarSocioeconomic parametersYearGDP per capita (US$) GDP growth, real (%) Inflation (%) Poverty rate (%)* Gini coefficient Human Development Index (ranking)1148 6.3 7.023.6 38.1 1482015 2016 2016 2011 2015 2017Demographic parametersLatest available year**20202050Total population (million)53.954.862.4Share of urban population*** (%)30.631.147.1Share of population aged 60 or older (%)9.710.318.5Old-age dependency ratio8.09.419.2Life expectancy at birth (years)66.066.870.5Life expectancy at age 60 (years)16.716.917.9Note: *Poverty headcount ratio at US$1.90 a day in 2011 PPP.**Population, share of urban population, share of the older population: 2018. Dependency and life expectancy: 2015.***Annual percentage of population at mid-year residing in urban areas.Data source: DESA Population Division 2018; IMF, 2018; World Bank, 2018; UNDP, 2018.167Figure CB1. Macroeconomic performance of Myanmar and Eastern Asia and the PacificPanel A. GDP growth, by decade averagePanel B. Stability of growth by decade Source: University of Groningen, 2013, World Bank, 2018.Table CB2. Labour market and pension indicatorsfor Myanmar, latest available yearLabour market indicatorsYearLabour force participation rate: aged 15 and older (%) Men WomenOwn-account and contributing family workers (% of labour force) Average monthly wage (US$) Pension indicators61.2201877.3201847.7201857.220171092015YearActive pension contributors (% of working-age population) Private sector scheme (SSB) Public sector schemeOld-age pension recipients (% of population older than 60) Public sector scheme Social pensionPublic social protection expenditure for older persons (% of GDP) Social pensionBenefit level, social pension % of GDP per capita)0.020180.020180.0201819.3201816.120183.220180.7 FY2014/150.120176.82017Sources: HAI, 2018; ILO, 2018a; ILO, 2018b; ILO, 2017. 168Figure CB2. Myanmar’s population pyramid, 2015 and 2050Source: DESA Population Division, 2018. A major challenge observed in other ASEAN Member States and in countries worldwide is the difficulty of enrolling informal workers in social security schemes, even when such schemes are in place. Myanmar has a high level of informal employment, estimated at 85.7 per cent of total employment – one of the highest among the ASEAN Member States (ILO, 2018c). This is composed of about 35 per cent wage employees without proper legal contracts and labour protection, effectively making them informal workers, 25 per cent own-account workers, 23 per cent unpaid family workers and about 2 per cent unregistered employers (ILO, 2018c; ILO, 2016). People in informal employment often receive low and irregular incomes, a fact that has a crucial role in how they access social security. Myanmar has a little more than 2 million immigrants, mainly working or living in Thailand, Malaysia, China and Singapore (DOP and UNFPA, 2017). Informal employment is high among migrant workers, including international migrants, internal rural-to-urban migrants and internal rural-to-rural migrants. While some may use official channels to migrate, such as through employment agencies, most people migrate irregularly due to the high financial costs, time costs and the administrative and documentation requirements. As a result, they end up working in jobs without adequate labour protection or social security, which can impact their future old-age income security. 3. Old-age pension challenges The challenges in Myanmar’s old-age pension system can be visualized in terms of coverage, adequacy and financial sustainability. Myanmar has lower coverage than other countries at a similar income level and spends less than other countries at the same stage of demographic transition. Around 16.1 per cent of the population aged 60 or older is covered by the public servant pension scheme and 0.6 per cent by the social pension. There are no pensioners in the private sector. The adequacy of pensions paid out by the public servant scheme is low, mainly due to low wages. The pensions do not cover the cost of living of pensioners, especially when retired public servants have to give up their free housing at the time of retirement and may not have saved enough to afford their own home. The average monthly pension in FY2013/14 was about MYK37,000, or US$37 (World Bank, 2016b, Myanmar Times, 2014). Because the scheme is entirely financed by the state budget, there are concerns about its future sustainability as the ageing process sets in (World Bank, 2016b). 4. Policy directions Myanmar is one of the few ASEAN Member States with no pension coverage for private sector workers. The design of a sustainable pension system composed of integrated contributory169and non-contributory schemes, progressively reaching the entire older population, can be an important measure in bringing the social protection system close to regional benchmarks (Clay et al., forthcoming). The system should aim to maintain financial sustainability while also increasing coverage and adequacy of benefits.Integration of the Social Security Board and public servant pension schemes could promote equity, easy movement between the public and private sectors for workers and administrative economies of scale. A contributory, defined-benefit scheme based on international standards and best practices could serve as the main mechanism to provide old-age income security to workers and could be implemented without jeopardizing the fiscal position (Clay et al., forthcoming). For those without access to the contributory schemes, the Universal Social Pension could be expanded from the current eligibility age of 85 to 65.Old-age income security in Myanmar: Policy messages• Shift from tax-financing to a contributory defined-benefit scheme for the public sector.• Put in place integrated contributory and non-contributory pension schemes for the formal sector.• Gradually expand the coverage of the Universal Social Pension to people aged 65 and older.A dialogue regarding the design of a pension system is taking place in Myanmar. One of the proposals is the establishment of a contributory defined-contribution pension scheme, such as acentral provident fund for the public and private sectors. This would be financed by contributions of 5–10 per cent of employees’ wages. The fund could be withdrawn for two purposes: (i) retirement and (ii) pre-retirement withdrawals, such as education, health care and housing (World Bank, 2016b). While a defined-contribution scheme or provident fund could provide additional security, thereby supplementing the primary defined-benefit pension scheme, they alone are not enough to guarantee income security in old age due to low benefit levels and insufficiently developed financial markets for investment in Myanmar (Clay et al., forthcoming).International experiences show that becausedefined-contribution schemes do not pool contributions, they maintain income inequality in countries where the gap between the rich and poor is significant. In a country with the current demographic and economic situation of Myanmar, the ILO should advise the Government to carefully assess all the potential impacts of creating a mandatory provident fund for workers in the private sector.The Government is also considering a parameter reform of the defined-benefit scheme for public servants, such as the extension of the reference wage period from the final year to the final three years or more. Other reform measures that could be considered are price indexation of pensions, and increase in the retirement age due to increasing life expectancy. Furthermore, the compensation pension, which serves as a termination benefit rather than old-age income security, should be separated from the pension system as far as financing is concerned and could be changed from a periodic benefit to a lump-sum payment (World Bank, 2016b).Workers in vulnerable employment make up 57.2 per cent of the labour force. They constitute selfemployed and contributing family workers who have irregular and interrupted participation in the labour market and can often experience periods of unemployment. Among wage workers, many170do not have access to social security. Such workers are usually not covered by contributory pension schemes (both defined benefit and defined contribution) in other countries and without access to pensions in old age, they are forced to work even in case of ill health or limited capacity to work. The best way to provide old-age income security to such workers is to ensure that the contributory system is complemented by a flat-rate, tax-funded pension.A preliminary policy note by Clay et al.(forthcoming) proposes an alternative option to ensure oldage income security for everyone, which combines a contributory, earnings-related defined-benefit scheme with a flat-rate, tax-funded pension. This combination should provide adequate income security to everyone in old age by meeting the minimum replacement rate of 45 per cent stipulated in Convention No. 128.Table CB3. Pension scheme parameters for MyanmarQualifying conditionsContribution years (CY) Pensionable agePension formulaBasic formulaMinimum benefit Maximum benefit Base salary (BS)Benefit indexationContributions % of gross earningsEmployee Self-employed Employer GovernmentPublic sector pensionSocial Security Board pension1060 (50 years with 25 years of service or 55 years with 20 years of service for earlyretirement)Pension: 1.5%*CY*BSLump sum: 50%*CY*BS15 6015 times the insured’s average monthly wage in the total contribution period is paid as a monthly benefit or a lump sum1.--Last salaryIt is not indexed but the Government has increased it periodically tocompensate for loss of purchasing powerover time.Full costSource: Clay et al., forthcoming; World Bank, 2016b.REFERENCESClay, T.; Idris, N.; Cunha, N.; Knox-Vydmanov, C. Forthcoming. A social protection floor for older people in Myanmar(ILO).Department of Population (DOP) of the Ministry of Labour, Immigration and Population of Myanmar; United Nations Population Fund (UNFPA). 2017. Overview of the results of the 2014 Population and Housing Census Myanmar (Nay Pyi Taw). Available at: http://themimu.info/sites/themimu.info/files/ documents/Report_Overview_Results_of_the_2014_Census_DOP_Dec2017.pdf[14 Feb. 2019].171Frontier. 2017. “Workers welcome changes to social security system”, 10 Nov. Available at: https:// frontiermyanmar.net/en/workers-welcome-changes-to-social-security-system[14 Feb. 2019].Griffiths, M.P. 2016. Resilience and community social organizations in rural Myanmar (Yangon). Available at: www.lift-fund.org[14 Feb. 2019].HelpAge International (HAI). 2013. The situation of older persons in Myanmar: Results from the 2012 survey of older persons(Yangon).—. 2017. “Myanmar’s national forum on universal social pension held in Pyidaungsu Hluttaw”. Available at: http://ageingasia.org/myanmars-national-forum-on-universal-social-pension-held-inpyidaungsu-hluttaw/[14 Feb. 2019].—. 2018. “Pension watch: Country profile on Myanmar”.Available at:www.pension-watch.net/ country-fact-file/myanmar[14 Feb. 2019].International Labour Organization. 2015. Myanmar labour force, child labour and school to work transition survey 2015: Executive summary report (Yangon). Available at: www.ilo.org/wcmsp5/ groups/public/---asia/---ro-bangkok/---ilo-yangon/documents/publication/wcms_516117.pdf[14 Feb. 2019].—. 2016. Report on Myanmar labour force survey 2015 (Yangon). Available at: www.mol.gov.mm/ en/wp-content/uploads/downloads/2017/03/LFS-English-Report-_17-11-2016.pdf[14 Feb. 2019].—. 2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].—. 2018c. Informality at a glance and snapshot by country: ASEAN and other countries from the Asia and the Pacific region (Geneva).International Monetary Fund (IMF). 2018. World economic outlook update, July 2018 (Washington, DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[14 Aug. 2019].Knodel, J. 2014. “The situation of older persons in Myanmar – An overview”, inBOLD, vol. 24, No. 3, pp. 9–16.Pwint, Z.P. 2014. “For civil servants, retirement brings problems, not peace”, in Myanmar Times, 23 June. Available at: www.mmtimes.com/index.php/in-depth/10794-for-civil-servants-retirementbrings-problems-not-peace.html[14 Feb. 2019].Nishino, Y.; Koehler, G. 2011. Social protection in Myanmar: Making the case for holistic policy reform. IDS working paper, no. 386 (London, Institute of Development Studies). Available at: http:// onlinelibrary.wiley.com/doi/10.1111/j.2040-0209.2011.00386_2.x/pdf[14 Feb. 2019].Ramkissoon, S. 2015. Technical options to implement a universal social pension in Myanmar (Kent, Development Pathways).Thant, H.2015. “Budget concerns cripple pension expansion”, in Myanmar Times, 15 Jan. Available172at:www.mmtimes.com/index.php/national-news/12805-budget-concerns-cripple-pension-expansion.html[14 Feb. 2019].U Kyaw Linn Htin. 2019. Implementation status of cash transfer programs and idea for electronic cash transfer (Nay Pyi Taw).United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019].United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[14 Aug. 2019].UNICEF. 2013. “Myanmar statistics”.Available at: www.unicef.org/infobycountry/myanmar_statistics. html[14 Feb. 2019].University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[14 Aug. 2019].World Health Organization (WHO). 2015. “Myanmar: WHO statistical profile”, January 2015.Available at: www.who.int/gho/countries/mmr.pdf[14 Feb. 2019].World Bank. 2016a. Live long and prosper – Ageing in East Asia and Pacific(Bangkok).—. 2016b. Building resilience, equity and opportunity in Myanmar: The role of social protection:Strengthening social security provision in Myanmar, Structural policy country notes Myanmar, Note 7(Washington, DC).—. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019].—. n.d. ASPIRE, Pensions database (Washington, DC).173COUNTRY BRIEF: PHILIPPINESElements of the old-age pension system Non-contributory social assistance • Social Pension for Indigent Senior Citizens:A tax-financed pension paid to Filipino citizensaged 60 or older who are “frail, sickly, or with disability” and without pension, income or any other support. Non-contributory social insurance • Pensions for special groups of public servants: Separate schemes for armed forces, judiciary and constitutional commissions, police, coast guard, fire protection personnel, etc. Contributory social insurance • Social Security System pension:A contributory, partiallyfunded, defined-benefit scheme with compulsory coverage of private sector workers and self-employed workers earning at least PHP1,000, and voluntary coverage of specific categories of persons. • Government Service Insurance Systempension:A contributory, partiallyfunded, definedbenefit scheme with compulsory coverage of public sector employees, except the special groups. • Pag-IBIG Fund (Home Development Mutual Fund.An individual savings account with compulsory coverage of private sector workers and self-employed workers earning at least PHP1,000, and voluntary coverage of specific categories of persons. Members receive all contributions and dividend earned at maturity. Voluntary schemes Occupational, private pension schemes:Various defined contribution schemes, such as the taxexempt Personal Equity and Retirement Account, that provide the option of supplementary savings, mainly for middle-income formal and self-employed workers.Legal framework • 1997: Social Security Act (Republic Act No. 1161), 1954, amended by Republic Act No. 8282 • 1997: Government Service Insurance Act (Republic Act No. 8291), amended by PresidentialDecree No. 1146 • 2010: Expanded Senior Citizens Act (Republic Act No. 9994)1741. Development of old-age income security policiesThe two major pension systems in the country, the Government Service Insurance System (GSIS) and the Social Security System (SSS), were established in the mid-twentieth century. Together, they comprise the second pension pillar. The GSIS for the public sector is not only the first social insurance scheme to be set up in the Philippines but also the oldest in the region (COSE and HAI, 2017). It was created in 1936 and started operations in 1937. The GSIS provides death, disability, work injury and separation benefits as well as emergency and housing loans. It is financed by a 21 per cent contribution from employees (9 per cent) and employers (12 per cent).Two decades later, the Social Security System for private sector employees and self-employed workers was created by Republic Act 1161 of 1954. The scheme started operations in 1957. It is a defined-benefit scheme that provides a monthly pension according to a formula (table CB3). Pensions are indexed at discretion, depending on inflation and the fund status (World Bank, 2016b).The SSS scheme is partially funded, with a gross contribution rate of 11 per cent (7.37 percent paid by employers and 3.63 per cent for employees). Contributions are calculated on an insurable earnings base of at least 1,000 Filipino pesos (PHP) and a maximum of PHP16,000. The contributions finance maternity, sickness, disability, survivor and death benefits, in addition to old-age pensions. The contribution rate may increase in the future to accommodate the new unemployment insurance and expanded maternity benefits. Self-employed persons shoulder the entire 11 per cent contribution. Despite mandatory coverage of self-employed persons, few of them are covered in practice (COSE and HAI, 2017).To encourage participation of low-income workers in informal employment in the SSS, the AlkanSSSya12saving mechanism was introduced. It allows participants to deposit just PHP11 (about US$0.20) per day, as and when they can, into designated “piggy banks” or saving boxes. The contribution amount of PHP11 per day or PHP330 per month is calculated based on a monthly salary credit of PHP3,000 (11 per cent of PHP3,000). And 40 per cent of the monthly salary credit of PHP3,000 corresponds to the minimum pension benefit of PHP1,200 (table CB3). Workers can declare a higher monthly salary credit and pay higher contributions to receive higher benefits. The sustainability of AlkanSSSya rests on the premise that (i) workers in informal employment have a continuing means of livelihood; (ii) an organization, such as the informal sector groups, handles the administrative tasks of counting the total contributions and filling the payment forms; and (iii) community support from informal sector groups to instil a culture of saving among workers (DAP, 2018).While public sector employees are covered by the GSIS, there are separate retirement schemes for members of the Judiciary and Constitutional Commissions, the Armed Forces, the Philippine National Police, the Bureau of Jail Management and Penology, the Bureau of Fire Protection and others. The Armed Forces of the Philippines’ Retirement and Separation Benefits System is a defined-benefit scheme with a lump-sum payment upon retirement and monthly pensions indexed on current wages. Being entirely funded by general taxation, the various schemes, such asthe Armed Forces one, have high costs and are thought to be unsustainable. A proposal to finance them through contributions and shift the management to a new fund under the GSIS is being studied (Cu, 2018; PIA, 2018).12 “AlkanSSSya” is a play on the word “alkansya”, which means piggy bank. In other words, AlkanSSSya is a piggybank for SSS.175One component of the third pension pillar is the mandatory PAG-IBIG Fund (or the Home Development Mutual Fund). The fund was set up in 1978. Primarily intended to provide housing support, it also serves as a provident fund, wherein worker and employer contributions, together with the dividend earned can be withdrawn either at the maturity date or upon retirement, death, migration, job loss or disability. The main use of the fund is for housing loans, followed by personal loans, which may leave little for retirement savings.In 2010, the Expanded Senior Citizens Act was passed, and in 2011, the Social Pension for Indigent Senior Citizens was implemented by the Department of Social Welfare and Development. This constitutes the first pension pillar. The Social Pension equals PHP500 per month, paid as a quarterly instalment.To be eligible, senior citizens are identified using the National Household Targeting System for Poverty Reduction, also known as Listahanan(World Bank, 2016b). They must be “frail, sickly or with disability” and not receive other pensions, income or financial support. Local government units apply their own eligibility and identification mechanisms as well, which may make the beneficiary selection process varying slightly across regions (ILO, forthcoming). The pension was provided to those aged 77 and older from 2011 to 2014. In January 2015, the eligibility age was lowered from 77 to 65 and in 2016, it was further lowered to 60 years, in compliance with Republic Act 9994 (COSE and HAI, 2017).2. Socioeconomic backgroundIn the 1950s and 1960s, the Philippines was one of Asia’s industrial powerhouses and a major exporter of consumer goods. The gains made during these times were lost to economic mismanagement and corruption in the Government in the 1970s and 1980s, when the country encountered negative growth rates (figure CB1). It was overtaken by neighbouring ASEAN Member States, such as Malaysia, Singapore and Thailand, which all grew at 4–8 per cent. This also contributed to almost a tenth of the rapidly-growing population, including some of the most talented citizens, to take up work in foreign countries.In more recent decades, the Philippine economy has been relatively resilient to global economic shocks due to low exposure to troubled international securities, lower dependence on exports, relatively resilient domestic consumption, large remittances from overseas Filipino workers and a rapidly expanding outsourcing industry (CIA, 2017).Nevertheless, economic growth needs to be more inclusive. The Gini coefficient for the Philippines was 40.1 in 2015, one of the highest in the region. In recent times, the Government prioritized the regularization of contractual workers and disguised employment relationships, which, if implemented effectively, could improve access of workers in informal employment to social security and pensions (Panganiban, 2018).176Table CB1. Socioeconomic and demographic data forthe PhilippinesSocioeconomic parameters GDP per capita (US$) GDP growth, real (%) Inflation (%) Poverty rate, (%)* Gini coefficient Human Development Index (ranking)Demographic parametersTotal population (million) Share of urban population*** (%)Share of population aged 60 or older (%) Old-age dependency ratio Life expectancy at birth (years)Life expectancy at age 60 (years)Latest year**available 2020 106.5 46.97.87.2 68.617.2Year2989 6.7 4.021.6 40.1 1132017 2017 2018 2015 2015 20182050109.7 47.4151.3 61.88.214.28.114.869.373.217.319.1Note: *Poverty headcount ratio at national poverty line. **Population, share of urban population, share of older people: 2018. Dependency and life expectancy: 2015. ***Annual percentage of population at mid-year residing in urban areas.Data source: DESA Population Division 2018; IMF, 2018; World Bank, 2018; UNDP, 2018.Unemployment in the country is high in regional comparison, standing at 5.3 per cent in 2018. At 16.4 per cent of the workforce in 2018, underemployment is another pressing concern, especially among self-employed persons and rural and agricultural workers (PSA, 2018b). These are workers who expressed intention to work more but were unable to find work. An estimated 36.6 per cent of the population is in vulnerable employment, such as self-employment, including low-paid work and unpaid family work(ILO, forthcoming; PSA, 2018b).177Figure CB1. Macroeconomic performance of the Philippines and Eastern Asia and the PacificPanel A. GDP growth, by decade averagePanel B. Stability of growth, by decadeSource: University of Groningen, 2013; World Bank, 2018.Table CB2. Labour market and pension indicatorsforthe Philippines, latest available yearLabour market indicators Labour force participation rate: aged 15 or older (%)Men WomenOwn-account and contributing family workers (% of labour force) Average monthly wage (US$) Pension indicators Active pension contributors (% of working-age population)Private sector defined-benefit scheme (SSS) Public sector defined-benefit scheme (GSIS) Old-age pension recipients (% of population older than 60) Contributory schemes Social Pension Public social protection expenditure for older persons (% of GDP) Social Pension Benefit level, Social Pension (% of GDP per capita)Year 60.9 77.0 57.2 36.6279 Year23.0 21.02.1 64 29 35 0.6 0.142018 2017 2017 201720122018 2018 2018 2018 2018 2018 2012 2018 2017Sources: ILO, 2018a; ILO, 2018b; ILO, 2017; PSA, 2018b. Figure CB2. Philippines’ population pyramid, 2015 and 2050Source: DESA Population Division, 2018. 178The Philippines’ population is estimated at around 107 million, making it the world’s twelfth most populous country. It is also one of the youngest in the region, with slow growth in old-age dependency rates. While the country’s birth rate remains high in regional and global comparison, fertility rates declined over the past 60 years, from 7.4 children per woman in 1950–55 to 3 children in 2010–15. The old-age dependency ratio is projected to double, from 7.2 per cent in 2015 to 14.8 per cent in 2050, as the share of older persons in the total population simultaneously changes from 7.8 per cent to 14.8 per cent. Urbanization is also increasing. At the beginning of the twentieth century, more than 80 per cent of the population lived in rural areas, but that proportion now stands at about 50 per cent (table CB3).Labour immigration to North America, the Middle East and South-Eastern Asia is significant (World Bank, 2016b). Estimates of the number of overseas workers vary between 10.2 million in 2013 (CFO, n.d.) and 2.3 million in 2017 (PSA, 2018a). Although migrants can voluntarily enrol for SSS pensions, few of them actually do and will be left without pension security in old age when they are not covered by pension systems in other countries. In 2015, 21.6 per cent of individuals in the Philippines lived below the national poverty line, indicating the large number of people facing difficulty in contributing to social security (World Bank, 2016b).3. Old-age pension challengesWhile the scope of programmes in the Philippines is comprehensive, effective coverage of workers remains low due to persistent poverty and labour informality. Of the total labour force of 44.1 million, about 14.2 million are actively paying members of SSS and 1.4 million of GSIS. Furthermore, low public expenditures on old-age security contributes to low pension coverage among olderpeople. About 1.9 million older people receive pensions from the SSS and 400,000 people draw their pension from the GSIS, thereby putting 29 per cent of the older population under contributory pension coverage (ILO, forthcoming;HAI, 2018). Another 2.8 million persons receive the non-contributory social pension, equivalent to 35 per cent of the older population. This means that 36 per cent of older people have no access to pensions (HAI, 2018). Most of them were informal or vulnerably employed workers.Although the Philippines’ contributory pension system has relatively higher coverage than other ASEAN Member States at similar economic status, it is still inadequate, with 65 per cent of the labour force not participating in the SSS or the GSIS (COSE and HAI, 2017). Most SSS and GSIS pensioners fall into the higher income brackets, meaning that low-income workers who have little savings and are most in need of pensions may not receive one. The percentage of contributors remained fairly steady between 2000 and 2016 (COSE and HAI, 2017).Non-standard employment in the Philippines is quite substantial. There is a sizeable number of contractual workers hired by public sector organizations who do not generally have access to GSIS or SSS benefits. Efforts have been made to enrol them as self-employed workers in the SSS, but their effective coverage is uncertain. Of an estimated 600,000 such contractual workers, only 200,000 have been enrolled in the SSS.Pension benefits, especially those received by SSS pensioners, are perceived as low, partly due to a wage ceiling of PHP16,000 per month for paying contributions (ILO, forthcoming). The wage ceiling has not kept up with rising inflation over time, leading to people in higher income spectra not being able to contribute towards higher benefits even if they wanted to. The pension benefit is not systematically indexed on inflation, but indexed on a discretionary basis, depending on observed inflation and the fund status (World Bank, 2016b). This has contributed to reducing the value of the179benefits over time. The Philippines has a gross replacement rate of about 38 per cent, just below the ILO target of 40 per cent and below the OECD average of about 54 per cent (World Bank, 2016b).Receipt of a contributory pension does not necessarily guarantee old-age income security. The average monthly SSS pension was PHP3,600 in September 2015, and 20 per cent received between PHP1,200 and PHP1,499, which is below the poverty line of PHP1,800 (COSE and HAI, 2017). Between 2000 and 2010, the real value of SSS pensions declined by 24 per cent, while in comparison, GSIS pensions increased by 3 per cent and the Armed Forces pension increased by 38 per cent. The SSS pensions are particularly low for women, possibly due to low-wage jobs performed by women. In September 2015, 44 per cent of women received pensions below PHP2,000 compared with 29 per cent of men (COSE and HAI, 2017).Although it is crucial to increase coverage of contributory pensions, persistent informality and poverty pose major challenges. According to official statistics estimate that 36.6 per cent of workers are vulnerably employed (as self-employed and unpaid family workers) (ILO, forthcoming). Other sources cite higher figures, of 70–75 per cent as informal or irregular workers (COSE and HAI, 2017).A large number of people work without proper contracts or social security in formal sector enterprises, effectively classifying them as informally employed. Workers in informal employment are more likely to have lower incomes and usually have irregular and unpredictable incomes. In the absence of employers, the entire 11 per cent of contributions must be shouldered by them.With their low incomes and thus low contributions, pension entitlements are also meagre. All these factors make contributing to social security a low priority for workers, with daily expenses, food and education expenditures usually taking precedence. Even if the level of contributions were to increase, it would take several years to translate into higher pensions. The only way to address this gap is through a universal tax-funded pension.The SSS faces the challenge of maintaining adequate benefits against a backdrop of limited contribution histories. A simulation by the World Bank (2016b) found that, at status quo, the SSS scheme will have greater benefit pay-outs than revenues by 2036, and it will face asset depletion by 2044. Because the old-age dependency ratio will more than double in the next 30 years, the benefit pay-outs will exceed the revenues by more than 1 per cent of GDP by about 2060.Another reason for the deterioration in fund sustainability is a benefit formula and qualifying conditions that over time, cannot be fully supported by the present contribution levels. The mediumterm financing issues, however, can be addressed through parameter reforms (World Bank, 2016b).The coverage of the Social Pension has increased rapidly, from 139,000 at inception to 2.8 million people in 2017. However, the scheme suffers from inclusion and exclusion errors in the targeting process. A review done by COSE and HAI (2017) found that many older recipients of the pension were not among the poorest, while many of those who were clearly poor were identified as nonpoor by the Listahanan database. As a result, Listahanan is now used as a starting reference point for local government units to select pension recipients. This may create grounds for favouritism and corruption.Furthermore, the benefit amount of PHP500 has not been adjusted with inflation since 2011, leading to a decrease in purchasing power of beneficiaries. It currently stands at a mere fraction of the poverty line of PHP1,800. The pension amount, as a share of per capita GDP, is one of the lowest worldwide – higher than Thailand and lower than Viet Nam and Brunei Darussalam. 1804. Policy directions At present, the pension system of the Philippines does not cover 36 per cent of older people or the “missing middle”, mainly consisting of informal workers probably living just above the poverty line. It is improbable that subsidizing their contributions under the SSS will lead to increased oldage pension coverage because many scheme members have been found to drop in and out of the scheme on an irregular basis and even paying 50 per cent of the contributions is unaffordable for many. Instead, it is recommended to expand coverage of the Social Pension, at least to vulnerable and near-poor older people and, over time, as a universal pension. This can tackle the inclusion and exclusion errors in targeting and bring about administrative cost-efficiency (COSE and HAI, 2017; World Bank 2016b).Old-age income security in Philippines: Policy messages • Improve the adequacy of benefits for the SSS and the social pension. • Expand the contributory base, particularly to non-standard and contractual work (withemployers paying their share of contributions). • Increase public investment in a sustainable pension system by establishing an integrated,multitiered system; consisting of a flat-rate, tax-funded pension for everyone (expansion of the current social pension) and a larger, contribution-financed pension based on member earnings (current SSS pension). • Conduct a parametric reform of the pension system.To build an equitable and sustainable pension system, it is necessary to increase public investment in the system. A more integrated system combining contributory and tax-funded elements can be considered. This can take the form of a multitiered system with a flat-rate, tax-funded pension for everyone (an expansion of the current Social Pension) and a larger, contribution-financed pension benefit based on member earnings (the current SSS pension). A recent World Bank study (2016b) proposed parameter reforms to the SSS pension scheme to improve its financial sustainability. The reform options include automatic indexing of pensions to inflation, harmonizing the benefit formulae, increasing the minimum retirement ageand increasing the contribution rate. The level of the social pension should also be increased for better adequacy. To increase contributory pension coverage and reach a sustainable pension system, it is necessary to improve the administrative and delivery mechanisms of the SSS and to take a more systemic view of labour market reforms towards creating better and decent jobs, particularly in the formal economy (World Bank, 2016b).181Table CB3. Pension scheme parameters for the PhilippinesSSSGSISSocial pensionQualifying conditionsPension formulaContribution years (CY)Pensionable ageBasic formulaSenior citizens who are frail, sickly, or with disability; donot receive any contributory pensions; do not have permanent income; or regularsupport from relatives1015None60–65 (mandatory 60–65 (mandatory age) unlessage); 55 for minersnot yet vestedHighest of three options:2.5%* (PHP700 + average monthly compensation) PHP300 monthly + 20% of average monthly salary credit (AMSC) for each year of service + 2% for each year of service exceeding 10 years 40% of AMSCTwo options: lump sum equal to basic monthly pension (BMP) * 18, plus BMP for life lump sum equal to BMP * 60 plus BMP for life after 5years PHP1200 monthly60 PHP500 per monthMinimum benefitPHP1200 per month (contribution period 10-20 years),PHP2400 (contribution period > 20 years)Maximum benefitSalary baseBenefit indexationContribution ratesEarnings baseEmployees Employers MinimumMaximumNoneHighest salary in last 5 years or averageof last 6 months, not revalued90% of average monthly compensation in last 3 yearsLast 3 years average salary, not revaluedAd hoc, as approved by SS Commission(60% of inflation over past 20 years)Ad hoc, as recommended by Actuary and approved byBoard3.63% 7.37% PHP1000 per month9% 12% Minimum wagePHP16000 per monthNoneAccording to periodic revision by the CongressNon-contributorySource: ILO forthcoming; World Bank, 2016b.182REFERENCESAquino, E.B. 2002. Private pension schemes in the Philippines: Regulatory practices (Geneva, Organization for Economic Co-operation and Development). Available at: www.oecd.org/finance/ private-pensions/2763673.pdf [12 Feb. 2019].Central Intelligence Agency (CIA). 2017, “World factbook: Philippines”. Available at: www.cia.gov/ library/publications/the-world-factbook/geos/rp.html [12 Feb. 2019].Coalition of Services of the Elderly (COSE) and HelpAge International (HAI). 2017. The feasibility of a universal social pension in the Philippines (Manila).Commission on Filipinos Overseas (CFO).n.d.“Stockestimationofoverseas Filipinos”. Available at: www.cfo.gov.ph/program-and-services/yearly-stock-estimation-of-overseas-filipinos.html [12 Feb. 2019].Cu, R. 2018. “Draft of pension system plan for AFP creates new unit to be run by GSIS” in Business Mirror, 2 Nov. Available at: https://businessmirror.com.ph/2018/11/02/draft-of-pension-systemplan-for-afp-creates-new-unit-to-be-run-by-gsis/ [12 Feb. 2019].Development Academy of the Philippines (DAP). 2018. AlkanSSSya program: Moving towards inclusive growth (Manila).Available at: www.dap.edu.ph/coe-psp/innov_initiatives/alkansssyaprogram-moving-towards-inclusive-growth/ [12 Feb. 2019].Gonzalez, E. 2007. “Social protection in the Philippines”, in Missing targets: An alternative MDG midterm report,Social Watch Philippines Report 2007 (Manila). Available at: http://www.socialwatch. org/node/12453 [12 Feb. 2019].Gonzalez, E.; Manasan, R. 2002. “Social protection in the Philippines”, in E. Adam; M. von Huff; M. John (eds):Social protection in Southeast and East Asia (Singapore, Friedrich-Ebert-Stiftung), pp. 180–229.HelpAge International (HAI). 2016. The Philippine social pension at four years: Insights and recommendations(Manila).Available at: http://www.helpage.org/download/56e195e41b33a/ [12 Feb. 2019].—.2018. “Pension watch: Country profile on Philippines”.Available at: www.pension-watch.net/ country-fact-file/philippines/ [12 Feb. 2019].International Labour Organization. 2014. Social protection for older persons: key policy trends and statistics(Geneva).—. 2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].—.Forthcoming. Assessment-based national dialogue in the Philippines: Establishing a social protection floor for all (Bangkok).International Monetary Fund (IMF). 2018. World economic outlook update, July 2018(Washington,183DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018 [14 Aug. 2019]. Organisation for Economic Co-operation and Development (OECD). 2015. Pensions at a glance (Paris). Panganiban, A. 2018. “Legal and illegal labor contracting”,inInquirer.net., 6 May. Available at: https:// opinion.inquirer.net/112965/legal-illegal-labor-contracting [12 Feb. 2019]. Philippine Information Agency (PIA). 2018.“GSIS to administer pension of newly-installed uniformed personnel”, 5 Sep. Available at: https://pia.gov.ph/news/articles/1012319 [12 Feb. 2019]. Philippine Statistics Authority (PSA). 2016. Current labor statistics January 2016 (Manila). —.2018a. 2017 survey on overseas Filipinos (Manila).Available at: https://psa.gov.ph/content/2017survey-overseas-filipinos-results-2017-survey-overseas-filipinos [12 Feb. 2019]. —.2018b. 2018 Annual labor and employment status (Manila).Available at: https://psa.gov.ph/ content/2018-annual-labor-and-employment-status [12 Feb. 2019]. Racelis, R.; Abrigo, M.R.; Salas, J.M. 2012. Filipino elderly living arrangements, work activity, and labor income as old-age support, Philippine Institute for Development Studies, Discussion Paper Series No. 2012-31 (Manila). United Nations, Department of Economic and Social Affairs (DESA), Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019]. United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[14 Aug. 2019]. University of Groningen. 2013. “Maddison Project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013 [14 Aug. 2019]. World Bank. 2016a. Live long and prosper – Ageing in East Asia and Pacific (Washington, DC). —.2016b. Review of the social security system: Considerations for strengthening sustainability and coverage (Washington, DC). Available at: https://openknowledge.worldbank.org/ handle/10986/24671 [12Feb. 2019]. —. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019]. —. n.d. ASPIRE, Pensions database (Washington, DC). World Health Organization (WHO). 2012. Health service delivery profile: Philippines (Manila, WHO and Department of Health, Government of the Philippines).184COUNTRY BRIEF: SINGAPOREElements of the old-age pension system Non-contributory social assistance • Silver Support Scheme: A means-tested retirement income supplement targeted atSingaporeans aged 65 years or older who had low incomes through life and now have little or no family support. Non-contributory social insurance • Pensions Scheme: A non-contributory superannuation scheme for civil servants that has largely been discontinued since 1986. • Savings and Employees Scheme: Defence budget-funded scheme for military officers and warrant officers. Contributory social insurance • Central Provident Fund: A fullyfunded individual savings account that is mandatory for employees, includes voluntary retirement savings for self-employed workers, and open only to citizens and permanent residents. Voluntary schemes • Private defined-contribution savings schemes:Available to expatriate and migrant workers who are not eligible for Central Provident Fund, as well as Singaporean workers who wish to buy additional insurance. • Supplementary Retirement Scheme: A voluntary, savings scheme with tax benefits, as a complement to the Central Provident Fund. • Housing Monetization Schemes (Lease Buyback Scheme for persons aged 65 and older, Silver Housing Bonus for persons aged 55 and older.It facilitates unlocking the value of HDB flats to supplement retirement income. Both schemes require seniors to top-up part of their proceeds to their Central Provident Fund Retirement Account, which will be streamed out as monthly pay-outs in their retirement years.Legal framework • 1956: Pensions Act (Ordinance 22), revised 2004 • 1978: Parliamentary Pensions Act (Act 24), revised 1996 • 2012: Parliamentary Pensions (Abolition) Act (Act 21) • 2013: Central Provident Fund Act (Chapter 36), Revised Edition • 2015: Silver Support Scheme Act, No. 281851. Development of old-age income security policiesThe Singaporean old-age income security system is one of the oldest in Asia. Its central element and the main vehicle of social security is the Central Provident Fund. The Fund aims to ensure financial security in old age by providing lifelong income and addressing housing and healthcare needs. It is a defined-contribution scheme, to which employees and employers make monthly contributions. Although the Central Provident Fund encourages personal responsibility and ownership for one’s retirement, other schemes such as the CPF LIFEemploys risk pooling and the Workfare Income Supplement redistributes income to an extent. It is complemented by government subsidies in various areas, including housing and healthcare, top-ups and other tax-funded social transfers targeted at persons in need of support.13 The Central Provident Fund is open only to Singaporean citizens and permanent residents.The Central Provident Fund was established in 1955 under British colonial rule to allow workers to save for their retirement. In 1968, the Public Housing Scheme was set up, which allowed workers to use their Central Provident Fund savings to pay for housing. As Singapore developed into a modern and advanced economy by the 1970s, wages rose and contribution rates were increased. Workers were also allowed to invest part of their savings to get higher returns. In 1984, a MediSave account was added so that workers could save for healthcare expenses in old age. In 1990, the MediShield healthcare insurance scheme was introduced to protect Singaporeans against high medical bills. MediShield was expanded in 2015 to MediShield Life to cover all Singaporeans, including those with pre-existing conditions. The Government provides subsidies for MediShield Life premiums to keep them affordable. ElderShield was introduced in 2002 as a basic long-term care insurance targeted at severe disability, especially during old age.ElderShield will be enhanced to CareShield Life in 2020 and will cover all Singaporeans born 1980 or later. Singaporeans born in 1979 or earlier can choose to join the scheme. Similar to MediShield Life, the Government will provide subsidies to Singaporeans for CareShield Life premiums.The Central Provident Fund provides three types of individual accounts for each member: an Ordinary Account to finance a home purchase, approved investments, term insurance and approved education courses; a Special Account for retirement (the member may invest a portion of the accumulating monies in retirement-related financial products); and a MediSave account for financing pre-defined medical costs and premiums for insurance schemes. At age 55, a member’s savings in the Ordinary Account and Special Account, up to a pre-defined retirement sum, specific to the member’s birth cohort, are transferred to a retirement account.Monies in the Ordinary Account and the Special Account, above the retirement sum, can be withdrawn in a lump sum from age 55. Monies in the Retirement Account are used to provide for monthly retirement pay-outs from a member’s pay-out eligibility age, which is currently 65. The accumulated contributions in the Ordinary Account earn an annual compound interest of at least 2.5 per cent, while those in the Special Account, the MediSave Account and the Retirement Account earn at least 4 per cent. An extra 1 per cent interest is provided on the first 60,000 Singaporean dollars (SGD) of Central Provident Fund savings for all members. Members aged 55 orolder earn an additional 1 per cent interest i.e. total of 2 per cent extra on the first SGD30,000.13 F or example,the Silver Support Scheme, U-Save rebates, GST Vouchers and Community Health Assistance Scheme,which are complemented by safety nets, such as MediFund and ComCare, provide further assistance to the neediest people.186In the initial decades of the Central Provident Fund, members could withdraw their entire savings upon retirement. Rising life expectancy,14 however, meant that retirees had more years in retirement, which would require more savings. There was high risk of people outliving their savings. As a result, the Minimum Sum Scheme15 was introduced in 1987 to allow retirees to receive monthly pension pay-outs from their Central Provident Fund savings rather than a lump-sum withdrawal. The Minimum Sum, an amount to be set aside for basic retirement needs in the Retirement Account, directly translated to a monthly pension for 20 years. In the 2000s, the Minimum Sum was gradually increased to meet at least the expenses of a member of a lower-middle income retiree household.With life expectancy rising further, the Minimum Sum Scheme was no longer sufficient. Thus, the Central Provident Fund LIFE annuity scheme was started in 2009. The scheme differs from the Minimum Sum Scheme in that it pays a monthly pension not just for 20 years but until death, with a bequest (a lump-sum grant comprising the remaining retirement savings) to survivors. Participation in CPFLIFE is automatic for younger members (aged 61 or younger in 2019) having more than a threshold Retirement Account balance. Opt-out is only possible on medical grounds or for people receiving other pensions. Reforms raising the minimum retirement savings amount and the introduction of the LIFE annuity scheme have helped to improve old-age income security for workers.To supplement the incomes of lower-wage workers and help them save for retirement, the Workfare Income Supplement Scheme was introduced in 2007. It provides Central Provident Fund top-ups and cash supplements to workers aged 35 or older with little household support, and with more going to older workers. For example, a 60-year-old worker earning SGD1,200 a month would be receiving Workfare Income Supplement benefits that amount to almost an extra 30 per cent of the worker’s monthly income. In 2017, 411,000 workers received SGD649 million of Workfare Income Supplementpay-outs.The Pioneer Generation and Merdeka Generation Packages were introduced in 2014 and 2019, respectively. These packages provide additional healthcare support for eligible Singaporeans born before 1960, including subsidies for outpatient medical expenses and MediShield Life insurance premiums, top-ups to MediSave accounts and additional support to protect older Singaporeans against the risk of severe disability.The Silver Support Scheme, introduced in 2016, provides a quarterly cash pay-out of SGD300– SGD750 to supplement the retirement income of Singaporeans aged 65 or older, who had low incomes through life and have little or no family support. To be eligible, a person should have a monthly household income (per person) of not more than SGD1,100, have low lifetime wages16 and satisfy certain housing type criteria. Older people living in smaller flats receive higher pay-outs. In 2018, SGD335 million of Silver Support pay-outs was disbursed to 153,000 older Singaporeans.In regard to civil servants, since 1972, new officers appointed to Division III and IV grades (administration, clerical, technical positions) were superannuated under the Central Provident Fund. On 1 April 1986, new officers in Division I and II grades (managerial, executive positions) were appointed on the full Central Provident Fundscheme.17 Virtually all civil servants are appointed on the full Central Provident Fundscheme today.14 To illustrate, the 2018 (preliminary) life expectancy at age 60 was 25.3 years while the statutory minimum retirement age was 62 and the re-employment age was 67.15 Later renamed as the Retirement Sum Scheme. 16 This is defined as total Central Provident Fund contributions of not more than SGD70,000 by age 55, or an averageannual net trade income of not more than SGD22,800 between the ages of 45 and 54 for self-employed persons. 17 There were two conversion exercises, in 1973 and 1986, for civil servants on the Pensions Scheme to convert to theCentral Provident Fund Scheme. Officers who opted to convert to the Central Provident Fund Scheme would have their accrued pension benefits frozen and payable on their retirement.187The Savings and Employee Retirement (SAVER) Plan and Premium Plan were established in 1998 and 2000 respectively in replacement of pension schemes. Fixed contributions are made monthly to military officers’ and warrant officers’ accounts to provide for a lump sum of money when they retire from the armed forces.2. Socioeconomic backgroundSingapore developed from its lower-middle income status to a high-income country in a relatively short period of about four decades, at the end of which, it emerged as a successful free market economy. The city-state pursued a growth strategy that focused on being an attractive business location despite regional and global environment changes. The economy relies mostly on exports, notably of consumer electronics, information technology, medical and optical devices, pharmaceuticals and financial services.According to the International Monetary Fund (2018), Singapore had the fourth-highest GDP per capita in the world on PPP terms in 2018, at 100,345 international dollars. Singapore’s growth slowed in previous years, reflecting the post-global financial crisis, but went up again in 2010 and stood at 3.1 per cent in 2018.With a Gini coefficient of 40.4 in 2018, inequality in Singapore was among the lowest in a decade. Unemployment is low. Labour market participation by women is lower than men but still higher than the global female labour force participation rate (ILO, 2018b).Table CB1. Socioeconomic and demographic data for SingaporeSocioeconomic parametersGDP per capita (US$) GDP growth, real (%) Inflation (%) Poverty rate (%) Gini coefficient* Human Development Index (ranking)64567 3.1 0.4 n.a.40.4 9Year 2018 2018 20182018 2018Demographic parametersLatest available year**20202050Total population (million) Share of urban population*** (%) Share of population aged 60 or older (%) Old-age dependency ratio Life expectancy at birth (years) Life expectancy at age 60 (years)5.65.96.610010010020.522.140.119.221.560.683.283.587.525.325.528.9Note: DESA Population Division projections have been used for the total population to maintain consistency across countries and to allow for representation of data until 2050. According to Singapore Department of Statistics, resident population stands at 4 million (2018) and is expected to increase to 4.1 million (2020) and 4.5 million (2040).*Data for 2018 are based on household income after accounting for government transfers and taxes. Data are preliminary.\**Population, share of urban population, share of the older population, dependency ratio, life expectancy (preliminary. 2018 (refers to resident population: Singapore citizens and permanent residents).***Singapore being a city-state, its entire population is urban.Data source: email correspondence with Singapore Department of Statistics, 2018; DESA Population Division, 2018; IMF, 2018; World Bank, 2018; UNDP, 2018.188Figure CB1. Macroeconomic performance of Singapore and Eastern Asia and the PacificPanel A. GDP growth, by decade averagePanel B. Stability of growth by decadeSource: email correspondence with Singapore Department of Statistics, 2018; University of Groningen, 2013, World Bank, 2018.Table CB2. Labour market and pension indicatorsfor Singapore, latest available yearLabour market indicators Labour force participation rate: aged 15 and older (%)Men Women Own-account and contributing family workers (% of labour force) Average monthly wage (US$) Pension indicators Active pension contributors (% of working-age population) Central Provident FundYear 67.7 75.6 60.29.3 2,87976.4 76.42018 2018 2018 2018 2015 Year 2018 2018Old-age pension recipients (% of population older than 65) Central Provident Fund SSSPublic social protection expenditure for older persons (% of GDP) Benefit level, social pension (% of GDP per capita)48.9 27.90.7 1.6-4.12018 2018 2011 2018Source: Comprehensive Labour Force Survey, Manpower Research & Statistics Department, MOM. Figure CB2. Singapore’s population pyramid, 2018 and 2050Source: Singapore Department of Statistics (2018 Resident Population); DESA Population Division (2050 Total Population Projections).189The estimated median age in Singapore will increase from 40.8 in 2018 to around 46.8 years by 2040. The resident total fertility rate has been gradually declining over the past decades and stood at 1.14 children per woman in 2018. Life expectancy at the age of 60 is estimated to rise from 25.3 (in 2018) to around 27.4 years by 2040 (DOS, 2018b; DOS, 2018c;DESA Population Division, 2018). These dynamics make Singapore one of the fastest ageing Asian countries.Singapore has high immigration rates, with an average net migration rate of 14.96 per 10,000 inhabitants between 2010 and 2015 (DESA Population Division, 2018). The ratio of the non-citizen population increased from 14 per cent in 1990 to 26 per cent in 2000 and further to 38 per cent in 2018 (DOS, 2018b).3. Old-age pension challengesThe demographic changes in Singapore and rapid ageing are not expected to strain the fiscal resources of the country but are likely to put additional pressure on the family as a traditional source of old-age income. Family is still regarded as the main source of income in old age in the form of transfers from and co-residence with children, ranking ahead of pensions, investment income and work (Allianz, 2014).Despite high contributions paid during their working lives, a section of current retirees has only modest incomes from the Central Provident Fund, partially due to the fact that they prioritize purchasing public housing as the single biggest use of savings rather than old-age income security (Ng, 2015). The Central Provident Fund savings were also used to meet approved education costs. In addition to these design features, not all current seniors worked regularly or had middle or higher incomes at the time of their contributions. As a result, more than 60 per cent of active Central Provident Fund members who turned 55 in 2018 were able to set aside the retirement sum for meeting basic needs in old age.Furthermore, Ng (2015) found that while poverty among older people not working had fallen over the years, it was the contrary for older people in the workforce. The relative poverty rate18 among working older people increased from 13 per cent in 1995 to 28 per cent in 2005 and to 41 per cent in 2011. Some older people prefer to work into their silver years rather than take assistance from their children or state assistance for the poor and vulnerable. This may point to principal deficiencies in having the Central Provident Fund as the main old-age income security system.Nonetheless, the Government has been putting in place several policy measures to adapt the scheme and ensure better old-age income security, including complementing the Central Provident Fund with several other means-tested, income-support schemes, as previously described.Studies aiming to estimate replacement rates produced a range of results (Ng, 2015). The OECD estimated, on the basis of 2011 data and legislation, that an average earner with stable income joining the Central Provident Fund in 2008 at the age of 20 years and retiring after a full career, would gain a gross pension of 11–13 per cent of lifetime earnings; in 2012 and 2013, it published net-replacement rates of between 24 per cent and 42 per cent of individual earnings for men, and of between 21 per cent and 38 per cent of individual earnings for women (OECD, 2013; OECD, 2012). The higher rates (for men and women alike) refer to model cases of scheme members with incomes that are 50 per centor 100 per cent of average earnings, the lower rates for those earning 200 per cent of the average.18 Defined as people with net incomes below 40 per cent of the median work income.190In comparison with other countries, Singapore’s income replacement rate may appear low. However, this low rate does not account for the fact that most older Singaporeans own their home or do not have to pay rent. This high rate of home ownership among older Singaporeans gives them the potential to monetize their homes for greater income in retirement if they so desire. Hence, an expenditure replacement benchmark may be a better gauge of what retirees may need.Although the Singaporean pension system mitigates absolute rather than relative poverty, a shift in the policy approach may be needed to enhance sustainability and fairness of the national pension system (Asher and Bali, 2013).It is not absolute but relative poverty that is an increasing concern in affluent societies like Singapore.Recognizing that for many Singaporeans, their Housing and Development Board, or HDB, flat is a nestegg for retirement, the Government provides a range of options for older people to unlock the value of their HDB flats to supplement their retirement income. Seniors aged 65 orolder can take up the Lease Buyback Scheme by selling part of their flat’s remaining lease to HDB, while living in the same flat. They will have to use part of their Lease Buyback Scheme proceeds to top-up their Central Provident FundRetirement Account. Alternatively, those aged 55 or older who have a right-sized to a three-room or smaller flat can take up the Silver Housing Bonus by topping-up part of the sales proceeds into the Central Provident Fund’s Retirement Account. Lastly, flat owners, both older and non-older people, can rent out their spare bedroom or flat for income.While the CPF LIFE scheme applies pooling mechanisms in conversion of accumulated savings to a pension annuity, pooling of risks and income redistribution between richer and poorer members of the society and across generations is still low compared with that in traditional statutory (definedbenefit, pay-as-you-go) social insurance systems. As a result, income inequality during the working life is many times carried forward during retirement. This is particularly relevant for women, who generally live longer but have lower labour force participation rates and, consequently, lower balances in their Central Provident Fund accounts. This leaves women at a disadvantage because they are more dependent on receiving support from children in their old age.Due to societal ageing, growing numbers of women will not be able to receive such support in the future and are more susceptible to falling into absolute poverty.Singapore is an example for other countries in the region with provident funds of potential measures that assist their members in dealing with some of the challenges of pure defined-contribution schemes. Elements included in recent years, such as the Silver Support Scheme and the CPF LIFE, are good examples of public intervention in improving old-age income security.Another important aspect to assess when describing the system relates to Singapore’s foreign workers. Of its 5.6 million population as of 2018, 1.6 million were non-residents (including foreign dependants, international students and foreign workers, or individuals who are not citizens or permanent residents). While foreign workers are not eligible for the Central Provident Fund19 and healthcare subsidies20 and benefits (Asher and Bali, 2013), employers are required to bear the costs and responsibilities for their foreign workers’ upkeep and maintenance under the Employment of Foreign Manpower Act throughout their employment period in Singapore. These include bearing the cost of medical treatment for their workers and purchasing and maintaining medical insurance19The Central Provident Fund is only extended to Singaporeans and foreigners who are permanent residents; foreigners and their employers are not required to make contributions.20 Foreigners may tap into the MediSave of approved Singaporean family members.191coverage of at least SGD15,000 per year per worker. Foreigners can also save for their retirement while working in Singapore through the Supplementary Retirement Scheme, which provides tax incentives for voluntary savings.4. Policy directionsSingapore’s pension system is a mature, efficient and well-governed system that being fullyfunded is “demographysafe”. This is important in the context of its demographic ageing. The system’s emphasis on personal responsibility and ownership of savings and the Central Provident Fund can be used to meet housing, healthcare and retirement needs. The main challenge will be to ensure that not too much capital is withdrawn before retirement, or in the case of housing, that there continue to be options for home owners to unlock the housing value for retirement needs.Old-age income security in Singapore: Policy messages• Assess the feasibility of expanding social risk pooling principles to a part or whole of Central Provident Fund contributions, including the feasibility of integrating a defined-benefit component.• Consider introducing a flat-rate, inflation-adjusted, universal social pension financed from the general government budget.Reforms that have increased the minimum amount to be left in the retirement account and the introduction of the Central Provident Fund LIFE annuity scheme (Pension Funds Online, n.d.) and the efforts to ensure that seniors have options to monetize their HDB flats, where about 80 per cent of Singaporeans live, are steps in the right direction.At the same time, there are challenges in having the Central Provident Fund as the main old-age income security system. To enhance the sustainability and fairness of Singapore’s pension system and ensure adequate pensions in old age, it may be useful to keep reinforcing social risk-pooling principles within the member contributions and to assess the feasibility of increasing the coverage of the Silver Support Scheme byexpanding it towards a universal, flat-rate, non-contributory social pension for all people linked to per capita income, whose value does not diminish over time.Having a flat benefit for everyone can not only help in maintaining absolute poverty at close to negligible levels by providing benefits for those in the lower brackets of income but also bring additional benefits for those who are covered under the Central Provident Fund. In the context of demographic ageing and the increasing burden on family support structures, it may also be necessary to consider a shift in policy focus from addressing absolute poverty to relative poverty.More open debate among all stakeholders could facilitate identifying the most appropriate policy choices for Singapore (Asher and Bali, 2013).The foreseeable social problems of Singapore, particularly in an ageing context, should be addressed by the Government by taking a greater role in financing, as in other countries with aged populations. This proposal would require overcoming the Government’s general “commitment to hold welfare expenditure to the barest minimum, offering assistance only to the handicapped and the aged” (Asher and Bali, 2013). Such reform, while requiring a major reform of the Government’s general revenue, could tap into the existing support parameters for the aged and is worth the effort because markets alone are not sufficient to address the social security problems of ageing populations.192Table CB3. Pension scheme parameters for SingaporeCentral Provident Fund Pension schemeQualifying conditionsMinimum withdrawal ageBasic formula5550 or 55Pensioners can receive their pensions in three ways:(a) a lump-sum payment; or(b) 40% pension commuted into a gratuity + 60% reduced monthly pension, or(c) monthly pension of up to two-thirdsthe last drawn pensionable monthly salaryPension FormulaContributionsMinimum benefitMaximum benefit Base salary Benefit indexationEmployeeEmployer% of gross earningsGovernmentEarnings baseMeasure Minimum MaximumNote: Pensionable officers who joined service on or after 1January 1995 will only be eligible for either option (a) or (c).Pension paid is based on the officer’s length of service and pensionable salaryon the eve of retirement.20% of earnings in excess of SGD75017% of earnings in excess of SGD50Under the Workfare Income Supplement Scheme, thoseaged 35 and older with monthly income SGD2000 and less receive up to 60% of their Workfare payment* in Central Provident Fundcontributions.n.a. Pension is not indexedn.a.n.a.n.a.*Maximum annual Workfare pay-out in cash and CentralProvident Fund currently at SGD3600. Maximumamount will be increased toSGD4000, for work done from January 2020.Gross incomen.a.SGD50n.a.SGD6000n.a.Source: CPF, n.d.; Asher and Bali 2013.193REFERENCESAllianz. 2014. “Providentfund systems in Asia – Challenges and reform directions”, presentation by Richard Wolf at the SAS Conference, June 2014, Singapore.Asher, M.G.; Bali, A.S. 2013. “Fairness and sustainability of pension arrangementsin Singapore: An assessment”, Malaysian Journal of Economic Studies, vol. 50, No. 2, pp. 174–191.Asher, M.G.; Rajan, R. 2000. Social protection in Singapore (Singapore, Friedrich-Ebert-Stiftung Foundation).Available at: http://library.fes.de/pdf-files/iez/01443007.pdf [10May 2019].Central Provident Fund (CPF). n.d. “Central Provident Fund”.Available at: www.cpf.gov.sg [10May 2019].—. n.d. Silver support scheme (Singapore).Available at:www.cpf.gov.sg/Members/Schemes/ schemes/retirement/silver-support-scheme[10May 2019].Cunico, K.; Lim, Y.; Han, J. 2017. “Ploughing on: The faces and insecurities of Singapore’s elderly working poor”,Channel News Asia, 7 May. Available at: www.channelnewsasia.com/news/cnainsider/ ploughing-on-the-faces-and-insecurities-of-singapore-s-elderly-8824490 [10May 2019].HelpAge International (HAI). 2018.“Pension watch: Country profile on Singapore”.Available at:http:// www.pension-watch.net/country-fact-file/singapore/[14 Aug. 2019].International Labour Organization. 2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].International Monetary Fund (IMF). 2018. World economic outlook update, July 2018(Washington, DC).Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[14 Aug. 2019].Ministry of Communication (MOC). 2017. “7 ways the Government is helping seniors”.Available at: www.gov.sg/news/content/7-ways-the-government-is-helping-seniors [14 Aug. 2019].Ministry of Manpower (MOM). 2016. “Factsheet on the silver support scheme”.Available at: www. mom.gov.sg/~/media/mom/documents/budget2016/factsheet-on-silver-support-scheme.pdf [14 Aug. 2019].Ng, K-H. 2015. Can Singapore’s Central Provident Fund still meet retirement income needs?.Organisation for Economic Co-operation and Development (OECD). 2012.Pensions at a glance Asia/ Pacific (Paris).—. 2013.Pensions at a glance Asia/Pacific(Paris).Pension Funds Online. n.d.“Country profile: Singapore”.Available at: www.pensionfundsonline.co.uk/ content/country-profiles/singapore/101 [14 Aug. 2019].Singapore Department of Statistics (DOS). 2019. Death and life expectancy(Singapore). Available at: https://www.singstat.gov.sg/find-data/search-by-theme/population/death-and-life-expectancy/ latest-data [17 Nov. 2019].194—. 2018a. Key Household Income Trends. Available at: www.singstat.gov.sg/-/media/files/ publications/households/pp-s25.pdf [14 Aug. 2019]. —. 2018b. Population Trends 2018. Available at: www.singstat.gov.sg/publications/populationtrends [14 Aug. 2019]. —. 2018c. Complete Life Tables for Singapore Resident Population, 2017-2018. www.singstat.gov. sg/publications/population/complete-life-table [14 Aug. 2019]. United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019]. United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[14 Aug. 2019]. University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[14 Aug. 2019]. World Bank. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/[14 Aug. 2019].195COUNTRY BRIEF: THAILANDElements of the old-age pension system Non-contributory social assistance • Old Age Allowance, or Bia Yung Cheep: Tax-financed universal old-age allowance paid topersons aged 60 or older, with progressively higher benefit levels based on age bands. Non-contributory social insurance • Civil servant pensions: Under the Government Pension Fund, these are primarily non-contributory defined-benefit pensions with a maximum replacement rate of 70 per cent and a lump-sum payment from a defined-contribution savings scheme upon retirement. Contributory social insurance • Partially funded, contributory defined-benefit scheme of the Social Security Office:It is mandatory participation for private sector companies with at least one employee and some public sector employees. • Provident fund: This is an occupational savings fund that is mandatory for employees in stocklisted companies but otherwise voluntary. When the fund is established under an agreement between the employer and employees, they contribute together at between 2 per cent and 15 per cent. • National Pension Fund: A mandatory provident fund based on a defined-contribution system for formal workers in the private sector. Employers and workers each contribute 3 per cent of the worker’s salary.Voluntary schemes • National Savings Fund:It is a fullyfunded, defined-contribution scheme with voluntaryparticipation for workers who are not contributing to any other pension fund or have dropped out of mandated schemes, including self-employed workers.Legal framework • 1996: Government Pension Fund Act • 2011: National Savings Fund Act • 2015: Amended Social Security Act No. 41961. Development of old-age income security policiesThe first formal private social insurance in Thailand dates to 1972 and consisted of a work injury insurance scheme. Two years later, the Workmen Compensation Fund was established by the Ministry of Interior, which introduced an“employer-liability system”and the protection of employees of industrial and commercial firms from work-related injury, illness, disability or death. Mandated old-age benefits for the private sector, with few exceptions, were introduced only in 1998, long after sickness, maternity, disability and death benefits had been established by Thailand’s first Social Security Act in 1990. Workers in informal employment or self-employment have been able to join the scheme on a voluntary basis since 1994 and contribute at a flat annual rate.By 2014, when the first cohort of contributors had completed the required 15 years of contribution to qualify for minimum old-age pension entitlements, the first pension pay-outs from the National Social Security Fund were made. Overall, these remain at very low levels, assuming a subsidy character in relation to private household income rather than an income guarantee for old-age persons.In 2011, two options were introduced under Article 40 of the Social Security Actfor participation of private sector workers in pension schemes: either (i) workers could choose to contribute only to the Social Security Office scheme or (ii) workers could combine it with a voluntary pension savings scheme. Contributions to the savings funds were fixed at a low flat rate and supplemented by the government through, also low, matching contributions. Since 2015, voluntary contributors can no longer choose benefit packages within the Social Security Fund scheme, which includes the old-age pensions. Instead, a voluntary National Savings Fund was created to cover self-employed people through individual savings accounts. The Government matches individuals’ contributions up to a certain amount.In 2018, a provident fund with mandatory participation for formal workers in the private sector was started. The details of its implementation are still being worked out. The National Pension Fund aims to increase retirement savings for private sector employees, especially low-income earners, who are not covered by voluntary provident funds. It has the potential to reach about 13 million workers, or 34 per cent of the labour force who are currently uncovered. The Fund mandates employers to match employee contributions starting with 3 per cent of the salary in the first year and with the contribution rate increasing every few years.In addition to contributory schemes tax-financed means-tested old-age pensions, known as the OldAge Allowance, were established in 1993 in Thailand by the central Government and implemented by local authorities at the district and provincial levels. These, however, differed widely in their interpretation of national guidelines and application of eligibility criteria, which resulted in abuse and leakages of programme resources, in addition to the worrisome exclusion error of more than half of the initial target group (Suwanrada, 2014). Since 2009, the eligibility criteria for receipt of pensions were broadened, and the old-age allowance was considered to be a universal scheme, even though eligibility restrictions with respect to citizenship, residential area, receipt of other social transfers and civil servants still remain.Other reforms to the benefit levels and administration of the programme were undertaken at the same time, but, in terms of adequacy of benefits, the old-age allowance and the Social Security Fund schemes face similar issues. The Government will allow simultaneous receipt of the old-age allowance and the Social Security Fund pensions from 2019 to benefit pensioners and facilitate administration of both schemes.Public employees are insured for old-age retirement income under four groups: (i) civil servants covered by the Government Pension Fund; (ii) officers who are covered by the Social Security Fund;197(iii) officers whose insurance is governed by the rules of the public institution where they work; and (iv) permanent employees who receive annuitized pensions funded by the Government.2. Socioeconomic backgroundFrom the 1960s until the Asian financial and economic crises that began in 1997, Thailand’s economy grew at an impressive average annual rate of 7.5 per cent. During 1999 and 2005, annual growth rates were kept steadily above 5 per cent (figure CB1). Remarkably, export-led economic growth has contributed to welfare improvements for the population on many fronts, including health, education and living standards. It has also led to industrial development and formalization of the labour market. Since 2009, economic growth has been more volatile, and the Thai political sphere became relatively less stable since 2013, which led to foreign investment and consumption trends to momentarily stagnate. From 2015, however, the economy has shown signs of steady recovery, maintaining annual GDP growth above 3 per cent and reaching unprecedented per capita income levels in 2017 (table CB1). Table CB1. Socioeconomic and demographic data for ThailandSocioeconomic parametersGDP per capita (US$) GDP growth, real (%) Inflation (%) Poverty rate (%)* Gini coefficient Human Development Index (ranking)6590 3.9 0.7 7.937.8 87Year2017 2017 2017 2017 2013 2015Demographic parametersLatest available year**20202050Total population (million)69.2Urban population share (%)49.9Share of population aged 60 or older (%)17.7Old-age dependency ratio14.8Life expectancy at birth (years)74.6Life expectancy at age 60 (years)21.969.465.451.469.519.135.118.250.075.682.122.225.7Note: *Poverty headcount ratio at national poverty line: approximately US$6.20 a day in 2011 PPP. **Population, share of urban population, share of the older population: 2018.Dependency and life expectancy: 2015.Data source: DESA Population Division 2018; IMF, 2018; World Bank, 2018; UNDP, 2018.Poverty and extreme poverty rates have declined substantially over the past 30 years in Thailand. The country overcame extreme poverty for the employed population in the 2010s, as measured by the US$1.90 at PPP per day international poverty line, and more than halved poverty from 2008 to 2015, as measured by the national poverty line (table CB1). Households in the North, North-East and the Deep South of Thailand have the lowest relative incomes, in general terms. Despite the significant progress in alleviating poverty, income inequality remains a substantial societal challenge. Income disparities are slightly higher in urban areas than rural areas. A little more than half of the total population lives in rural areas (table CB1), including the vast majority of the poor.198Figure CB1. Macroeconomic performance of Thailand and Eastern Asia and the PacificPanel A. GDP growth, by decade averagePanel B. Stability of growth by decade Data source: World Bank, 2018;University of Groningen, 2013.Table CB2. Labour market and pension indicatorsfor Thailand, latest available yearLabour market indicatorsLabour force participation rate: aged 15 or older (%)68.1Men76.9Women59.8Own-account and contributing family workers (% of labour force)30.9Average monthly wage (US$)412Year 2017 2017 2017 2017 2017Pension indicatorsYearActive pension contributors (% of working age population:) Private sector defined-benefit scheme (Social Security Fund)Old-age pension recipients (% of population older than 60) Contributory schemes Old-age allowancePublic social protection expenditure for older persons (% of GDP)n/a33.6201583.020168.2201674.820162.22015Contributory private sector pension and voluntary provident fund (Social Security Fund/NSF)0.82016Old-age allowance0.52016Civil servant pension (GPF)1.22016Average benefit, Old-Age Allowance (% of minimum wage)10.32016Data sources: ILO, 2018a; ILO, 2018b; ILO, 2017.199Figure CB2. Thailand’s population pyramid, 2020 and 2050Data source: UN Population Division, 2018. Thailand’s working-age population represented 83 per cent of total population in 2017 and the labour force was 68 percent. Both men’s and women’s participation rate shrunk roughly 3 percentage points in the past five years, but the men’s rate remained 1.3 times higher than for women (ILO, 2018b). Women’s lower representation in the labour market highlights actual gender-based vulnerabilities with regards to low pension levels in the old-age allowance because it shows that theyare likely to have more limited sources of retirement income and to depend more on the universal pensions. The share of services in the Thai economy has risen since 2014 and represented 46 per cent of total employment in 2017. The self-employed represented 31 per cent of the Thai labour force in 2017 (table CB2) share that has decreased in the past five years, during which the share of wage employees increased 7 percentage points (ILO, 2018b). 3. Old-age pension challenges Thailand’s pension system faces some of the typical challenges among middle-income economies in the region. First, the Social Security Office contributory pensions for private sector workers require parameter reform to adjust to demographic trends (figure CB2)and scheme maturation. And second, appropriate models for increasing coverage of the substantial informal sector need to be in order and implemented in the near future. The Social Security Fund pension benefits are not automatically indexed (to inflation, for instance), with adjustments being made on an adhoc basis only, allowing the scheme to run the common risk of benefit devaluation over time. In addition and due to the fact that the ceiling for contribution based on a salary of 15,000 Thai baht (THB) was never adjusted since the launch of the scheme in 1999, the insurable earnings for an important segment of the members is relatively low. This also sheds light on the low relevance of the pensions currently provided and their subsidy character. In practice, the first pensioners have accumulated relatively reduced years of contribution, making the replacement levels lower in relation to entitlements that will be based on longer periods of contribution. Additional policy measures will be required to increase coverage of contributory pensions, including to workers in informalemployment, that provide more adequate benefit levels than the current 200schemes. An estimated 33.6 per cent of the working-age population actively contributed to the Social Security Fund scheme in 2015, leaving substantial coverage and contribution-base gaps, especially among the share of workers in own-account or contributing family work who are highly dependent on the old-age allowance as a source of retirement income.With regards to coverage of old-age, tax-funded benefits, the country has reached a remarkable level of coverage through its universal old-age allowance, which is an almost unique achievement among countries at a similar stage of economic development in the ASEAN region. Overall, pension coverage is high, at an estimated 83 per cent of persons aged 60 or older covered in 2016, primarily due to the (nominally) universal social pensions (table CB2). The observed discrepancy to full coverage, with 17 percent uncovered, is likely to be a result of two factors: (i) self-selected abstinence due to modest benefit levels, especially given that the share of recipients is significantly lower in richer households and (ii) administrative obstacles faced by some of the eligible.The old-age allowance is not able to sufficiently protect older people from income poverty, especially with the average benefit level corresponding to merely 10 per cent of the minimum wage in 2016 (table CB2). In comparison, the Social Security Office private pension scheme generally provides, in theory, a more adequate replacement rate, in line with the 50 per cent target set out by government policies on old-age income security (table CB3). Data from surveys on behalf of the government indicate that 42 per cent of persons aged 60 or older have insufficient income to maintain preretirement living standards. This illustrates one of the most pressing issues faced by the Thai pension system, which is the adequacy of pension benefits.Moreover, as mentioned, providing adequate protection from old-age poverty will require not only a balanced approach of increasing the benefit levels of the social pensions (old-age allowance) but also of expanding the contribution base in the informal sector. Despite Thailand’s fairly advanced stage of economic development as an upper-middle-income economy, around 48 per cent of the labour force in 2017 was outside formal employment, either as own-account workers or contributing family workers (ILO, 2018b).The life expectancy of pensioners is expected to increase only modestly, in relative terms, from the current 22.2 years at age 60, to 25.7 years at age 60 by 2050. However, the share of the population aged 60 years or older will reach 35.1 percent in 2050 (table CB1), higher than the current figure in countries currently experiencing the severest ageing, such as Japan, with a corresponding share of 33.6 percent.Consequently, expenditures on social protection for the older population, which remain at rather modest levels (table CB2), will inevitably need to increase substantially over the coming decades.Another challenge has to do with the long-term financial (un)sustainability of the Social Security Fund pensions. Benefit payments from the Fund are predicted to outgrow contributions by 2034, with reserve depletion by 2054.21 As a result, gradual parameter reforms to increase the pensionable age from 55 and to increase the low contribution rates will be necessary in the near future.4. Policy directionsThe Ministry of Finance is at the planning stage of a wide-ranging pension reform, with the aim to establish a multitiered system.21 Actuarial valuation of Thailand Social Security administered by the Social Security Office (ILO, 2016).201Old-age income security in Thailand: Policy messages • Improve the adequacy of benefits, especially for the old-age allowance. • Extend the contribution base. • Harmonize the system. • Conduct a parametric reform of the system.Looking at the current system gaps, it would be extremely important to ensure that the reformed designof the pension system considers the different components that promote an integrated approach in which tax-funded benefits and contributory-based schemes complement each other to guarantee minimum levels of protection.Table CB3. Pension schemes parameters for ThailandQualifying conditionsContribution years (CY) Pensionable ageBasic formulaPension formulaMinimum benefit Maximum benefitBase salary (BS) Benefit indexationEmployeesContributions% of gross earningsEmployers GovernmentEarnings baseMeasure Minimum MaximumPrivate sector defined benefit (Social Security Fund)Voluntary savings (National Savings Fund)15?5560(20%+1.5%*(CY-15))*BSAnnuity (guaranteed rate of return equal to 1 year fixed-term deposit)inferred from minimum earnings baseTHB600inferred from maximum earnings base-5-yr non-indexed avg. individual wage---3% (incl. family benefit)THB50 per month to THB13200 per year3% (incl. family benefit)-1% (incl. family benefit)50–100% matching contribution up toTHB600–THB1,200 per year, dependingon age?-THB1650-THB15000-Universal social pension Qualifying conditionsBenefit amountLocal currency % of GDP per capita %of minimum wageAge 60 600–1000, depending on age3.3–5.5% 9.2–15.3%Source: ILO, 2016.202REFERENCESAsian Development Bank (ADB). n.d. “Thailand”. Available at: www.adb.org/countries/thailand/ poverty[10Feb. 2019].HelpAge International (HAI). 2018. “Pension watch: Country profile on Thailand”. Available at:http:// www.pension-watch.net/country-fact-file/thailand/[14 Aug. 2019].International Labour Organization. 2013. Thailand: A labour market profile(Bangkok, Regional Office for Asia and the Pacific). Available at: www.ilo.org/wcmsp5/groups/public/---asia/---ro-bangkok/ documents/publication/wcms_205099.pdf[8 Oct. 2018].—. 2016. Report to the government: Actuarial valuation of Thailand social security administered by the Social Security Office as of 31 December 2013 (Geneva, Social Protection Department).—. 2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].International Monetary Fund (IMF). 2018. World economic outlook update, July 2018 (Washington, DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[14 Aug. 2019].Lloyd-Sherlock, P.; Pot, A.M.; Sasat, S.; Morales-Martinez, F.2017. “Volunteer provision of long-term care for older people in Thailand and Costa Rica”, Bulletin of the World Health Organization. Available at: www.who.int/bulletin/online_first/BLT.16.187526.pdf?ua=1[8 Oct. 2018].Organisation for Economic Co-operation and Development (OECD). 2015. Pensions at a glance(Paris).Suwanrada, W. 2014. “National experiences from Thailand regarding long-term care of older persons”,presentation at the Regional Expert Consultation on Long-term Care of Older Persons 9-10 Dec. 2014, Bangkok. Available at: www.unescapsdd.org/files/documents/Item per cent203 per cent20 per cent28b per cent29 per cent20Long-term per cent20care per cent20in per cent20Thailand. pdf[8 Oct. 2018].United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019].United Nations Development Programme (UNDP). 2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/ files/2018_human_development_statistical_update.pdf[14 Aug. 2019].University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[14 Aug. 2019].World Bank. 2016. Live long and prosper – Ageing in East Asia and Pacific(Bangkok).—. 2017. “Thailand: Overview”. Available at: www.worldbank.org/en/country/thailand[8 Oct. 2018].—.2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019].—. n.d. ASPIRE, Pensions database (Washington, DC).203COUNTRY BRIEF: VIET NAMElements of the old-age pension systemNon-contributory social assistance• Tax-funded social pensions: A pension-tested benefit for people aged 80 and older and a means-tested benefit for people aged 60–79 living in poor households without family support or caregiver.Non-contributory social insurance• Civil servants’ pension scheme: A non-contributory pre-1995 government scheme, which is being phased out.• Merit allowance:A pension scheme for veterans implemented separately.Contributory social insurance• Mandatory social insurance: A contributory scheme for public and private sector employees with at least a one-month contract, including household workers; employees in agriculture, fishing and salt production; employees of cooperatives and unions; police and military personnel; part-time workers in communes, wards, and townships; and foreign citizens legally working in Viet Nam.• Voluntary social insurance: A scheme for people without mandatory social insurance coverage, such as self-employed workers. Members of the scheme have their premiums partially supported by government subsidies based on the poverty line in rural areas: for the period 2018–20, 30 percent of premiums subsidized for members of poor households (=VND46,200per month), 25 percent for members of households living just above the poverty line (=VND38,500 per month), and 10 percent for others (=VND15,400 per month).Legal framework• 2002: Labour Code – Law No. 35/2002/QH10 and No. 10/2012/QH13, 2012 • 2006: Social Insurance Law22 No. 71/2006/QH11, and No. 58/2014/QH13, 2014• 2009: Law on the Elderly 39/2009/QH12 and related Decree 06/ND-CP, 2011• 2007: Social Assistance Decree No. 67/NÐ-CP, No. 13/NÐ-CP, 2010 and No. 136/NÐ-CP, 2013• 2017: Decision on Master Plan for Social Assistance Reform (MPSAR) 488/QÐ-TTg• 2018: Resolution No. 28-NQ/TW, of the Fifth Conference of the Party Central Committee XXII on reforming social insurance policy22 The legal framework is complemented by several decrees and circulars: Decree No. 134/2015/ND-CP dated 29 Dec. 2015 on detailing a number of articles of the Law on Social Insurance. Resolution No. 125 / NQ-CP dated 8 Oct. 2018 promulgating the Government’s Action Program to implement Reso-lution No. 28-NQ / TW. Resolution No. 102 / NQ-CP dated 3 Aug. 2018 on assigning targets for development of social insurance participants. Decree No. 143/2018 / ND-CP, dated 15 Oct. 2018 on detailing the Law on Social Insurance and the Law on Occupa-tional Safety and Health on compulsory insurance for employees who are foreigners working in Viet Nam. Decree No. 115/2015 / ND-CP dated 11 Nov. 2015 detailing a number of articles of the Law on Social Insurance oncompulsory social insurance. Circular No. 01/2016 / TT-BLĐTBXH, dated 18 Feb. 2016 of the Ministry of Labour, Invalids and Social Affairs detail-ing and guiding the implementation of a number of articles of the Law on Social Insurance regarding voluntary social insurance. Circular No. 59/2015 / TT-BLĐTBXH, dated 29 Dec. 2015 of the Ministry of Labour, Invalids and Social Affairs detailing and guiding the implementation of a number of articles of the Law on social insurance regarding compulsory social insurance.2041. Development of old-age income security policiesSocial insurance pensions first emerged in the 1930s for French colonial officers. Following the 1945 revolution, the new Constitution of Viet Nam provided for assistance to “aged or disabled workers unable to work”. A retirement allowance for civil servants and workers in state-owned enterprises was first legislated in 1947 but not implemented until 1961 due to war. The pension was largely state financed with minor contributions from state-owned enterprises and by the late 1980s covered about 25 per cent of the labour force.23A contribution-financed social insurance system covering the public and private sectors was first provisioned in the 1992 Constitution in the course of major economic liberalization. The economic reforms resulted in a major shift of employment from the public to the private sector.24 The private sector gained social insurance coverage in 1993, later formalized in the 1994 Labour Code mandating coverage for workers in companies with at least 10 employees. The target replacement rate was set at a comparatively high 75 percent.25 The 2002 Labour Code expanded coverage to all workers with at least a three-month labour contract. Voluntary social insurance was implemented in 2008 following the promulgation of the 2006 Social Insurance Law.The first parametric reforms to safeguard the financial sustainability of the long-term fund were implemented between 2010 and 2014, gradually raising contribution rates from 5 per cent and 10 per cent to 8 per cent and 14 per cent for employees and employers, respectively. Another round of parametric reforms is expected to take place, based on the 2014 Social Insurance Law.The present reforms are implemented in a context of uncertainty regarding the financial sustainability of the social insurance fund. Prior to the current round of reforms, a deficit was estimated to arise as early as 2021, with reserve depletion by 2034 (ILO, 2013). Although these reforms reflect a recognition of the need to reduce the generosity of social insurance pensions in the long term, an exceptional 8 per cent pension was introduced in 2015, financed from general government revenues and social insurance fund reserves.26 This increase was granted to all pensioners regardless of pension benefit level.In parallel, a pension-tested social pension was first implemented in 2002. The eligibility age has gradually been lowered from 90 years to 80 years of age. With this pension-tested scheme, Viet Nam achieved universal old-age protection for people aged 80 and older. A means-tested benefit is available to people aged 60–79, covering about 95,000persons (ILO, forthcomingc).Resolution No. 28-NQ/TW of 2018 provides a set of directions for reforming social insurance policy and increasing effective and adequate coverage. It sets the target of increasing contributory social security coverage to 35 per cent of the working age population in 2021, 45 per cent in 2025 and 60 per cent in 2030; increasing social pension coverage to 45 per cent in 2021, 55 per cent in 2025 and 60 per cent in 2030 and increasing member satisfaction to 80 per cent in 2021, 85 per cent in 2012 and 90 per cent in 2030. It includes also a target to increase participation of workers in informal employment in contributory social insurance. The Resolution also proposes to build a multitier social insurance system, enhance linkages among social insurance policies and implement flexible mechanisms to achieve coverage expansion.23 Equivalent to the share of public sector employment, which was low compared with other socialist economies as agriculture accounted for the majority of the labour force.24 Public sector employment fell from close to 25 per cent in 1986 to about 15 per cent in 1991 (World Bank, 1994). 25 Normally reached after 30 contribution years for men and 25 years for women. 26 Resolution 78/2014/QH13, adopted by the National Assembly as part of the 2015 budget after higher than expectedrevenues in 2014. The increase applies to social insurance pensioners and merit allowance recipients (veterans).205The Resolution also establishes several other priorities such as reforming the pension formula, changing the conditions for access to lump-sum benefits; expanding the coverage of the mandatory scheme to additional groups of workers, reforming the system parameters (such as retirement ageing) to ensure the long-term sustainability of the fund; and changing the pension indexation mechanism. All these modifications are currently being discussed and expected to be inserted in the forthcoming revision of the social insurance law.2. Socioeconomic backgroundFollowing important economic and political reforms launched in 1986 under Đổi Mới, Viet Nam has developed from one of the world’s poorest nations to a lower-middle-income country. Since 1990, the growth in GDP per capita has been among the fastest in the world, averaging 6.4 per cent a year in the 2000s. The country’s medium-term outlook remains favourable, with GDP expanding by 6.8 per cent in 2017, while the country’s fundamental drivers of growth – resilient domestic demand and export-oriented manufacturing – remain in force.Economic growth has been accompanied by a sharp reduction in poverty. More than half of the population lived on less than US$1.90 a day in 1993, extreme poverty concerned 7 per cent of the population in 2015. During the same period, the share of individuals living below the national poverty line went down, from about 60 per cent to 13 per cent. Through advances in basic service provision, the population is significantly more educated and healthier than 20 years ago (World Bank, 2017).Inequality, represented by the Gini index, increased from 35.7 to 38.7 in the 20-year period from 1992 to 2012, and then declined to 35.3 in 2016. The Government is committed to reforms. The Socio-Economic Development Strategy 2011–2020 highlights the need for structural reforms, environmental sustainability, social equity and emerging issues of macroeconomic stability (World Bank, 2017).Table CB1. Socioeconomic and demographic data for Viet NamSocioeconomic parametersYearGDP per capita (US$) GDP growth, real (%) Inflation (%) Poverty rate (%)* Gini coefficient Human Development Index (ranking)Demographic parametersTotal population (million) Share of urban population (%)Share of population aged 60 or older (%)Old-age dependency ratio Life expectancy at birth (years) Life expectancy at age 60 (years)Latest available year** 2020 96.5 35.911.69.6 75.6 22.32343 6.8 3.5 7.035.3 116205098.4 37.312.52017 2017 2017 2015 2016 2018114.6 57.328.311.734.976.682.122.926.1Note: *Poverty headcount ratio at US$1.90 a day in 2011 PPP. **Population, share of urban population, share of the older population: 2018. Dependency and life expectancy: 2015.206Data source: DESA Population Division 2018; IMF, 2018; World Bank, 2018; UNDP,2018. Figure CB1. Macroeconomic performance of Viet Nam and Eastern Asia and the PacificPanel A. GDP growth, by decade averagePanel B. Stability of growth by decade Source: University of Groningen, 2013; World Bank, 2018.Table CB2. Labour market and pension indicatorsfor Viet Nam, latest available yearLabour market indicatorsLabour force participation rate: aged 15 and older (%) Men WomenOwn-account and contributing family workers (% of labour force) Average monthly wage (US$) Pension indicatorsActive pension contributors (% of population aged 15–64) Contributory defined-benefit pension*Old-age pension recipients (% of population older than statutory pension age) Social pension Contributory defined-benefit pensionPublic social protection expenditure for older persons (% of GDP) Social pensionBenefit level, social pension (% of GDP per capita)Year76.1201781.1201771.5201756.120172122015Year17.1201617.1201639.9201513.0201816.120185.520150.120166.62016Note: *Includes public and private sector scheme members.Source: HAI, 2018; ILO, 2018b; ILO, 2017; ILO, forthcoming a. 207Figure CB2. Viet Nam’s population pyramid, 2015 and 2050Source: DESA Population Division, 2018. Today, Viet Nam is still a demographically young country with an old-age dependency ratio of just 10 per cent. However, the country is entering a rapid ageing transition that is among the fastest worldwide. By 2050, the number of people 60 years or older will almost triple, from 11.6 per cent to more than 28.3 per cent, or 32 million people. The number of people older than 80 will also triple to more than 6 per cent of the population. Additional pressure on the pension system will arise from increases in life expectancy. The average life expectancy at age 60 is expected to increase from 22.3years to 26.1 years until 2050 (table CB1). The remarkable economic growth and poverty reduction over the past two decades have not been shared equally by older people, especially those outside the formal economy. Many of the people working in informal employment continue working into their retirement as long as they can. Almost all older people in Viet Nam have married, but around 40 per cent are widowed. Primarily, women are affected by old-age poverty and isolation because the gender imbalance is more pronounced in older age groups. Women comprise 59 per cent of all older people, increasing to 70 per cent among those aged 80 years or older. However, because women are less likely to be employed in the formal economy, they are less able to access social insurance pensions from the Vietnamese Social Security. In 2016, 26 per cent of men aged 65 years and older but only 12 per cent of women accessed a Vietnamese Social Security pension. Older women are more at risk of income insecurity and isolation than older men; more than 60 per cent of women aged 65 years orolder are single or widowed, compared with only 17.8 per cent of men, and the proportion rises with age. The majority of old people live in low-income households. In 2016, almost 68 per cent of older people lived in households in which the per capita income was less than 100,000 Vietnamese dong (VND) (US$4.50) per day, while 31.5 per cent were living on less than VND50,000 (US$2.25). Specific categories of older people are more likely to live in low-income households. About 72.5 per cent of older people live in rural and mountainous areas, often without proper access to infrastructure and services (HAI, 2015). Furthermore, 81 per cent of older people in rural areas and 88 per cent of older ethnic minority groups were living on less than VND100,000 per day in 2016 (ILO, forthcoming b). 208According to the national surveillance ofolder persons in Viet Nam, conducted in 2011, 56 per cent of respondents self-assessed their health conditions as weak or very weak. Nearly four of every 10 individuals in the 60 and older age group did not own a health insurance card (HAI, 2015).3. Old-age pension challengesSimilar to other middle-income countries in the region, Viet Nam faces the dual challenges of improving the financial sustainability of the pension system and expanding coverage to a substantial informal sector, including the large number of people who are already old and do not benefit from any pension. Both challenges carry urgency due to the rapid pace of demographic ageing.Viet Nam’s labour market is characterized by a high degree of non-wage and informal work, with only about 41 per cent of working adults being in wage work (ILO, 2018b). The share of wage employees in the labour force has doubled since the beginning of the millennium, largely as a result of a shift away from agriculture and a reduction in the share of contributing family workers. There is some potential for a continued trend towards formalization, leading to social security coverage expansion in the future.A significant number of employees, at 5.4 million, work without employment contracts or social insurance coverage in the formal sector (ILO, forthcoming c), thereby illustrating the need for enhanced compliance and enforcement in the formal sector. Viet Nam has only limited scope to increase the number of social security contributors through higher labour force participation. Labour force participation rates are already among the highest in the region, with only a modest gender gap.Coverage is relatively high in the 80 and older age group due to the pension-tested scheme, which benefits 1.35 million, or 70 per cent of the 80 and older age group (HAI, 2018). A worrying gap remains in the 60–79 age group, where around 70 percent of people do not receive any income from government schemes. Social pensions are paid to only 2.3 per cent of this age group due to narrow eligibility requirements.Active coverage of contributory social security in the working-age population increased slightly from 19.9 percent in 2010 to 28.5 percent in 2017. Of this, just 1.7 per centwas voluntary in 2017. Key issues preventing coverage expansion are a persistently large share of informal work in the labour force, slow expansion of voluntary membership and only 70 per cent compliance with mandatory membership. In 2017, the Master Plan for Social Assistance Reform was approved, which aims to expand coverage by gradually lowering the minimum eligibility age for those receiving the pensiontested tax-funded pension.Still, the social pension benefit levels are low, when compared with the minimum standard of living. The benefit is set at VND270,000 per month (US$11.6), which may explain the high poverty rates among people aged 80 orolder, even though they receive a social pension. The value of the meanstested pension for people aged between 60 and 79, living in poverty and with no family support is VND405,000 per month (US$18.5), but only reaches around 95,000 older persons (ILO, forthcoming c). The majority of social pensioners receive only VND270,000, which is equivalent to 18 per cent of the near-poverty line and contributes very little in income support for older people.The contributory pension system offers generous benefits in regional and global comparison.27 The pension design is such that the maximum replacement rate is 75 per cent of the reference wage27 Gross replacement rates in most countries in the region lie in the range of 35–50 per cent. The average in OECD countries is about 65 per cent.209after 35 years of contributions. People who have not contributed for 20 years and those who have discontinued contribution payments for at least one year can receive a lump-sum payment.A large number of insured persons, around 600,000, opted for the lump-sum payment in recent years and thus do not have adequate protection in their old age. Inadequate protection will however become a financial burden for the Government in the future (ILO, forthcoming a).Another challenge affecting the sustainability of the pension system is that private sector workers partially finance civil servant pensions. While private sector pensions are based on average career earnings, civil servant pensions are based on the average earnings of the last few years, which are generally much higher. The majority of pensioners in the system are civil servants with replacement rates estimated at 107 per cent. While former private sector workers only realize an average replacement rate of 42 per cent (Peyron-Bista et al., forthcoming).Viet Nam’s public social protection expenditure for the older population is estimated at 5.5 per cent, well above the Asian-Pacific average of 2 per cent.Expenditure on social pensions remains low, which results in low benefits, high minimum eligibility age of the pension-tested benefit and narrow selection criteria for the means-tested social pension. As Viet Nam moves towards an ageing society and a more mature pension system, the financial sustainability of the contributory pension system will have to be ensured through reform measures. Over time, life expectancy will improve, which implies that people will require pensions for a longer time (or will extend their working careers if their health conditions allow so). In combination with a decreasing ratio of contributors to beneficiaries, this trend may pose financial challenges to the system(ILO, forthcoming a).4. Policy directionsIn addition to the last round of parametric reforms under the 2014 social insurance law, further reforms will be necessary to safeguard the long-term health of the social insurance fund, adjusting such system parameters as the retirement age or accrual rates. Proposals for a retirement age increase have been repeatedly considered by the Government, including for the renewed Labour Code in 2019.Old-age income security in Viet Nam: Policy messages• Gradually increase the retirement age for men and womenin line with the ageing population and towards gender equality in retirement.• Introduce parametric reform of the existing defined-benefit pension scheme by reducing the accrual rate to per annum.• Introduce a multitier system that consists of a basic, universal or pension-tested tier for all residents, a social insurance tier for contributors and a voluntary savings scheme.• Introduce a voluntary notional defined contribution scheme.• Reform the system parameters (such as reduction in qualifying contribution years).• Change the pension indexation mechanism.210The issue of unequal retirement ages for women and men has been the subject of government debates since 2000. During its most recent country review in 2015, the Committee on the Convention on the Elimination of All Forms of Discrimination Against Women followed a joint UN recommendation to suggest equalizing retirement ages (UN, 2012). There is some evidence that women’s earlier retirement in the civil service prevents them from reaching the highest salary grades, leading to average pension of about two-thirds of men in the same career track (UN, 2012).One policy option is to gradually increase the retirement age to 65 for both men and women. A parametric reform of the existing defined-benefit pension scheme is also proposed that envisages a gradual reduction of the accrual rate to 1.5 per cent per annum. The aim is to keep the contribution rates in the long run at a level that is deemed to be sustainable (Cichon et al., forthcoming).Expansion of social insurance coverage to workers who are currently in informal employment (largely self-employed and contributing family workers), features prominently in the Master Plan for Social Assistance Reform and is a central element of the Master Plan for Social Insurance Reform. Mandatory social insurance membership also needs more effective enforcement and incentive policies, especially for SMEs, short-term contract workers and family businesses. A reduction of the eligibility age for the pension-tested benefit, along with continuous increases in the benefit amount, is being considered.To cover the “missing middle” (informal workers), better links between the tax-funded scheme and contributory schemes are required. This can take the form of a multitiered system consisting (i) a basic, pension-tested scheme for all residents; (ii) a social insurance defined-benefit pension for contributors largely from the formal sector; and (iii) a voluntary defined-contribution scheme complementary to those covered by the mandatory second tier and that is open to all people. The voluntary defined-contribution scheme is intended to help higher-income earners achieve a higher replacement rate (Cichon et al., forthcoming).Alternatively, the first tier could be a pension-tested scheme covering all residents who would not qualify for a contributory pension. A gradual approach based on the reduction of the eligibility age of the current pension-tested scheme could be a solution to reach the ILO Resolution No. 28 targets. According to ILO estimates, reaching those targets would require the eligibility age to be reduced from 80 to 67 years by 2030.Table CB3. Pension scheme parameters for Viet NamQualifying conditionsPension formulaContributory pension (defined benefit)Contribution years (CY) Pensionable age Basic formula Minimum benefit Maximum benefitBase salary (BS)Benefit indexationCurrentReform (2018)2055 females, 60 malesFemale: (45%+3%*(CY-15))*BS Male: (45%+2%*(CY-15))*BSFemale: (45%+2%*(CY-15))*BS Male: gradual transition until 2022 to(45%+2%*(CY-20))*BSCivil servant minimum wage75%*BSAverage lifetime salary, indexed on CPI. Public sector: averaging period 5-10 years depending on year of scheme entry (gradually increasing to lifetime earnings by 2025). Indexed on pub-lic sector base wage for those who joined before 2016.Law allows for indexation on CPI/GDP depending on budget capacity. Not yet reflected in implementation regulations.211Contributions% of gross earningsEmployees EmployersGovernmentEarnings baseMeasure Minimum Maximum8% (incl. disability + survivors)14% (incl. disability + survivors)Subsidies as needed, cost of pre-1995 public sector pensioners, contributions for civil servants who started working before 1995. Law allows for contribution subsidies for voluntary members depending on budget capacity. Not yet reflected in implementation regulations.Salary plus allowances Salary plus allowances and bonusesCivil servant minimum wage. Voluntary members: rural poverty line.20 times civil servant minimum wageQualifying conditions Benefit amountTax funded pensionsLocal currency% of GDP per capitaper cent of minimum wageAge 80+: pension-tested 270000 6.6%11.3%Age 60+: means-tested, having no family support/no caregiver 540000 13.2%22.5%Note: Higher benefit amounts apply to older people with disabilities.Source: ILO, Forthcoming a.REFERENCESCichon, M.; Peyron-Bista, C.; Clay, T. forthcoming. Options for the reform of the social insurance scheme in Viet Nam (Bangkok, ILO).General Statistics Office of Viet Nam (GSO). 2016. Report on labour force survey 2016(Hanoi).HelpAge International (HAI). 2015. “Ageing population in Viet Nam”. Available at: http://ageingasia. org/ageing-population-vietnam/ [14 Aug. 2019].—.2018. “Pension watch: Country profile on Viet Nam”.Available at:www.pension-watch.net/ pensions/country-fact-file/viet-nam [14 Aug. 2019].Hoi, L.V.; Thang, P.; Lindholm, L. 2011. Elderly care in daily living in rural Viet Nam: Need and its socioeconomic determinants. BMC Geriatr. 2011; 11: 81. Available at: www.ncbi.nlm.nih.gov/pmc/ articles/PMC3239225/[14 Aug. 2019].International Labour Organization (ILO). 2013. Actuarial valuation of the public pension scheme of the Viet Nam Social Security Fund(Hanoi).—.2017. World social protection report 2017–19: Universal social protection to achieve the Sustainable Development Goals (Geneva).—. 2018a. “ILO social security inquiry database”.Available at:www.ilo.org/sesame/IFPSES. SSDBMenu[14 Aug. 2019].—. 2018b. “ILOSTATdatabase”. Available at:www.ilo.org/ilostat/[14 Aug. 2019].—. Forthcoming a. Actuarial valuation as at 31 December 2015 of long-term benefits administered by the Viet Nam Social Security Organization (Hanoi).—. Forthcoming b. Comprehensive old age pension system for Viet Nam: Delivering on Resolution 28 (Hanoi).212—. Forthcomingc. Exploring reform options for a multi-tier pension system in Viet Nam (Hanoi). —; United Nations Population Fund (UNFPA). 2014. Income security for older persons in Viet Nam: Social pensions,policy brief (Hanoi). International Monetary Fund (IMF). 2018. World economic outlook update, July 2018 (Washington, DC). Available at:www.imf.org/en/Publications/WEO/Issues/2018/07/02/world-economic-outlookupdate-july-2018[14 Aug. 2019]. Organisation for Economic Co-operation and Development (OECD). 2015. Pensions at a glance (Paris). OXFAM. 2017. Even it up: How to tackle inequality in Viet Nam, Oxfam briefing paper (Hanoi). Available at: www.oxfam.org/sites/www.oxfam.org/files/file_attachments/bp-Viet Nam-inequality120117-en.pdf[14Aug. 2019]. Peyron-Bista, C.; Thi, N.; Giang, Q.; Hai, D.N. forthcoming. Discussion on options to the extension of social insurance coverage in Viet Nam, Report to the Department of Social Insurance, Ministry of Labour,Invalids and Social Affairs of Viet Nam (Bangkok, ILO). United Nations (UN). 2012. Convention on the elimination of all forms of discrimination against women, women’s rights and retirement age in Viet Nam, UN Discussion Paper (Hanoi). Available at: http://un.org.vn/en/publications/doc_download/276-un-discussion-paper-cedaw-womens-rightsand-retirement-age-in-viet-nam.html [14 Aug. 2019]. United Nations Development Programme (UNDP). 2015. Social Assistance in Viet Nam: a review and proposals for reform. —.2018. Human development indices and indicators: 2018 statistical update (New York). Available at: http://hdr.undp.org/sites/default/files/2018_human_development_statistical_update.pdf[14 Aug. 2019]. United Nations Department of Economic and Social Affairs (DESA) Population Division. 2018. World population prospects: The 2017 revision (New York). Available at:https://esa.un.org/unpd/wpp/ DataQuery/ [14Aug. 2019]. University of Groningen. 2013. “Maddison project database”. Available at: www.rug.nl/ggdc/ historicaldevelopment/maddison/releases/maddison-project-database-2013[14 Aug. 2019]. World Bank. 1995. Viet Nam: Poverty assessment and strategy(Washington, DC). —. 2017 “Viet Nam: Overview”. Available at: www.worldbank.org/en/country/Viet Nam/ overview[25Feb. 2019]. World Bank. 2018. “Data bank”. Available at: http://databank.worldbank.org/data/ [14 Aug. 2019]. —. n.d. ASPIRE, Pensions database (Washington, DC).213OLD-AGE INCOME SECURITY IN ASEAN MEMBER STATESPOLICY TRENDS, CHALLENGES AND OPPORTUNITIES214
关注微信
