SOCIAL PROTECTION AND JOBS GLOBAL PRACTICE NOVEMBER 2020 THE MALAYSIA DEVELOPMENT EXPERIENCE SERIES A Silver Lining Productive and Inclusive Aging for Malaysia CONNECT WITH US wbg.org/Malaysia @WorldBankMalaysia @WB_AsiaPacific http://bit.ly/WB_blogsMY NOVEMBER 2020 THE MALAYSIA DEVELOPMENT EXPERIENCE SERIES A Silver Lining Productive and Inclusive Aging for Malaysia SOCIAL PROTECTION AND JOBS GLOBAL PRACTICE © 2020 International Bank for Reconstruction and Development / The World Bank Sasana Kijang, 2 Jalan Dato Onn, Kuala Lumpur 50480, Malaysia Some rights reserved. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and permissions: This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution: Please cite the work as follows: World Bank (2020) ‘A Silver Lining: Productive and Inclusive Aging for Malaysia,’ Washington, DC: The World Bank. Third-party content: The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Table of Contents Acknowledgements 5 List of Figures, Tables and Boxes 6 List of Acronyms 10 Executive Summary 12 Chapter 1. Introduction 22 Chapter 2. Context 32 2.1 Demographic Context 33 Nationwide Demographic Trends 33 Subnational Demographic Trends 39 2.2 Macroeconomic Impacts 42 Impacts on Growth 42 Impacts on Public and Private Savings 46 2.3 Socioeconomic Context 48 Household Composition and Income 48 Incidence and Depth of Poverty 51 Chapter 3. Policy Areas 60 3.1 Employment 61 Labor Market Participation 61 Productive Characteristics and Types of Jobs 71 3.2 Income Security for Older Persons 80 Contributory Social Insurance 80 Non-Contributory Social Assistance 87 3.3 Aged Care 93 Demand, Provision and Financing 93 Quality Assurance and Human Resources 100 Governance of Private Sector Providers 107 Chapter 4. Conclusions and Recommendations 112 References 130 Methodological Annex 134 A. Definitions 135 B. Qualitative Research 136 C. Long-Term Growth Model 137 D. Aged Care Policies, Legislation and Programs 140 E. Additional References 143 A Silver Lining: Productive and Inclusive Aging for Malaysia 3 4 A Silver Lining: Productive and Inclusive Aging for Malaysia Acknowledgements This report was prepared by a team from the World Bank’s Social Protection and Jobs Global Practice led by Achim Schmillen. The report has four chapters. Chapter 1 was prepared by Dewen Wang and Achim Schmillen. For Chapter 2, Section 1 was prepared by Wei Aun Yap, Section 2 by Mahama Samir Bandaogo with inputs from Jorge Guzman Correa, and Section 3 by Kenneth Simler and Zainab Ali Ahmad. For Chapter 3, Section 1 was prepared by Amanina Abdur Rahman with inputs from Alyssa Farha Jasmin, Section 2 by Mark Dorfman, and Section 3 by Dewen Wang and Wei Aun Yap with inputs from Amanina Abdur Rahman. Chapter 4 was prepared by Achim Schmillen with inputs from all team members. Overall guidance was provided by Firas Raad, Philip O’Keefe and Gabriel Demombynes. The authors thank peer reviewers Gustavo Demarco, Johannes Koettl and Veronica Silva as well as Elena Glinskaya, Mei Ling Tan, Norman Loayza and Richard Record for helpful discussions, comments and suggestions, Joshua Foong for leading external communications and facilitating a mission to Perak State, and Li Shen Liew for first-rate administrative support. The authors are grateful to the Department of Statistics, the Ministry of Health and the Ministry of Women, Family and Community Development for data access, to GreenZebras Sdn Bhd for the implementation of qualitative field research, and to Juliana French for drafting a background paper on the findings from this research. Very special thanks are due to all participants in the qualitative field research, who graciously shared their perspective and experiences. Kane Chong and Francis Sim designed the report and its cover, and it was copy-edited by Jeannette Goon and V Ming Chin. During different phases of the project, officials of Malaysia’s Government from the Economic Planning Unit, the Ministry of Health, the Ministry of Human Resources, the Ministry of Women, Family and Community Development, the Employees’ Provident Fund, the Institute for Labour Market Information and Analysis, the Malaysia Productivity Corporation, the Social Security Organization, and the Perak State Executive Council provided invaluable suggestions and feedback. The project benefitted from consultations with the Alzheimer’s Disease Foundation Malaysia, the Association for Residential Aged Care Operators of Malaysia, the Central Welfare Council of Malaysia, the Goldenage Welfare Association Malaysia, Homage, the Malaysian Healthy Aging Society, the Malaysian Research Institute on Ageing, the National Council of Senior Citizens Organizations Malaysia, the Social Wellbeing Research Centre, Vriens & Partners, and the World Health Organization, and from site visits to Greenacres Retirement Village, Pusat Aktiviti Warga Emas Putrajaya, Pusat Jagaan Sri Mesra, Rumah Seri Kenangan Cheras, and Twilight Menglembu Care Centre. A Silver Lining: Productive and Inclusive Aging for Malaysia 5 List of Figures, Tables and Boxes List of Figures Figure 0.1 Transition from aging to aged and super-aged status by country, Years 13 Figure 0.2 Old-age dependency ratio, Percentage 13 Figure 0.3 Average age of retirement by gender and country, Years 16 Figure 0.4 Active EPF members age 54 and estimated pensions, Number and RM 16 Figure 0.5 Share of households with older persons covered by JKM programs and beneficiary incidence, 17 Percentage Figure 0.6 Projected annual net new persons with ADLs limitations or IADLs dependencies, Number 17 Figure 1.1 Malaysia’s aging framework 24 Figure 2.1 Population age 65 and above, Percentage and Number (’000) 33 Figure 2.2 Transition from aging to aged, and super-aged status by country, Years 33 Figure 2.3 Population age 60 and above and 65 and above, Percentage 34 Figure 2.4 Total fertility rate and male and female life expectancy at birth, Percentage and Years 34 Figure 2.5 Prospective and chronological age, Years 37 Figure 2.6 Disability-adjusted old-age dependency ratio, Percentage 37 Figure 2.7 Total and old-age dependency ratios, Percentage 38 Figure 2.8 Change in working age population from 2020 to 2050, Percentage 38 Figure 2.9 Net migration, Numbers (’000) 39 Figure 2.10 GNI per capita and old-age dependency ratio, 2010 US$ and Percentage 39 Figure 2.11 Transition from aging to aged, and aged to super-aged status by ethnicity and gender, Years 40 Figure 2.12 Transition from aging to aged, and aged to super-aged status by state, Years 40 Figure 2.13 Population age 65 and above by state, 2020, Persons 41 Figure 2.14 Population age 65 and above by state, 2040, Persons 41 Figure 2.15 Projected headline and per capita GDP growth, Percentage 43 Figure 2.16 Headline GDP growth by economy, Percentage 43 Figure 2.17 Baseline and alternative projections for population growth rate, Percentage 44 Figure 2.18 Baseline and alternative projections for headline GDP growth, Percentage 44 Figure 2.19 Baseline and alternative projections for ratio of persons in working age to overall population, Percentage 45 Figure 2.20 Baseline and alternative projections for headline GDP growth, Percentage 45 Figure 2.21 Projected spending on pension, Percentage of GDP 46 Figure 2.22 Projected health spending, Percentage of GDP 46 Figure 2.23 Baseline and alternative projections for savings rates, Percentage of GDP 47 Figure 2.24 Baseline and alternative projections for GDP per capita, 2010 US$ 47 Figure 2.25 Older men receiving income by income source and age group, Percentage 50 Figure 2.26 Older women receiving income by income source and age group, Percentage 50 Figure 2.27 Poverty incidence curves for households with and without older persons 52 6 A Silver Lining: Productive and Inclusive Aging for Malaysia List of Figures, Tables and Boxes Figure 2.28 Household composition among households with at least one older person, Percentage 54 Figure 2.29 Older persons living in households with only older persons by age group, Percentage 54 Figure 2.30 Older persons living in households with only older persons by location, Percentage 55 Figure 2.31 Older persons living in households with only older persons by gender, Percentage 55 Figure 2.32 B20 poverty rates among households with and without older persons, Percentage 57 Figure 2.33 B40 poverty rates among households with and without older persons, Percentage 57 Figure 2.34 B20 poverty rates among older persons by age group and gender, Percentage 58 Figure 2.35 B40 poverty rates among older persons by age group and gender, Percentage 58 Figure 3.1 Baseline and alternative projections for growth in working age to population ratio, Percentage 61 Figure 3.2 Baseline and alternative projections for GDP per capita, 2010 US$ 61 Figure 3.3 Life expectancy and age-related labor and pensions policy parameters, Years 62 Figure 3.4 Employment rates of men age 20-24 and 55-64 by district, Percentage 65 Figure 3.5 Employment rates of women age 20-24 and 55-64 by district, Percentage 65 Figure 3.6 Employment rate by age, Percentage 67 Figure 3.7 Employment rate by gender and age, Percentage 67 Figure 3.8 Average age of retirement by gender and country, Years 69 Figure 3.9 Gender gap in employment of workers age 55-64 by country, Percentage points 69 Figure 3.10 Employment rate of men by ethnicity and age, Percentage 69 Figure 3.11 Employment rate of women by ethnicity and age, Percentage 69 Figure 3.12 Employment rate of urban and rural men by age, Percentage 70 Figure 3.13 Employment rate of urban and rural women by age, Percentage 70 Figure 3.14 Share of men by educational attainment and age, Percentage 71 Figure 3.15 Share of women by educational attainment and age, Percentage 71 Figure 3.16 Share of older men employed by educational attainment and skill level, Percentage 73 Figure 3.17 Share of older women employed by educational attainment and skill level, Percentage 73 Figure 3.18 Share of men employed by sector and age, Percentage 73 Figure 3.19 Share of women employed by sector and age, Percentage 73 Figure 3.20 Share of workers by status in employment, gender and age, Percentage 75 Figure 3.21 Share of part-time workers by gender and age, Percentage 75 Figure 3.22 Average weekly working hours of part-time workers by age, Number 76 Figure 3.23 Part-time workers willing to work longer hours by age, Percentage 76 Figure 3.24 Monthly employment income by gender and age conditional on workers’ education, RM 78 Figure 3.25 Difference in average monthly employment income between workers aged 40 to 49 and 50 to 59, 78 Percentage Figure 3.26 Average accumulated employment income from age 20 to 54/59 by location and gender, RM 79 Figure 3.27 Average accumulated employment income from age 20 to 54/59 by education and gender, RM 79 Figure 3.28 Stylized overview of Malaysia’s social protection system 80 A Silver Lining: Productive and Inclusive Aging for Malaysia 7 List of Figures, Tables and Boxes Figure 3.29 Coverage of retirement savings institutions, Share of labor force 82 Figure 3.30 Share of own-account workers among all workers by gender, Percentage 82 Figure 3.31 Active EPF members age 54 and estimated pensions, Number and RM 84 Figure 3.32 Simulated replacement rates by EPF minimum withdrawal age, Percentage 84 Figure 3.33 Labor force covered by retirement savings institutions, Percentage 86 Figure 3.34 Working age individuals contributing to EPF and average contributions per capita per month, 86 Percentage and RM Figure 3.35 Beneficiaries of JKM social assistance programs, Number (‘000) 88 Figure 3.36 Expenditures on JKM social assistance programs, RM (million) 88 Figure 3.37 Share of households with older persons covered by JKM programs and beneficiary incidence, 89 Percentage Figure 3.38 Illustration of the tradeoffs in policy parameters for social pension 89 Figure 3.39 Simulated fiscal cost of social pension, Percentage of revenue 92 Figure 3.40 Adequacy of social pensions for B20, Percentage of average income 92 Figure 3.41 Projected net annual increase in older persons by age group, Number and Percentage 94 Figure 3.42 Projected annual net new persons with ADLs limitations and IADLs dependencies, Number 94 Figure 3.43 Building blocks of an inclusive aged care system 95 Figure 3.44 RSK residents, Number 97 Figure 3.45 Licensed private residential aged care centers by state, Number and Number per population age 65 97 and above Figure 3.46 Public spending on aged care by year, RM (million) 99 Figure 3.47 Spending on aged care by country, Percentage of GDP 99 Figure 3.48 Aged care workers by occupation, Number (’000) 103 Figure 3.49 Aged care workers by occupation, Percentage of total employment 103 Figure 3.50 Aged care workers by occupation and age group, Percentage 103 Figure 3.51 Aged care workers by occupation and gender, Percentage 103 Figure 3.52 Aged care workers by occupation and education, Percentage 105 Figure 3.53 Aged care workers by occupation and citizenship, Percentage 105 Figure 3.54 Typical licensing process for a private residential aged care center 109 Figure 4.1 Stylized representation of coverage and adequacy of old-age income protection based on current 117 setup Figure 4.2 Stylized representation of coverage and adequacy of old-age income protection based on integrated 117 vision Figure 4.3 Female labor force participation rate by country and country group, Percentage 119 Figure 4.4 Baseline and alternative projections for female labor force participation rate, Percentage 119 Figure 4.5 Baseline and alternative projections for growth in human capital, Percentage 119 Figure 4.6 Baseline and alternative projections for headline GDP growth rate, Percentage 119 Figure 4.7 Simulated EPF balances by age under phased withdrawal, RM 124 8 A Silver Lining: Productive and Inclusive Aging for Malaysia List of Figures, Tables and Boxes List of Tables Table 0.1 Contributors to decline in headline GDP growth, Percentage 14 Table 0.2 Poverty rates for households with and without older persons, Percentage 14 Table 0.3 Overview of policy recommendations 20 Table 1.1 Challenges and opportunities of aging 24 Table 1.2 Aging policies in select East Asian countries 28 Table 2.1 Projected headline GDP growth and drivers of growth 44 Table 2.2 Composition of households with at least one older person, Percentage 49 Table 2.3 Composition of households with at least one older person by age of household head, Percentage 49 Table 2.4 Poverty rates for those living in households with and without older persons, Percentage 51 Table 2.5 Poverty rates for households headed by older and non-older persons, Percentage 53 Table 2.6 Poverty rates by household composition, Percentage 53 Table 2.7 Poverty rates for households headed by older persons by location, Percentage 59 Table 2.8 Poverty rates for households headed by older persons by ethnicity, Percentage 59 Table 2.9 Poverty gaps for households with and without older persons, Percentage 59 Table 3.1 Employment rate of persons age 50 to 74 by country 63 Table 3.2 Types of aged care provision in Malaysia 95 Table 3.3 Number of aged care facilities and beneficiaries 96 Table 4.1 Overview of policy recommendations 114 Table 4.2 Possible transition phasing 124 Table A.1 Definitions 132 List of Boxes Box 1.1 The silver economy 26 Box 2.1 Measuring aging 36 Box 2.2 Trends in co-residence, living arrangements and poverty among older persons 54 Box 3.1 The lump of labor fallacy 64 Box 3.2 Distinguishing between age and cohort effects 68 Box 3.3 The agricultural labor market 74 Box 3.4 COVID-19 and the protection of older persons 101 Box 3.5 Licensing of private residential aged care centers 108 Box 3.6 Developing public-private partnerships (PPP) in the aged care sector 111 Box 4.1 Mitigating the macroeconomic impacts of aging 118 Box 4.2 Expanding pension coverage by offering matching contributions 122 Box 4.3 Using needs assessments to determine access to and scope of aged care services 128 A Silver Lining: Productive and Inclusive Aging for Malaysia 9 List of Acronyms ADLs Activities of Daily Living ALMPs Active Labor Market Policies ASEAN Association of Southeast Asian Nations B20 Bottom 20 B40 Bottom 40 BA Public Assistance (Bantuan Am) BOT Financial Assistance for Older Persons (Bantuan Orang Tua) BSH Cost of Living Aid (Bantuan Sara Hidup) DOSM Department of Statistics Malaysia EIS Employment Insurance System EPF Employees Provident Fund GDP Gross Domestic Product GNI Gross National Income HRDF Human Resource Development Fund IADLs Instrumental Activities of Daily Living IMF International Monetary Fund JKM Department of Social Welfare (Jabatan Kebajikan Masyarakat) JPK Department of Skills Development (Jabatan Pembangunan Kemahiran) KWAP Public Service Pension Fund (Kumpulan Wang Persaraan) LTAT Armed Forces Pension Board Fund (Lembaga Tabung Angkatan Tentera) LTGM Long-Term Growth Model MOF Ministry of Finance MOH Ministry of Health MOHR Ministry of Human Resources MySPC Malaysia Social Protection Council NCDs Noncommunicable diseases NGO Non-Governmental Organization NOSS National Occupational Skills Standard NRI Natural Rate of Interest OECD Organization for Economic Co-operation and Development PAWE Activity Centre for Older Persons (Pusat Aktiviti Warga Emas) PLI Poverty Line Income 10 A Silver Lining: Productive and Inclusive Aging for Malaysia List of Acronyms PPP Public Private Partnership PRS Private Retirement Scheme PWDs Persons with Disabilities RE Rumah Ehsan RLE Remaining Life Expectancy RM Ringgit Malaysia RS Rumah Sejahtera RSK Rumah Seri Kenangan RWT Rumah Warga Tua SOCSO Social Security Organization TFP Total Factor Productivity UN United Nations UN WPP United Nations World Population Prospects UPWE Senior Citizens Care Unit (Unit Penyayang Warga Emas) US$ United States Dollar A Silver Lining: Productive and Inclusive Aging for Malaysia 11 Executive Summary 12 A Silver Lining: Productive and Inclusive Aging for Malaysia Executive Summary Introduction and Context In 2020, Malaysia passes a crucial milestone in its demographic trajectory and becomes an aging society. Driven by a precipitous decline in fertility accompanied by a sustained rise in life expectancy, in recent years Malaysia has seen an uptick in the pace of demographic change. As a result, in 2020 it will transition to an aging society which is defined per the international convention as having 7 percent or more of the population age 65 and above (World Bank 2016). Henceforth, Malaysia will age rapidly. After 24 years, it will become an aged society, with the share of the population age 65 and above reaching 14 percent. Only 12 years later, the country will become super-aged, with the share reaching 20 percent (see Figure 0.1). Malaysia’s transition from an aging to an aged society will happen at a similar pace to Japan’s, whereas it took 115 years in France, 73 years in Australia and 69 years in the United States. In parallel, the old-age dependency ratio—the number of persons over the age of 65 compared with the population age 15 to 64—is projected to rise substantially from 10.4 percent in 2020 to 21.7 percent in 2040 (see Figure 0.2), accompanied by a net increase in the number of older persons by 130,000 to 210,000 per year. Although life expectancy has increased, not all of the increased years of life are spent in good health. Malaysia’s healthy life expectancy at birth is estimated at 66.6 years in 2016, about a decade less than life expectancy. This divergence is partly due to the rising prevalence of noncommunicable diseases (NCDs). For example, the prevalence of diabetes among adults increased from 11.2 percent in 2011 to 18.3 percent in 2019. FIGURE 0.1: Transition from aging to aged and FIGURE 0.2: Old-age dependency ratio, super-aged status by country, Years Percentage Lao PDR 25 Cambodia Philippines Myanmar 20 Brunei Darussalam Old-age dependency ratio (%) Indonesia Malaysia Vietnam 15 Singapore Thailand Turkey 10 China Republic of Korea Japan Australia 5 France United Kingdom 1950 2000 2050 0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 Aging duration Super-aging duration Source: Authors’ calculations based on UN WPP 2019. Source: UN WPP 2019. Note: Data for Australia, France and the United Kingdom are truncated as these were aging societies even before 1950. Blue bars represent the transition period from aging to aged. Orange bars represent the transition period from aged to super-aged. Rapid aging will be one of the most crucial megatrends affecting Malaysia in coming decades, raising policy challenges in areas such as employment, income security, health care, and aged care. According to macroeconomic projections undertaken for this report, aggregate economic growth in Malaysia is expected to decline from 4.5 percent in 2019 to about 1.8 percent by 2050 and demographic changes A Silver Lining: Productive and Inclusive Aging for Malaysia 13 Executive Summary in the form of slower population growth and a shrinking working-age to total population ratio will account for about a third of the decline (see Table 0.1). Equally important, aging threatens to increase poverty and vulnerability among older persons, especially those who are frail and in need of aged care and long-term care services. Today, poverty rates among this population group are only slightly higher than among the population as a whole—using the bottom 20 (B20) per capita income threshold as the poverty line. In 2016, the poverty rate was 21 percent for households with older persons and 19.7 percent for households without them (see Table 0.2). However, the poverty incidence for older persons rises with age and there are concerns that the vulnerability of older persons will increase as the total dependency ratio increases and families might accordingly struggle to cope. Furthermore, without strengthened public health care, the prevalence of disabilities accumulated over a life span will further increase care needs. In addition, more and more older persons are living alone, and there are concerns that household incomes might become more volatile both globally and in Malaysia. All these challenges are becoming more pressing as Malaysia is likely to transition from an upper middle-income to a high-income economy within the next few years, despite a COVID-19 induced recession in 2020, and the aspirations of its expanding middle class regarding jobs, income, and well-being continue to grow (see World Bank forthcoming). TABLE 0.1: Contributors to decline in headline TABLE 0.2: Poverty rates for households with GDP growth, Percentage and without older persons, Percentage Contributor Share (%) Household B20 poverty B40 poverty type line line Population growth 19 Growth in total dependency ratio 12 No older 19.7 39.3 persons present Total factor productivity growth 12 Human capital growth 2 Older persons 21.0 42.3 present Public invest. (falling effectiveness) 43 Private capital to output ratio (falling All households 20.0 40.0 12 effectiveness) Baseline 100 Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on LTGM. Amenities Survey 2016 (DOSM). The policy challenges created by aging should not be underestimated, but neither should the opportunities of a silver economy. Harnessing the productive potential of persons age 50 and above can help mitigate the growth effects of aging and address talent shortages while the aged care sector can become a new driver for economic growth, job creation, improved social services, and a better quality of life. In addition, recent studies have highlighted the economic potential offered by an aging society. In this context, the term silver economy has been coined to encompass all sources of opportunities that arise from economic activities to serve the needs and demands of older persons. Studies from high income countries demonstrate that older persons can potentially be relatively wealthy consumers compared with young generations. Meeting the increasing needs and demands of older persons will lead to the creation of new markets and business opportunities to spur growth, create jobs, and increase tax revenues. While rapid aging is crucially important and will become even more so in coming decades, through the right choice of policies, the Government of Malaysia can help the country achieve productive and inclusive aging. Aware of both the challenges and opportunities of aging, relevant policy documents already 14 A Silver Lining: Productive and Inclusive Aging for Malaysia Executive Summary reflect inclusive and productive, as well as healthy aging as key policy objectives. In 1995, the National Policy for the Elderly was Malaysia’s first strategic policy on aging. To implement this policy, the National Advisory and Consultative Council for Older Persons was set up in 1996 and a Plan of Action for the Older Persons was formulated in 1998. Based on a review of the first strategic policy and the Plan of Action, the National Policy for Older Persons and the Plan of Action for Older Persons were introduced in 2011 with the goal to allow older persons to have a high sense of self-worth and dignity through development, health and well-being, as well as an enabling and supportive environment. More recently, the 11th Malaysia Plan has included initiatives to improve the supportive environment for older persons and to promote active aging while the Mid-Term Review of the Plan has emphasized the need to enhance older persons’ living environments. The COVID-19 pandemic has resulted in an unprecedented crisis with an enormous health and human toll, as well as exacerbated many of the policy challenges raised by aging both in the short term and more structurally. There are concerns that the COVID-19 crisis is creating a “new normal” for the care, income support and employment of older persons. COVID-19 infections in several aged care homes in Malaysia, including in unlicensed homes, show that older persons living in institutional settings are an exceptionally vulnerable population group. They also demonstrate the importance of service standards, in particular, health and safety standards, as part of a broader response to protect this group. In addition, the crisis may exacerbate the economic vulnerability of older persons while reducing their savings for retirement. This reduction is due to declining contributions to the Employees’ Provident Fund (EPF)—the retirement savings institution with the largest coverage in Malaysia—as well as withdrawals from EPF accounts and declines in asset prices. Finally, the very elevated rates of unemployment and underemployment due to the crisis lend urgency to policies that allow workers who have lost their employment, including older workers, to return to work. Recognizing that there is no evidence that increased employment among late-career workers negatively affects the employment prospects of younger workers, required policies might include the provision of enhanced and tailored opportunities for upskilling and reskilling that take into account both temporary and potentially permanent changes in labor demand brought about by the crisis. The report analyzes Malaysia’s demographic, socioeconomic and macroeconomic contexts, as well as three select key policy areas where critical knowledge gaps exist–employment, income security, and aged care. Although the report recognizes that aging cuts across multiple sectors and requires a life cycle perspective, not all sectors are comprehensively addressed. Nutrition, basic education, and health care are notable omissions, not because these are any less important than employment, income security, and aged care, but because the focus of the report is on select policy areas with large knowledge gaps. Similarly, the report recognizes that aging is a cross-cutting and multi-dimensional issue with productive aging, inclusive aging and healthy aging as key, equally important dimensions, but does not discuss healthy aging in detail. Methodologically, the report relies on a mixed methods approach that combines quantitative, qualitative and institutional research. The quantitative research is conducted using recent, nationally representative data from the Department of Statistics Malaysia in addition to international data sources. Complementing the quantitative investigation, qualitative and institutional analyses are used to improve the understanding of the constraints behind the most significant challenges identified by quantitative data. The qualitative analysis focuses on aged care and employs semi-structured interviews with diverse stakeholders to answer questions regarding the roles and responsibilities of older persons, families, communities, NGOs, the private sector, and the Government. The interviews also take into account caregivers’ and older persons’ experiences, preferences, ideas, concerns, and expectations. The institutional analysis encompasses a desk review of relevant laws, policies, and regulations. A Silver Lining: Productive and Inclusive Aging for Malaysia 15 Executive Summary Findings Compared to other upper middle- and high-income countries, the employment rate of those age 50 to 74 in Malaysia is low, especially among women, and some persons in this age group struggle to find flexible, productive and inclusive employment. In Malaysia, the employment rate—calculated as the share of persons employed at a given age—for persons age 50 to 74 stands at 45.9 percent, which is lower than in most other upper middle- and high-income countries. The differences between Malaysia and comparator countries are, to a large extent, driven by Malaysia’s relatively low employment rates for persons between age 50 and 64, as well as for women (see Figure 0.3). In turn, the low employment rate among persons between age 50 and 64 is arguably at least partly driven by the relatively low minimum retirement age and low EPF minimum withdrawal age while women face constraints related to childcare, the legal environment, and gender norms and attitudes. As workers age, prevalent types of employment also change. Both self-employment and part-time work are relatively more common among older workers. Outside of agriculture, 19 percent of employed men and women age 40 to 44 are self-employed, but this is the case for 43.1 percent of employed men and half of employed women age 60 to 64. While these changes in the types of jobs are arguably at least partly due to workers’ preferences, there is clear room for an improved policy environment to allow more of them to find flexible, productive and inclusive employment. Also of note, there is no evidence that increased employment among older workers negatively affects the employment prospects of younger workers either in Malaysia or internationally. FIGURE 0.3: Average age of retirement by FIGURE 0.4: Active EPF members age 54 and gender and country, Years estimated pensions, Number and RM 74 25,000 4,500 72 4,000 Estimated monthly pension (RM) Active EPF members age 54 20,000 3,500 70 Average age of retirement 68 3,000 15,000 66 2,500 64 2,000 10,000 1,500 62 5,000 1,000 60 500 58 0 0 56 less than 50 50-100 100-150 150-200 200-250 250-500 500-1,000 1,000+ 54 Malaysia Germany United Kingdom OECD 2018 Canada Australia Turkey United States Chile Mexico Japan Republic of Korea EPF account balances (RM'000) Number of active members age 54 Male Female Indexed annuity (amount per month) Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on EPF. Amenities Survey 2016 (DOSM) and OECD. Adequacy and coverage of Malaysia’s system of contributory social insurance and non-contributory social assistance for older persons are major challenges. EPF as Malaysia’s main private-sector retirement savings scheme is a defined-contribution provident fund. Seventy percent of EPF contributions are placed in Account 1, dedicated to retirement savings, while the remaining 30 percent goes into Account 2, which can be used for broader purposes. The reliance on a defined-contribution provident fund places Malaysia in the relatively unusual situation that—outside of the civil service pension system—institutions providing 16 A Silver Lining: Productive and Inclusive Aging for Malaysia Executive Summary old-age income security pose no fiscal risks. At the same time, in the absence of a broad non-contributory social assistance program for older persons, almost all old-age income security risks rest with these persons and their families. In fact, at 60.8 percent, participation in contributory retirement savings institutions is low, especially when compared to an aspirational peer group of high-income countries. Coverage is especially low among lower-income households, with less than a fifth of working-age B20 being active EPF contributors. In addition, the majority of EPF contributors will only receive very low benefits in retirement, because participation in covered employment is intermittent and the minimum withdrawal age is low. Almost three quarters of workers at age 54 have balances of under RM250,000 (see Figure 0.4). Translated into an indexed annuity, almost three quarters of workers will have a monthly benefit of less than RM1,050, only slightly more than the poverty line income (PLI) of RM980 before its recent revision. As most older persons in Malaysia have inadequate EPF benefits or none at all, it falls either to families or to very limited non- contributory social assistance to protect older persons against poverty and destitution. The Department of Social Welfare (Jabatan Kebajikan Masyarakat or JKM) provides cash assistance for poor and vulnerable older persons through the Financial Assistance for Older Persons (Bantuan Orang Tua or BOT) program, among others. Although JKM programs are targeted toward the poor, their budget allocation and coverage are both small. Only slightly more than a quarter of households with members age 65 and older in the B20 receive assistance from JKM (see Figure 0.5). Rapid aging accompanied by a changing socioeconomic context and increasing prevalence of NCDs will threaten existing aged care arrangements. As in many other countries in East Asia, familial provision of aged care has long played a critical role in Malaysia whereby frail older persons get support and care from their spouses, children, and other relatives. However, this traditional arrangement shows signs of weakening as the typical household structure changes from large, multi-generational households toward small nuclear ones. While co-residence remains common, the share of three-generation households declined from 41.1 percent in 2004 to 30.7 percent in 2016. The decline in co-residence may affect households’ ability to informally provide care to older household members. In parallel, there are more and more persons who FIGURE 0.5: Share of households with older FIGURE 0.6: Projected annual net new persons persons covered by JKM programs and with ADLs limitations or IADLs dependencies, beneficiary incidence, Percentage Number 40 140,000 Net new persons with functional limitations to ADLs Share of older persons receiving JKM transfers 35 120,000 and beneficiary incidence (%) 30 100,000 and dependency in IADLs 25 80,000 20 15 60,000 10 40,000 5 20,000 0 1 2 3 4 5 6 7 8 9 10 0 Income deciles 2020 2025 2030 2035 2040 2045 2050 Share of households with older persons receiving JKM transfers (%) Bene ciary incidence (%) Functional limitation to ADLs Dependency in IADLs Source: Authors’ calculations based on Household Income and Basic Source: Authors based on UN WPP. Amenities Survey 2016 (DOSM). A Silver Lining: Productive and Inclusive Aging for Malaysia 17 Executive Summary have a high probability of having functional limitations in activities of daily living (ADLs) and dependency in instrumental activities of daily living (IADLs).1 The number of persons with ADLs functional limitations is projected to more than double from 2020 to 2040 while the number of persons with IADLs dependency is projected to increase by about one million per decade (see Figure 0.6), reflecting the lifelong impact of a high prevalence of NCDs in addition to demographic trends. Already today, an estimated 250,000 older persons have dementia while many care needs are unmet—according to the National Health and Morbidity Survey 2018, 30.8 percent of those age 60 and above are malnourished. On top of that, the survey also found that 30.8 percent of those in that same age group have poor social support. Meanwhile, in terms of formal provision, institutional care is very limited and home- or community-based care is still at an early stage of development. Public financing of aged care is very limited as well and weighted toward institutional care with relatively low quality of services for a small number of destitute older persons. Coverage of private institutional care is low and uneven across space. While there are some experiments with a focus on urban areas to explore viable business and service delivery models for home, community-based and institutional care, there are also significant challenges with regard to service quality, human resources, and governance arrangements. Aging poses particular challenges for women’s employment, income security and aged care needs. Across all age groups, employment rates are significantly lower for women than for men. The gap is largest between age 50 and 60 as women on average retire earlier. In 2016, 59.7 percent of men age 60 but only 17.9 percent of women age 60 were employed. Arguably, gender differences in employment rates are due to constraints related to childcare, the legal environment, and norms and attitudes (see World Bank 2019a). Women also have both lower rates of EPF coverage and lower balances. In addition to the low employment rates among women, this is because a relatively large share of women who participate in the labor market is self-employed or engaged as unpaid family workers and thus often not covered by social insurance. Moreover, women who are covered by social insurance on average have lower EPF balances than men, partly due to the prevailing gender wage gap. As a result, the average EPF balance at age 54 is RM177,000 for women and RM233,000 for men. With regard to aged care, since women on average live longer, they have particularly pronounced care needs. In addition, the large majority of aged care workers are women. This is the case for 55.8 percent, 77.4 percent, and 93.6 percent of personal care workers, health workers, and domestic workers, respectively. Thus, strengthened arrangements for training and qualifications to upgrade skills of aged care workers can have particularly beneficial effects for women. Policy Options An integrated, interagency policy approach can address challenges and grasp opportunities in a systematic and mutually reinforcing way. Reducing some of the barriers to productive and inclusive aging will require clear prioritization and finely calibrated policy approaches. At the same time, many of the recommended policy approaches to achieve productive and inclusive aging will address more than one barrier and different approaches will be mutually reinforcing and can create opportunities for a silver economy. For instance, an increase in the minimum retirement age to 65 can not only improve employment opportunities for older workers, but also foster old-age income security. In addition to that, more productive and inclusive aging is certainly a cross-sectoral endeavor; some of the most crucial policy approaches such as strengthening the governance of the aged care sector will necessitate cooperation between different 1 ADLs correspond to essential elements of self-care such as feeding, toileting, mobility, dressing, and cleaning, and hence limitations raise the need for substantive care services. IADLs are instrumental to enabling independent living—such as food preparation, shopping, housekeeping, and managing finances—which may not be required daily and necessitate lower intensity care services such home help. 18 A Silver Lining: Productive and Inclusive Aging for Malaysia Executive Summary ministries and agencies. Therefore, a systematic and integrated interagency approach to the promotion of productive and inclusive aging—as well as one that adopts a life cycle perspective, mainstreams approaches to address the specific constraints faced by women, and is mindful of political economy considerations—will have the most pronounced and sustainable impact. Ideally, this approach would be guided by an integrated strategy and supervised by a steering committee. The promotion of productive and inclusive aging will require policies that foster the productive employment of all workers, provide minimum income protection to all older persons, and build an inclusive aged care system. General policy directions as well specific short-term, medium-term and long- term policy options are laid out in Table 0.3, with the caveat that this distinction is not always entirely clear- cut and that for some of the long-term policy options to be effective, implantation needs to begin in the short or medium term. In the future, Malaysians will have to work longer in order to ensure adequate financial protection in old age. But they will also be more able to do so, being healthier for longer, with less physically demanding occupations, and with more digitally-enabled work places. As in nearly all high-income countries, longer working lives will in turn require gradual adjustments to the minimum retirement age in line with increasing longevity, and in the EPF minimum withdrawal age, among other policy parameters. In parallel, policies are needed that foster workers’ productive employment—such as enhanced opportunities for training and lifelong learning. In addition, the provision of minimum income protection of older workers will require improvements in the coverage and adequacy of social insurance schemes, as well as the introduction of a modest, broadly targeted tax-financed social pension. Finally, for the aged care sector to become a new driver of economic growth, it will be crucially important to create an enabling market and regulatory environment for private aged care provision, to strengthen the governance of the sector, and to selectively increase public financing in line with available fiscal space. A Silver Lining: Productive and Inclusive Aging for Malaysia 19 Executive Summary TABLE 0.3: Overview of policy recommendations General Policy Short-term Medium-term Long-term Directions Policy Options Policy Options Policy Options Gradually raise the min. Provide enhanced Develop regulations for retirement age to 65; link opportunities for training older workers’ productive 1 it to life expectancy at and lifelong learning. and flexible employment. Foster the productive retirement thereafter. employment of all Improve women’s Consider piloting workers, including Facilitate the adoption economic opportunities incentives to encourage older workers. of age-management through better childcare the employment of older strategies by companies. and other measures. workers. Adjust the minimum Develop an integrated Introduce a modest, withdrawal age for EPF vision of a system for old- broadly targeted tax- Account 1 and rethink the 2 age income protection. financed social pension. role of Account 2. Provide minimum income protection to Broaden EPF coverage to Explore annuitization Establish a one-stop shop all older persons. more of the self-employed options and longevity for EPF, SOCSO and HRDF and informal sector insurance for EPF contributions. workers. balances. Develop a systematic Upgrade skills of aged Streamline and harmonize and actionable aged care care workers through licensing requirements for strategy based on a solid strengthened training aged care facilities. 3 assessment. arrangements. Build an inclusive aged Strategically reorient Selectively increase public care system. Improve quality standards public financing toward financing to incentivize of aged care homes to home and community- market-oriented care ensure health and safety. based aged care. provision. POLICY DIRECTION 1 Foster the productive employment of all workers, including older workers, through a comprehensive set of demand- and supply-side policies. A first policy option to foster productive employment in the short term is to provide enhanced opportunities for training and lifelong learning that take into account the specific circumstances of older workers, such as their generally relatively low level of education. In addition, policies could encourage and facilitate the adoption of age-management strategies encompassing work organization, work equipment, working time policies, and training, among other aspects. In the medium term, it is recommended to develop a regulatory framework for the productive and flexible employment of older workers and to facilitate part-time and other flexible forms of employment. In order not to exacerbate existing gender imbalances in Malaysia’s labor market, it will also be important to address women’s specific constraints to labor market participation. Relevant initiatives include better availability, quality and affordability of childcare, reforms of the legal environment and improved support for parents in line with international legal norms, and policies that address gender norms and attitudes (see World Bank 2019b). A long-term policy option is to gradually increase the minimum retirement age to 65, and thereafter link it to life expectancy. Macroeconomic projections indicate that increasing the retirement age to 65 could raise GDP growth by 0.3 percentage points per year while a review of the relevant international experience and data for Malaysia finds no evidence that increased employment among older workers negatively affects the employment prospects of younger workers. Also in the long term, these initiatives could be coupled with the piloting of financial incentives to encourage the employment of older workers—such as targeted, conditional, and time-bound wage subsidies. 20 A Silver Lining: Productive and Inclusive Aging for Malaysia Executive Summary POLICY DIRECTION 2 Provide minimum income protection to all older persons through increased coverage and adequacy of savings and social insurance schemes, as well as a modest, broadly targeted tax- financed social pension. In the short term, an integrated vision of a system for old-age income security that provides minimum protection to all older persons and policy measures in the areas of both social insurance and social assistance can be developed. With regard to social insurance, proactive measures can improve the coverage and adequacy of relevant schemes, including EPF retirement savings and the Social Security Organisation’s work injury, disability and survivorship insurance schemes. Several measures could increase the proportion of workers actively contributing to EPF and ensure that retirement savings could be sustained for longer. To increase coverage, oversight could be improved, including through a requirement for the registration of all workers as a condition for the granting of business licenses and/or government contracts. To raise average balances, in the long term the minimum withdrawal age for EPF could gradually be increased to 65 through a well-considered transition process, for instance over a period of 20 years (which of course would need to begin in the short or medium term). In addition, consideration could be given to converting all contributions to EPF wholly into retirement savings, to transitioning to phased withdrawals of EPF balances, and to exploring longevity insurance and annuitization options. Over time, the EPF contribution rate could be reduced as the benefit eligibility age increases, provided that there is an observed increase in coverage and adequacy. However, even with all these measures it is unlikely that social insurance schemes can ever cover the entire labor force. Thus, a modest, broadly targeted tax-financed social pension may also be required. In the short or medium term, one practical option would be to introduce a social pension of RM350 per month targeted at the B40 age 65 and over and using the BSH implementation system. POLICY DIRECTION 3 Build an inclusive aged care system with an enabling market and regulatory environment for private not-for-profit provision, strengthened governance, and selectively increased public financing. In the short term, the development of an inclusive aged care system could be guided by a systematic and actionable aged care strategy which in turn could be based on a comprehensive assessment and diagnostic. In light of the COVID-19 pandemic, it will be particularly crucial to continuously improve the aged care infrastructure and service standards to ensure the health and safety of residents of aged care homes. In the medium term, institutional barriers could be removed and licensing requirements and processes for existing unlicensed aged care facilities and potential new market entrants could be streamlined and harmonized. Furthermore, the Government’s role has already begun to shift from a “supplier and provider” to a “purchaser and regulator” of aged care services. However, this process could be accelerated, including through the establishment of public-private partnerships, and through shifting to results- instead of input- based commissioning. In the long term, it will also be important to strengthen arrangements for training and qualifications to upgrade the skills of aged care workers. For publicly financed aged care services, an approach that combines needs assessments and means tests to determine the eligibility of older persons could be considered. Finally, it is recommended to increase and strategically reorient public financing for aged care services. In this context, public financing could shift toward home and community-based aged care, address imbalances across space, mobilize private and social investments, and extend the coverage of low-income households. A Silver Lining: Productive and Inclusive Aging for Malaysia 21 CHAPTER 1 Introduction 22 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 1: Introduction As the result of a profound development process that has included social and economic progress, health improvements, and technological advances, Malaysia is becoming an aging society in 2020. Since the formation of today’s Malaysia in 1963, the country has been transformed from a low-income economy dominated by agriculture and the exploitation of natural resources to a country at the cusp of high-income status with a strong and diversified manufacturing and services sector (see Abdur Rahman and Schmillen forthcoming). This development process has led to widespread though uneven income growth, and radically improved nutrition, health care and social service provision. Together with technological advances and changing social norms and attitudes, it has also led to a significant decline in fertility rates and increases in life expectancy. The average number of children born to a woman over her lifetime for the period from 1960 to 1964 was 6.4 but is two now (and continuing to fall), while life expectancy at birth has increased by more than two decades since the 1950s. In turn, the share of the population age 65 and above has increased from 3.4 percent in 1970 to 7.2 percent in 2020. This means Malaysia is now following the trajectory of many other countries that have undergone an economic transformation from low- to upper middle- or high-income status and a demographic transition from high fertility and mortality to low fertility and mortality. As a result, it is becoming an aging society according to the common definition of having more than 7 percent of the population age 65 and above.2 A balanced view of aging recognizes both the challenges and the opportunities that this process will present. Aging is not simply a change in a country’s demographic structure but also causes changes in social and economic structures, which will then affect economic growth and the conditions of families and communities. No doubt, in Malaysia as elsewhere, aging will bring significant challenges. In terms of macroeconomic impacts, there are concerns that aging may have a negative impact on long-term growth due to a shrinking labor force, lower public savings, and potentially slower technological adoption. In addition, due to aging a higher proportion of older persons may need public income support after they exit the labor market and more health care and social care services when they get frailer and more dependent. While these challenges should not be underestimated, a balanced approach recognizes that aging can also offer promising opportunities—this is the “silver lining” alluded to in the title of this report. For instance, harnessing the productive potential of older workers can help mitigate the growth effects of aging and address talent shortages. Many aging societies have promoted the development of the aged care sector as a new driver for economic growth, job creation, improved social services, and a better quality of life. Moreover, welfare improvements of older persons can also generate positive externalities for the whole society. For instance, they can create opportunities for older persons to take care of their grandchildren and family members, or to become volunteers. This will in turn promote social cohesion and social solidarity. Aging is not only about “older persons” or about “the old” versus “the young.” Rather, to manage aging requires a life cycle perspective on policy design that recognizes the importance and consequences of behavioral change across the life cycle of individuals (see World Bank 2016). For example, the health of children affects their health as adults, saving for old age needs to start early, and the burden of taxation to provide services and benefits needs to be spread fairly across generations. In turn, better labor market prospects for older workers can improve a society’s overall productivity and tax intake; better health promotion, primary health care services and health management can reduce morbidity rates and the prevalence of non-communicable diseases (NCDs) and other diseases, among older persons, and thus lower health expenditures; and better quality and more accessible aged care services can improve the quality of life of older persons and create jobs for younger ones. 2 For a list of working definitions of technical terms used in this report, see Annex A. A Silver Lining: Productive and Inclusive Aging for Malaysia 23 Chapter 1: Introduction The challenges and opportunities of aging for individuals, the economy and society are often two sides of the same coin. For instance, as people live longer, questions about how they can stay healthy and financially secure need to be answered. However, if the answers to these questions can be found, older persons can then make meaningful contributions to the well-being of their families and society more broadly. Thus, smart policies will consider a balanced approach that examines both the challenges and opportunities of aging to reduce costs and maximize benefits (see Table 1.1): • From a demographic perspective, aging can either lead to financial or care burdens or increase the number of healthy and productive persons in a society. On the one hand, the aging process increases a society’s average age and the old-age dependency ratio, defined as the number of persons over age 65, compared with the working-age population age 15 to 64. This can cause financial, care and other burdens to families and society. On the other hand, with improvements in health and aged care provision, more older persons will live healthily and independently. With the right labor and skills policies, they can also have longer and more productive working lives and productively contribute to the economy and society (see Lee and Mason 2006). • From an economic perspective, aging may have a negative impact on long-term growth but may also nurture a silver economy. According to canonical growth models, aging leads to a shrinking labor force and this in turn reduces economic growth. While this effect may to some extent be offset by human capital accumulation, aging may potentially also slow down economic growth due to its negative impacts on public savings due to increased health- and pension-related spending (whereas the impacts on TABLE 1.1: Challenges and opportunities FIGURE 1.1: Malaysia’s aging framework of aging Dimensions Challenges Opportunities Demography • Disabilities and • Healthier Economy dementia longevity and (Productive self-fulfillment aging) • Demographic debt • Demographic dividend3 Environment Health Economy • Smaller • Human capital (Supportive (Healthy workforce accumulation aging) aging) Wellbeing • Increasing • Silver economy of older public spending persons • Job creation • Lower economic and tax growth contributions Society • Old-age • Resources and Spiritual Social poverty social capital (Positive (Active aging) aging) • Burden to • Social families and contributions communities and cohesion Source: Authors. Source: Authors based on National Policy for Older Persons. 3 The concept of demographic dividend was originally introduced by the United Nations Population Fund and refers to the economic growth potential that can result from shifts in the population age structure when the proportion of working age people in the total population is high, which offers opportunities because relatively more people are productive and can contribute to growth (see Section 2.2). Lee and Mason (2006) further developed the concept by differentiating between a first and a second demographic dividend, where the second demographic dividend comes mainly from augmented human capital accumulation and can thus be pronounced relatively late in a society’s aging process. 24 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 1: Introduction private savings are theoretically ambiguous). Indeed, many empirical studies for both the United States and Europe have found that economic growth slows roughly one-to-one with declines in labor force or population growth (see Lee and Mason 2017). Concurrently, there are other recent studies that have highlighted the economic potential offered by an aging society. In this context, the term silver economy has been coined to encompass all sources of opportunities that arise from economic activities to serve the needs and demands of older persons. Studies from both high income countries and emerging economies in Asia demonstrate that older persons can potentially be relatively wealthy consumers compared with young generations. This can lead to the creation of new markets and business opportunities to spur growth, create jobs and increase tax revenues (See Box 1.1). In addition, it has long been recognized that enabling more older workers to stay productive and in the labor market can be a key source of growth for aging societies. • From a societal perspective, old-age poverty may increase if there are more older persons without adequate income support or care, while at the same time older persons may take care of other family members or become volunteers. Aging can cause financial and care pressures for both older persons as care recipients and their families as informal caregivers. With economic transformation and rapid urbanization, many younger people move from rural areas to towns and cities, leaving older persons behind and putting pressure on rural communities. At the same time, older persons also have assets such as their social capital—that is, their networks of relationships—that allow them to make productive contributions to their families and communities. For instance, they can provide support within their own households by taking care of grandchildren or other family members, as well as participate in community activities as volunteers. They also often have historical memories of local communities and events, which can strengthen the communities’ culture, social cohesion and social solidarity. World Bank projections indicate that Malaysia is likely to transition from an upper middle-income economy to a high-income economy within the next few years, despite a COVID-19 induced recession in 2020 (see World Bank forthcoming). Navigating the journey to high-income country status raises a number of questions about not just the speed of Malaysia’s growth, but also the quality of growth and its sustainability. Most significantly, there is a growing sense that the aspirations of Malaysia’s middle-class are not being met and that the economy does not produce enough well-paid, high-quality jobs. There is also a widespread sense that the proceeds of growth have not been equitably shared between the richest and the poorest, and that increases in the cost of living are outstripping incomes, especially in urban areas. Malaysia’s transition to high-income status and the growing aspirations of its citizens regarding jobs, income and well-being lend additional urgency to addressing the challenges posed and opportunities offered by aging. While the rapid aging process is crucially important and will become even more so in coming decades, demography is not destiny. Various analysts and observers have highlighted that aging is one of the key megatrends that will shape tomorrow’s world (see McKinsey 2015, Deloitte 2017 and World Economic Forum 2017). However, through the right mix of policies, governments can help societies adapt to rapid aging and improve the well-being of all. The Government of Malaysia has already highlighted the importance of inclusive and productive, as well as healthy aging in its relevant policy documents. In 1995, the National Policy for the Elderly was Malaysia’s first strategic policy on aging. To implement this policy, the National Advisory and Consultative Council for Older Persons was set up in 1996 and a Plan of Action for the Older Persons was formulated in A Silver Lining: Productive and Inclusive Aging for Malaysia 25 Chapter 1: Introduction BOX 1.1 The Silver Economy The term silver economy has been coined to encompass all sources of opportunities that arise from economic activities that serve the needs and demands of older persons. While there is no common definition of the term silver economy, usually it is meant to include both the products and services that are directly purchased by older persons and any other economic activities that are indirectly induced by these purchases. Thus, it encompasses the unique cross-section of economic activities related to the production, consumption and trade of goods and services relevant for older persons. Various estimates from Europe, the United States and beyond show that the silver economy is very large and growing. Bank of America Merrill Lynch (2014) estimates that in 2014, the global silver economy stood at US$7 trillion, which would have made it the world’s third largest economy if it had been a country. The same study estimates that by 2020 the global purchasing power of older persons age 50 and above would reach US$15 trillion. Similarly, a set of studies by the European Commission (2015a and 2015b) reports that in 2015, the silver economy contributed over EUR4.2 trillion to the European Union (EU)’s gross domestic product (GDP) and sustained over 78 million jobs across the EU. This would be equivalent to 29 percent of EU GDP and 35 percent of its employment. According to the same set of studies, by 2025 the EU’s silver economy is projected to reach a GDP contribution of EUR6.4 trillion and 88 million jobs. This would be equivalent to 32 percent of the EU’s GDP and 38 percent of its employment. A study for the United States by Accius and Suh (2019) argues that the silver economy contributes US$8.3 trillion to the United States’ economy each year, or 40 percent of its GDP. The same study estimates that the silver economy contributes a further US$745 billion per year worth of unpaid non-market activities such as volunteering and caregiving. In addition, the study argues that the silver economy contributed US$2.1 trillion in tax revenue to the United States in 2018 and that this tax revenue will increase four times by 2050. Finally, according to the study, in 2018 the silver economy supported 88.6 million jobs in the United States either directly or indirectly, and this is projected to grow to 102.8 million jobs in 2050. Some recent studies indicate that the silver economy has even more potential in Asia than in Europe. A report by the World Data Lab (2019) estimates that the total annual spending power (in 2011 US$ adjusted for purchasing power parity) by persons age 60 and over in Asia will increase from US$4.2 trillion in 2019 to US$8.6 trillion in 2030. In comparison, the total annual spending power by persons age 60 and over in Europe will rise from US$3.7 trillion to US$5.2 trillion. The report’s projections also show that in spite of this impressive rise, older persons in Asia will still account for a smaller share of overall purchasing power than in Europe—15 percent as compared to 28 percent in Europe. But even though the silver economy’s share of overall purchasing power remains largest in high-income economies, given the size of emerging economies in Asia the silver economy in these economies is very sizeable. Again according to World Data Lab (2019), by 2030 the spending power of Chinese seniors is expected to grow to $3.2 trillion while the Indian silver economy is expected to experience explosive growth, to almost $1.4 trillion in spending power. Also at the country level, World Data Lab (2019) projects that Malaysia is expected to experience around 40 percent growth of the total annual purchasing power by persons age 60 and over from 2019 to 2030. 26 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 1: Introduction 1998. Based on a review of the first strategic policy and the Plan of Action, the National Policy for Older Persons and the Plan of Action for Older Persons were introduced in 2011. The National Policy for Older Persons presents a conceptual framework with five dimensions and corresponding long-term policy goals (see Figure 1.1). While this framework is largely focused on the well-being of older persons instead of providing a wider strategy for how Malaysia’s society can manage aging, it nevertheless reflects a strong commitment from the Government to create a policy environment that allows older persons to enjoy a respected life and be independent with dignity by optimizing their self-potential through a productive, healthy, active, positive, and supporting aging process. Following the introduction of the National Policy for Older Persons, the National Advisory and Consultative Council for Older Persons was strengthened. In addition, promotion and advocacy, lifelong learning, safety and security, governance and shared responsibility, intergenerational solidarity, and research and development were identified as six priority areas for interventions. Following the introduction of the National Policy for Older Persons, aging has been incorporated into Malaysia’s five-year national development plans and specific policies, regulations and programs have been further introduced. The 11th Malaysia Plan has included specific initiatives to improve the supportive environment for older persons and to promote active aging. These have been reinforced in the Mid-Term Review of the Plan which has emphasized the need to enhance older persons’ living environment. More specifically, the 11th Malaysia Plan has included a sub-strategy for “enhancing the living environment for the elderly”. This sub-strategy highlights specific initiatives regarding, for example, elderly-friendly infrastructure, the streamlining of social protection, enhanced research capacity to support active aging, and the expansion of lifelong learning for older persons. While these initiatives are closely aligned with the National Policy for Older Persons, at least in part they are at a relatively initial stage, leaving more room for further development and refinement for the 12th Malaysia Plan currently under preparation. Complementing the National Policy for Older Persons and the discussion of aging in five-year national development plans, specific policies, laws and regulations of relevance for the aging process include the National Housing Policy of 2011 that states the need for the Government and private sector to continue to provide affordable housing for specific targets groups including older persons, the Third National Plan of Action for Nutrition of Malaysia 2016 to 2025 that includes nutritional guidance for older persons, and various laws and regulations covering aged care and health care for older persons. As Malaysia continues to manage the challenges and opportunities of aging, it can learn from international experiences and lessons, including from other countries in East Asia. As Malaysia ages, it faces similar challenges to other aged or aged countries in East Asia. In fact, some countries in the region are further advanced in the aging process than Malaysia, including Australia, China, Japan, the Republic of Korea, Singapore and Thailand. Malaysia can learn from these countries’ experiences in managing the challenges and opportunities of aging, including in terms of changes and innovations in the relevant policy environment. For instance, starting in the 1960s, Japan has introduced a series of pension, health, long-term insurance, housing and transportation policies for aging and worked toward achieving an “age-free society” in which older persons will not be automatically considered senior citizens and will be encouraged to stay healthy and keep working. Korea has also introduced strategic plans and multiple active aging programs, in addition to a long-term care insurance system, and is now promoting an age-friendly economy. Australia is a good example for how to prepare for an aging society through evidence-based policy formulation. Since 2002, the Australian Government has produced a series of roughly five-yearly Intergenerational Reports to assess the long-term sustainability of current policies and the changes to Australia’s population size and age profile that may impact economic growth, the workforce and public finances over the next 40 years. Singapore has also developed a relatively advanced policy environment for aging through continuous policy and legislative reforms. Compared to these countries, Thailand and China are still at an early stage of the aging process, which is more comparable to that of Malaysia. Nevertheless, they have formulated long-term A Silver Lining: Productive and Inclusive Aging for Malaysia 27 Chapter 1: Introduction aging plans and piloted various relevant policies and programs. All these countries’ experiences and lessons can be helpful for Malaysia as the country further develops its long-term aging strategies, updates its laws and regulations, and pilots and scales up specific programs (see Table 1.2). Aging is a cross-cutting and multi-dimensional issue with productive aging, inclusive aging and healthy aging as key dimensions. International experiences and lessons suggest that it is best practice to develop an integrated approach to formulating aging-related policies and regulations. These policies and regulations would capture the relevant demographic, social and economic context, identify constraints and TABLE 1.2: Aging policies in select East Asian countries Country Policies The Aged Persons Homes Act 1954; the Residential Aged Care Services in Australia 2008; the Aged Care Packages in the Community 2008; the Home Care Package 1997; the Aged Care Act Australia 1997, amended in 2006, 2013 and 2016; and the Aged Care Quality and Safety Commission Act 2018. Aging and Aged Care Five-Year Development Plans since 2001, 10 th (2001-2005), 11th (2006-2010), 12th (2011-2015), and 13th (2016-2020); Long-Term Aging Strategies (2019—2022/2035/2050); the China Elderly Rights Protection Law 1995, amended in 2015 and 2018; National Aged Care System Pilots 2016—ongoing; and Long-Term Care Insurance pilots 2016—ongoing. Universal Health Insurance Coverage (1961) and Universal Pension Coverage (1961); the Elderly Welfare Act (1963); the “Gold Plan” (1990-2000) for a massive investment for long-term care services; Decentralization in Social Welfare (1990); Act on Promotion of Construction of Accessible Japan Building (1994); Pension Reform (1994); the Long-Term Care Insurance Scheme (2000); Act on Promotion of Accessible Public Transportation (2000); Comprehensive Reform in Tax and Social Security Reform (2012-); Promotion of Community-Based Integrated Care; Promoting Aging in Place and Building an “Age-Free Society”. Senior Welfare Act 1981, amended in 2012, 2013, and 2015; Act on Prohibition of Age Discrimination in Employment and Aged Employment Promotion 1991; Act of The Long-Term Care Security for Republic the Elderly 2007; Universal Health Coverage 1989; Long-Term Care Insurance System 2008; Aging of Korea Society and Population Master Plan established every five years: 1st (2006-2010), 2nd (2011-2015), 3rd (2016-2020); Active aging and social welfare programs for older persons implemented by multiple government ministries; Leaping toward an “Age-Friendly” economy. Action Plan of Successful Aging 2016; Whole of Government Approach, Ministerial Committee on Aging, Age Planning Office and Agency of Integrated Care under MOH; Health care 2020 Singapore Masterplan; Pioneer Generation Package; Maintenance of Parents Act and Mental Capacity Act; Reemployment Act 2012 (extending retirement age); and immigration policies for foreign domestic workers and care staff. The First National Older Persons Plan 1986-2001; The Second National Older Persons Plan 2001- 2021; the National Health Security Act 2002 for a universal health coverage; the Older Persons Thailand Act 2003 for elderly right including elderly allowance; the National Commission on the Elderly 2003; the National Long-Term Care Plan 2011; and the Elderly Fund to provide non-contributory pension, assistance and aged care services. Source: Authors. 28 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 1: Introduction policy priorities, and implement tailored and sequential programs and interventions to promote productive, inclusive, and healthy aging. According to Malaysia’s National Policy for Older Persons, productive aging is defined as the capacity of older persons to contribute through paid or voluntary work that brings meaning and satisfaction. Again according to the National Policy, healthy aging refers to efforts toward a healthy lifestyle, a better health system, and the creation of an environment and community which prioritizes health. Inclusive aging is defined in this report as the pursuit of an inclusive society where older persons have access to equitable health and social services, can participate in social and economic activities, and share the society’s prosperity. All these three key dimensions of aging are important and interlinked. For instance, if older persons are healthy this creates the physical and mental preconditions that allow them to be active and productive. In addition, if older persons are productive this provides them with better income security and allows them a healthy lifestyle. Finally, if aging is inclusive the gains of social and economic development are more widely shared, which in turn creates productive opportunities for vulnerable older persons. Against this backdrop, this report aims to contribute to a considered debate of the policy choices needed to achieve productive and inclusive aging in Malaysia. The report recognizes the importance of healthy aging but does not discuss it in detail. Instead, it focuses on productive and inclusive aging and the linkages and interactions between these two dimensions and healthy aging. In addition to highlighting the criticality of the issue for Malaysia’s future development, in Chapter 1 the report adopts a balanced view that highlights both challenges and opportunities of aging. In Chapter 2, it follows a three-part approach to document the relevant context and state of preparedness and in Chapter 3 it investigates employment, income security for older persons, and aged care as three key policy areas. Finally, in Chapter 4 it develops a set of actionable policy recommendations. A comprehensive analysis of some other relevant policy areas, in particular, the health sector and the fiscal sustainability of the civil service pension system are left for future research. Similarly, while the report recognizes that managing aging requires a life cycle perspective on policy design it does not try to comprehensively address all relevant policy areas such as children’s nutrition, human capital accumulation and medical care. Instead, the focus is on policies either directly affecting older persons, such as those related to aged care, or those with close linkages or spillovers between older persons and others, such as on employment and old-age income security. An even more in-depth analysis of some of the issues investigated in the report is also left for future research. For instance, the report recommends the development of a systematic and actionable aged care strategy but also recognizes that building on the analysis of this report, further in-depth analyses including a comprehensive assessment and diagnostic and benchmarking against international experiences and best practices would be needed to inform the development of such a strategy. Chapter 2 of this report consists of three sections and documents the relevant context and state of preparedness including with regard to demographic trends, socioeconomic circumstances and macroeconomic impacts of aging. Section 2.1 highlights key aging trends in Malaysia and compares its demographic transition with those in other countries in East Asia. It also examines the factors underlying Malaysia’s aging trends and discusses past, present and future developments in the size of the working-age population and total and old-age dependency ratios. In addition, it documents detailed patterns of aging across groups and space. Section 2.2 uses the World Bank’s Long-Term Growth Model (LTGM) to analyze the impacts of aging on the long-term path of headline and per capita economic growth, private savings and other macroeconomic variables. Throughout the report, the LTGM is also used to quantitatively explore some of the policy reforms and structural changes that could mitigate the impacts of aging on long-term growth. Section 2.3 explores key patterns of well-being of older persons in Malaysia, including with regard to income, household composition, and the incidence and depth of poverty. It also investigates how these patterns of well-being differ between population groups and across space and how they have evolved over time. A Silver Lining: Productive and Inclusive Aging for Malaysia 29 Chapter 1: Introduction Chapter 3 also consists of three sections and investigates employment, income security for older persons, and aged care as key policy areas. Thus, Chapter 3 does not aim to be a comprehensive treatment of possible policy implications of aging in Malaysia, but focuses on selected key areas of high policy interest and on closing critical knowledge gaps. Section 3.1 aims to build an understanding of the current labor market environment for older workers in Malaysia. This will be the basis for actionable policy recommendations to extend productive working lives. For this purpose, it investigates recent trends in the employment of older workers including with regard to labor market participation patterns by gender, ethnicity, urban-rural location, and education. In addition, the section elaborates on the types of jobs that older workers hold and patterns with regard to earnings at any particular age and across workers’ entire careers. Section 3.2 evaluates the performance of Malaysia’s system of contributory and non-contributory social protection programs to achieve adequacy, coverage, and affordability of old-age income protections. It builds on and extends the relevant analysis in the Malaysia Economic Monitor Surviving the Storm (see World Bank 2020) by covering relevant institutions beyond the EPF, evaluating the performance of social assistance programs implemented by the Department of Social Welfare (Jabatan Kebajikan Masyarakat or JKM) specifically for older persons, and providing a more extensive discussion of considerations and options for a non-contributory “social pension.” Section 3.3 reviews aged care policies, laws and programs, and discusses crucial areas such as aged care provision and financing, quality assurance and human resources, as well as the governance of private for-profit and not-for-profit providers. The section documents that aging, accompanied by changing socioeconomic circumstances and expectations of older persons, is expected to increase the demand for a variety of aged care services and to pose a significant challenge to existing aged care arrangements in Malaysia. Chapter 4 uses the documentation of the context and investigations of policy key areas to develop a set of actionable policy recommendations. The policy recommendations recognize that the promotion of productive and inclusive aging will require a comprehensive, interagency policy approach that addresses challenges and constraints in a systematic and mutually reinforcing way, as well as an approach that adopts a 30 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 1: Introduction life cycle perspective and mainstreams interventions to address specific constraints that women face. Based on this recognition, Chapter 4 argues that fostering the productive employment of all workers, including older workers, will be crucial for mitigating the growth impacts of aging and addressing persistent talent shortages. In addition, improvements in the income security of older persons will require an integrated vision of a system for old-age income protection with policy measures in the areas of both social insurance and social assistance. With regard to social insurance, proactive measures can improve the coverage and adequacy of relevant schemes. However, even with all these measures it is unlikely that social insurance schemes can ever reach full coverage of the labor force. Thus, a modest, broadly targeted tax-financed social pension may also be required. Finally, for the aged care sector to become a new driver of economic growth, it will be crucially important to create an enabling market and regulatory environment for private not-for- profit aged care provision, to strengthen the governance of the sector, and to strategically and selectively increase public financing in line with available fiscal space. The report relies on a mixed-methods approach to investigate the factors that constrain the achievement of productive and inclusive aging and to develop actionable policy recommendations. The mixed-methods approach combines quantitative, qualitative and institutional research. The quantitative research is conducted using recent, nationally representative data from the Household Income and Expenditure Survey and the Labour Force Survey, both of which are conducted by the Department of Statistics Malaysia. In addition, international data sources (the World Bank’s World Development Indicators and the United Nations’ Population Prospects, for example) are used to compare Malaysia with other countries. Complementing the quantitative investigation, qualitative and institutional analyses are used to improve the understanding of the constraints behind the most significant challenges identified by quantitative data, in particular in the area of aged care where significant knowledge gaps exist. The qualitative analysis employs semi-structured interviews with diverse stakeholders to answer three related questions. First, what are views on the roles and responsibilities of older persons, families, communities, non-governmental organizations (NGOs), the private sector, and the Government in providing and financing aged care? Second, what are the views of family caregivers and care workers with regard to their skills, remuneration, working conditions, quality of services, and Government support? Third, what are the experiences, preferences, ideas, concerns, and expectations of older persons with regard to aged care?4 Finally, the institutional analysis encompasses a desk review of relevant laws and regulations in areas such as employment, pensions and aged care. The report recognizes that while aging is a long-term process, the policy challenges it raises are deeply intertwined with and exacerbated by short- and medium-term phenomena such as the COVID-19 pandemic. The year 2020 not only marks a crucial milestone in Malaysia’s development trajectory because it becomes an aging country, but also because it is the year during which the COVID-19 pandemic presents arguably the greatest economic, health and social challenges in the country’s history. As discussed in this report, the increased rates of unemployment and underemployment due to the COVID-19 pandemic lend urgency to the policies that allow the productive and inclusive employment of older workers, including with regard to the provision of enhanced and tailored opportunities for lifelong learning. In addition, the COVID-19 pandemic may increase the vulnerability of older persons while reducing their savings for retirement through declining EPF contribution rates, withdrawals from EPF accounts, and declines in asset prices. In fact, reduced contribution rates to and early withdrawal modalities from EPF accounts formed an important part of Malaysia’s response to COVID-19. Finally, COVID-19 infections in several aged care homes in Malaysia demonstrate the importance of service standards, in particular health and safety standards, to protect older persons living in institutional settings as an exceptionally vulnerable population group. 4 For the precise methodological approach of the qualitative analysis, see Annex B. The complete sets of findings and implications from the qualitative analyses are documented in a background paper to this report by French (2020). A Silver Lining: Productive and Inclusive Aging for Malaysia 31 CHAPTER 2 Context 32 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context 2.1 Demographic Context Nationwide Demographic Trends Malaysia becomes an aging society in 2020, according to projections by both the Department of Statistics Malaysia (DOSM) and the United Nations World Population Prospects (UN WPP). 5 An aging society is defined as a society with 7 percent or more of the population age 65 and above. Thus, following a recent uptick in the pace of demographic change the year 2020 marks an important demographic milestone for Malaysia. For much of Malaysia’s developmental history since its formation in 1963, the country was demographically “young”—less than 4 percent of the population was age 65 and above until twenty years ago (see Figure 2.1). In the past, the pace of demographic change was also relatively slow. It took Malaysia 36 years for the proportion of older persons to increase by just one percentage point from 3.3 percent in 1970 to 4.3 percent in 2006. However, around the year 2010, the pace of aging began to accelerate. As a result, it took just six years from 2006 to 2012 for the proportion of older persons to increase by a further one percentage point to reach 5.3 percent. Based on current projections, the proportion of older persons will increase by more than another percentage point to 8.2 percent in 2023, which is only three years away. The pace of aging in Malaysia is nearly identical to Japan’s. Among countries in the Association of Southeast Asian Nations (ASEAN), Thailand, Singapore, and Vietnam became aging nations ahead of Malaysia, and all ASEAN countries except Lao PDR and Cambodia will make this transition by 2030. The FIGURE 2.1: Population age 65 and above, FIGURE 2.2: Transition from aging to aged, and Percentage and Number (’000) super-aged status by country, Years 16 7,000 Lao PDR Cambodia 14 6,000 Philippines Myanmar 12 Population age 65+ (’000) 5,000 Brunei Darussalam Population age 65+ (%) Indonesia 10 4,000 Malaysia 8 Vietnam 3,000 Singapore 6 Thailand Turkey 2,000 4 China Republic of Korea 2 1,000 Japan Australia 0 0 France 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 United Kingdom 1950 2000 2050 Population age 65+ (’000) Percent of population age 65+ Aging duration Super-aging duration Source: Authors’ calculations based on DOSM. Source: Authors’ calculations based on UN WPP. Note: Data for Australia, France and United Kingdom are truncated as these were aging societies even before 1950. Blue bars represent the transition period from aging to aged. Orange bars represent the transition period from aged to super-aged. 5 Unless otherwise stated, demographic projections are based on the medium fertility projection variant of the UN WPP’s 2019 Revision. This variant incorporates information on net international migration including flows of labor migrants, irregular migrants, and refugees (see United Nations 2019). A Silver Lining: Productive and Inclusive Aging for Malaysia 33 Chapter 2: Context pace of aging in Malaysia is so rapid that in the 24 years from 2020 to 2044, Malaysia is projected to reach the next demographic milestone—that of an aged nation (more than 14 percent of the population age 65 and over). This same transition, from aging to aged nation, took 115 years in France, 69 years in the United States (see Kinsella and He 2009), and 24 years in Japan. Japan’s and Malaysia’s subsequent transition from aged to super-aged (more than 20 percent of the population age 65 and over) is projected to take 11 and 12 years, respectively, and thus to follow a remarkably similar pace. It is worth noting that while Japan underwent its demographic transition at a similar pace to Malaysia, it did so at a more advanced stage of economic development and that several East Asian countries will make the same transition even more rapidly than Malaysia (for instance China, the Republic of Korea, Singapore, Thailand and Vietnam, see Figure 2.2). Nevertheless, the magnitude of the demographic challenge faced by Malaysia should not be underestimated. Projections indicate a net increase per year of between 130,000 to 210,000 older persons between 2020 and 2044. As a result, the number of older persons will reach 5.6 million by 2044. Malaysia’s overall aging narrative remains unchanged irrespective of how aging is measured and of the precise cutoff to differentiate between older and non-older persons. This report recognizes potential shortcomings of relying on chronological age to measure individual and population aging processes and therefore, in Box 2.1, complements this concept with newer measures of aging—prospective age and characteristics-based measures. Nevertheless, the report generally uses chronological age to measure individual and population aging and relies on age 65 as the cutoff to define older persons, in alignment with international norms and age categories published by DOSM. While the cutoff of 65 makes it possible to contextualize the current state and pace of Malaysia’s demographics across countries, it is worth noting that policymakers in Malaysia often use 60 as the cutoff to differentiate between older and non-older persons and define an “aging society” as one with 15 percent or more of the population age 60 and above, a threshold expected to be reached in 2030. What is most important is that the aging narrative remains unchanged irrespective of cutoff age used (see Figure 2.3). FIGURE 2.3: Population age 60 and above and FIGURE 2.4: Total fertility rate and male and 65 and above, Percentage female life expectancy at birth, Percentage and Years 25 7 80 6 75 20 Life expectancy at birth 5 Population age 60+ / 65+ (%) 70 Total fertility rate 4 15 65 3 60 10 2 55 1 5 0 50 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 2018 0 Total fertility rate 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 Life expectancy at birth (Male) Population age 60+ (%) Population age 65+ (%) Life expectancy at birth (Female) Source: Authors’ calculations based on DOSM. Source: Authors’ calculations based on UN WPP. 34 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context Malaysia’s rapid pace of aging has been driven by a precipitous decline in the total fertility rate accompanied by a sustained rise in life expectancy, especially for women. Malaysia’s total fertility rate (that is, the average number of children born to a woman over her lifetime) for the period from 1960 to 1964 was 6.4, but the fertility rate dropped to the replacement level of 2.1 in the period from 2010 to 2015. At two, it is now below the replacement rate (see Figure 2.4). In parallel, life expectancy at birth has increased by more than two decades since the 1950s. The increase in female life expectancy has been particularly impressive. It now stands at 78 years, four years longer than males. Although life expectancy has increased, not all of the increased years of life are spent in good health or free of disabilities, underscoring the importance of healthy aging over the life cycle. Malaysia’s healthy life expectancy at birth is estimated at 66.6 years in 2016, about a decade less than life expectancy, according to the WHO Global Health Observatory. This divergence is partly due to the fact that the prevalence of NCDs is high and rising. For example, the prevalence of diabetes among adults increased from 11.2 percent in 2011 to 18.3 percent in 2019, according to the National Health and Morbidity Survey 2019. The prevalence of diabetes also rises with age—among those age 60 and above, 41.5 percent have diabetes, whether diagnosed or not. NCDs frequently lead to complications such as strokes which in turn can result in disability and dependence (also see Section 3.3). Unsurprisingly and perhaps due to other contributing factors such as a lack of nutrition, literacy or education, the prevalence of IADLs limitations among those age 60 and above is high (42.9 percent in 2018, according to the National Health and Morbidity Survey 2018). Adjusting for the prevalence of disabilities, Malaysia’s old-age dependency ratio already seems to be close to that of France (see Box 2.1). A Silver Lining: Productive and Inclusive Aging for Malaysia 35 Chapter 2: Context BOX 2.1 Measuring Aging Chronological age is the most common and straightforward but not always the best measurement of aging, as it looks backward rather than forward. Sanderson and Scherbov (2019) introduce prospective age as an alternative, forward-looking measure—persons with the same prospective age, across different periods of time, have the same expected remaining life expectancy (RLE). Figure 2.5 uses an RLE duration of 18 years to calculate prospective age trends for Malaysia.6 This is because 18 years was the RLE of a person age 55 in the early 1950s, around the time of the EPF Act of 1951 was passed. The figure shows that a person age 55 in 1953 would have been expected to live another 18 years—as would a person age 64 in 2023 and a person age 68 in 2053. By contrast, chronological age at 55 remains unchanged over time. Prospective age can further be used to measure aging dynamically, using the population’s life expectancy to adjust the age structure and relevant parameters such as the old-age dependency ratio. In effect, this measure “corrects” chronological age by taking into consideration life expectancy and its relevance to economic, labor, and pension policies. 6 Many countries use an RLE duration of 15 years to compare prospective age, reflecting labor and pensions parameters at the point that social security systems for older persons were introduced in many high-income countries. 36 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context FIGURE 2.5: Prospective and chronological age, FIGURE 2.6: Disability-adjusted old-age Years dependency ratio, Percentage7 70 67.9 Malaysia Prospective and chronological age (Years) 64.3 65 France Czech Republic 60 Slovak Republic 55 Hungary 55.0 55.0 Spain 50 1953 1963 1973 1983 1993 2003 2013 2023 2033 2043 2053 Belgium Mid year of period 0 10 20 30 40 Prospective age Chronological age Disability-adjusted old-age dependency ratio (%) Source: Authors’ calculations based on UN WPP. Sources: Authors’ calculations based on UN WPP, National Health and Note: Prospective age—where RLE is fixed to the RLE of a ‘standard’ Morbidity Survey 2018 and Bussolo et al 2015. year/period. 1950–1955 is the period used here. Characteristics-based measures of aging, as a generalization of prospective age, uses characteristics of aging to define a person’s age, as older persons are a very heterogeneous group with a diverse range of functional capacities (WHO 2015). Characteristics-based measures of aging adjust the old-age dependency ratios—defined as the number of persons over the age of 65, compared with the working-age population age 15 to 64—by disabilities which characterize old age rather than by chronological age. Thus, characteristics-based measures highlight that aging not only pertains to the end of life but to the whole life cycle. Although it may not always be possible to compare different countries using characteristics-based measures of aging due to differences in survey instruments and the environment (more or less enabling for those with disabilities), Malaysia’s disability-adjusted old-age dependency ratio of 10.6 percent already seems to be close to that of France, which stands at 12.1 percent (see Figure 2.6). This reinforces the criticality of extending disability-free lifespans through strengthened public health care and emphasizing health promotion and risk factor reduction for persons of all ages. There are other concepts and measures of aging, such as biological age (see Skirbekk 2019). The concept of biological age relies on a set of biomarkers to measure aging more precisely than chronological age which disregards aspects such as economic productivity, health, or functional capacities. While the concept of biological age is intuitively appealing, its application has proven technically challenging. 7 Note: Full comparability between Malaysia and other countries may be affected as different survey instruments are used. The disability-adjusted old-age dependency ratio is defined as the number of adults who are at least 20 years of age and who have disabilities or dependencies in IADLs, divided by the number of adults who are at least 20 years of age and who do not have disabilities. A Silver Lining: Productive and Inclusive Aging for Malaysia 37 Chapter 2: Context From a demographic perspective, Malaysia’s best years are now. As Malaysia transitions from a relatively young to an aging nation, it is currently in a demographic sweet spot. The total dependency ratio—the number of persons age 0 to 14 and over the age of 65, compared with the working-age population of 15 to 64—stands at 44 percent. This is at the lowest it has been since at least 1970 (see Figure 2.7). This is due to a rapid decline in the young-age dependency ratio—the ratio of individuals age 14 or below compared to the working-age population—as fertility rates have dropped, as well as a previously moderate increase in the old-age dependency ratio—the proportion of individuals age 65 and above. However, after this period, the total dependency ratio will rise again. This is driven by both an increase in the absolute number of older persons, as well as a decrease in the relative, but not necessarily absolute, number of persons of working age. Accordingly, the proportion of the working-age population is projected to decrease by 3 percentage points—from 69 percent of the population in 2020 to 66 percent in 2050—(see Figure 2.8) even though the working-age population is projected to increase by 4.4 million over the same time period. This increase in the total dependency ratio has critical implications for economic growth. As the burden of caring for the young and older persons falls on fewer and fewer working-age adults, the growth of Malaysia’s economy will have to be driven by productivity growth and the expansion of economically productive years, not the increasing share of the working-age population (see Section 2.2). FIGURE 2.7: Total and old-age dependency FIGURE 2.8: Change in working-age population ratios, Percentage from 2020 to 2050, Percentage 100 ASEAN 90 Lao PDR 80 Philippines 70 Dependecy ratio (%) Cambodia 60 Myanmar 50 Malaysia 40 Indonesia 30 Vietnam 20 Brunei Darussalam Thailand 10 Singapore 0 Republic of Korea 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 2014 2018 2022 2026 2030 2034 2038 -25 -20 -15 -10 -5 0 5 10 Total dependency ratio (%) Old-age dependency ratio (%) Change in working age population (%) Source: UN WPP. Source: Authors’ calculations based on UN WPP. Net inward migration of working-age adults into Malaysia since the 1980s has moderated the pace of aging and created more favorable demographics—for now. Over 90 percent of noncitizens in Malaysia in 2020 are of working age. Hence, the net inward migration of 3 million persons since the 1980s has helped to moderate the pace of aging in Malaysia (see Figure 2.9). In fact, if noncitizens were excluded from the calculation of the total dependency ratio, this would amount to 50 percent in 2020. In other words, two working age adults would on average need to support one dependent, young or old. However, if noncitizens are included in calculations, the total dependency ratio is 44 percent, that is, more than two working age adults support one dependent. If there were henceforth no more inflows of working-age noncitizens into Malaysia, this would result in less favorable demographics. 38 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context FIGURE 2.9: Net migration, Numbers (’000) FIGURE 2.10: GNI per capita and old-age dependency ratio, 2010 US$ and Percentage 1,000 GNI per capita (constant 2010 US$) 800 20,000 2018 10,000 Net migration ('000s) 600 2017 5,000 400 2,500 1960 200 1,000 1995 500 0 0 5 10 15 20 -200 Old-age dependency ratio (% of working-age population) 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 Republic of Korea Singapore Japan East Asia and Paci c Mid year of period Malaysia Source: Authors’ calculations based on UN WPP. Source: Authors’ calculations based on World Bank World Development Indicators. Malaysia has not grown old before it has grown rich to the extent of other countries in East Asia and the Pacific—but the window for further demographics-driven economic growth is closing rapidly. Compared with the East Asia and Pacific region as a whole, Malaysia’s level of economic development has been above average for any given old-age dependency ratio. For instance, at an old-age dependency ratio of 10 percent, the typical country in East Asia and Pacific had a gross national income (GNI) per capita of US$5,000 while Malaysia’s GNI per capita was US$10,000. However, while Malaysia underwent rapid economic development during the period when its old-age dependency ratio was low, the subsequent rise in the old-age dependency ratio means that the window for further demographics-driven economic growth is closing quickly, which makes immediate action an imperative. Put differently, GNI per capita increased steadily during the period from 1960 to 2000 when the old-age dependency ratio was just over 6 percent but has recently inflected towards a lower economic growth trajectory, with relatively smaller increases in GNI per capita for any given increase in the old-age dependency ratio. In addition, while Malaysia has not “grown old before it has grown rich,” it is also not on a growth trajectory required to catch up with high-income countries in East Asia and Pacific such as Japan, Singapore, and the Republic of Korea (see Figure 2.10). Subnational Demographic Trends Aging patterns within Malaysia exhibit pronounced heterogeneities—for instance, there is more than a decade of difference between the stage of aging across ethnicities. Malaysia’s Chinese community was already an aging society with 7 percent or more persons age 65 and above by 2010 while the Bumiputera community, encompassing Malays and other indigenous peoples, is projected to become an aging society only in 2023, which is 13 years later (see Figure 2.11). These differences reflect a smaller fall in the total fertility rate among the Bumiputera community to 2.8 by 2010, compared with the fall in fertility rate among Chinese to 1.8 that same year (see Zarinah Mahari 2011). The Indian community reached the threshold of A Silver Lining: Productive and Inclusive Aging for Malaysia 39 Chapter 2: Context an aging society in 2018. Across ethnicities, women in general are at a more advanced stage of aging than men, by approximately two years. The combined interethnic and gender differences in the aging process are so large that, for example, Chinese women, as a community, will become aged around the same time that Bumiputera men become aging. FIGURE 2.11: Transition from aging to aged, FIGURE 2.12: Transition from aging to aged, and and aged to super-aged status by ethnicity and aged to super-aged status by state, Years gender, Years Putrajaya Bumiputera Males Sabah Labuan Bumiputera Females Terengganu Kelantan Indian Males Selangor Kuala Lumpur Indian Females Johor Pahang Chinese Males Sarawak Negeri Sembilan Chinese Females Kedah Melaka Indonesia Penang Perlis Malaysia Perak Indonesia Vietnam Malaysia Vietnam Singapore Singapore 2010 2020 2030 2040 2000 2010 2020 2030 2040 Source: Authors’ calculations based on DOSM and UN WPP. Source: Authors’ calculations based on DOSM and UN WPP. Note: Data for Chinese men and women are truncated as these were aging Note: Data for Perlis and Perak are truncated as these were aging societies population groups even before 2010. Blue bars represent the transition even before 2010. Blue bars represent the transition period from aging to period from aging to aged, orange bars the transition period from aged to aged, orange bars the transition period from aged to super-aged. super-aged. There are large geographic differences; up to two decades separate the stage of aging across states. Two states—Perlis and Perak—had already become aging societies by 2010 while in the same year, only 2.9 and 5.5 percent of residents in Sabah and Kelantan were age 65 or older, respectively. This means the demographic profile of Perlis and Perak is more comparable to that of Singapore than to those of some other states, while Sabah and Kelantan are more demographically comparable to Indonesia (see Figure 2.12). Perhaps unsurprisingly, Sabah is only projected to become an aging society in 2029, approximately two decades after Perlis and Perak.8 In addition to the stage of aging, the pace of aging also varies substantially across states. For instance, relative to the rest of the country, Selangor is younger and has a larger proportion of working-age adults. As a consequence, Selangor is only projected to become an aging society in 2023, three years after Malaysia as a whole.9 However, the pace of aging in Selangor is projected to be so rapid that the state will transition to an aged society after only 15 years. Thus, it is projected to become an aged society in 2038, six years before Malaysia as a whole. Selangor currently accounts for the largest number of older persons in Malaysia, followed by Johor, but Sabah and Sarawak are catching up rapidly (see Figure 2.13 and Figure 2.14). Selangor, Johor and Perak—three relatively well-off states—have the largest number of older persons currently, while Sarawak and Sabah in East Malaysia come fourth and seventh, respectively. Until 2040, differential demographic changes are projected to result in a quadrupling of the number of older persons from 174,000 to 627,000 8 The literature has also documented wide variations in proportion of older persons below the level of the state, that is, at the district and sub-district level (see Hamid 2019). 9 Internal net migration projections are based on a transition probability matrix for the period 2000 to 2010 (see DOSM 2016). 40 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context FIGURE 2.13: Population age 65 and above by FIGURE 2.14: Population age 65 and above by state, 2020, Persons state, 2040, Persons Sabah Sarawak 627,225 552,721 Perak Sarawak Kedah 274,218 226,785 190,470 WP Kuala Selangor Kedah Lumpur 402,936 388,332 367,123 WP Selangor Kuala 1,319,868 Perak Pulau Pinang Lumpur 458,016 184,263 137,570 Pahang 283,182 Pulau Johor Sabah Pahang Johor Pinang Kelantan 294,488 174,021 131,258 742,248 414,324 238,293 Source: Authors’ calculations based on DOSM. Source: Authors’ calculations based on DOSM. Note: National population aged 65 and above = 2.4 million. Note: National population aged 65 and above = 6.0 million. in Sabah, one of Malaysia’s poorer states. This will cause it to leapfrog to become the state with the third highest number of older persons. In parallel, Sarawak will become the state with the fourth highest number of older persons. More broadly, unequal demographic trajectories superimposed on unequal economic development across states is an aspect of Malaysia’s aging journey which will need to be addressed. Several states such as Kedah, Perlis, Pahang, and Perak are already both relatively poorer and older than Malaysia as a whole. In addition, Sabah is relatively poor but is still relatively young for now. Hence, it may have a short window of opportunity left to leverage demographics-driven growth. A Silver Lining: Productive and Inclusive Aging for Malaysia 41 Chapter 2: Context 2.2 Macroeconomic Impacts Impacts on Growth In Malaysia, like several other Asian countries, a demographic dividend contributed up to a third of headline economic growth between the 1960s and the 1990s (see Bloom, Caning and Finlay 2010 and Bloom and Williamson 1998). In several Asian countries, favorable demographics, fueled by the decline in the young-age dependency ratio and resulting rise in the share of the working age to total population ratio, contributed to fast economic growth rates between the 1960s and the 1990s. The contribution to growth of favorable demographics has come to be known as the demographic dividend. As discussed in Section 2.1, Malaysia was one of the countries in East Asia that benefited from these favorable demographics for much of its independent history. However, as also discussed in Section 2.1 Malaysia like many countries in the region now has a population that is aging faster than its level income may suggest, especially compared to many European countries (see World Bank Group 2016). Today, Malaysia has a population that is aging fast, leading to negative long-term growth impacts unless proactive mitigation policies are adopted. One of the most direct impacts of aging is the decline in the share of the working-age population. As detailed in Section 2.1, this is primarily driven by a decline in the fertility rate and a rise in longevity; as the fertility rate falls and a larger portion of the population is age 65 and older, the working-age population as a fraction of the total population shrinks. This implies that to some extent, aging is an integral part of the development process; as countries become richer, fertility rate tends to fall and with access to better health care and nutrition people tend to live longer. However, the process of aging can still be much faster in some countries than in others. For example, Malaysia is aging faster compared to the Republic of Korea at the same level of income. Malaysia’s gross national income (GNI) per capita in 2018 was about the same as Korea’s GNI per capita in 1991, but in that year the proportion of the population age 65 and over in Korea was 5.5 percent, compared to 6.9 percent in Malaysia (see Section 2.1). Using the World Bank Long-Term Growth Model (LTGM), it is possible to analyze how changes in the share of the working-age population will impact long-term economic growth in Malaysia. The LTGM is an extension of the Solow-Swan growth model where the key building blocks include saving, investment and productivity. The model is developed and maintained by the Macroeconomics and Growth Team of the World Bank’s Development Research Group. See Annex C for details on the model and its calibration to the Malaysian context. In the baseline scenario, aggregate economic growth in Malaysia is expected to decline from 4.5 percent in 2019 to about 1.8 percent by 2050 (see Figure 2.15). Under an assumption of continuity of recent trends in the economy, headline growth is expected to slow down substantially. Growth in GDP per capita is projected to fall substantially as well, from 3.2 percent in 2019 to 1.5 percent in 2050. This decline in long-term economic growth, as projected in this baseline scenario is not unique to Malaysia, but is typical for economies transitioning to high-income status (see Figure 2.16) which Malaysia is expected to achieve by 2024 based on the same baseline projection. The projected decline in growth can be attributed to a fall in the following five growth drivers: (i) population growth, (ii) growth in the ratio of persons in working age to the overall population, (iii) total factor productivity (TFP) growth, (iv) human capital growth, and (v) effectiveness of public and private investment. 42 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context FIGURE 2.15: Projected headline and per capita FIGURE 2.16: Headline GDP growth by economy, GDP growth, Percentage Percentage 5 14 12 4 10 GDP growth (%) GDP growth (%) 3 8 6 2 4 1 2 0 0 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 1960 1970 1980 1990 2000 2010 Headline GDP growth (%) Per capita GDP growth (%) Hong Kong SAR, China Republic of Korea Singapore Taiwan, China Source: Authors’ calculations based on LTGM. Source: Authors’ calculations based on World Bank World Development Indicators. Through alternative scenarios it is possible to isolate the long-term contribution of demographic changes to the projected decline in Malaysia’s economic growth. In the LTGM, demographic changes are captured by population growth and the growth of the working-age to total population ratio. Both of these growth drivers are projected to decline under the baseline scenario. However, this baseline scenario ignores potential countervailing developmental or policy changes such as the potential for further increases in female labor participation (see Lee and Chung 2008, Canning et al. 2009 and World Bank 2020). By sticking to the baseline assumption that aging will cause a decline in the labor force, two alternative scenarios in which the growth rate of the population and the working-age to total population ratio are constant can be used to reveal an approximation of the long-term growth impact of aging. These alternative scenarios assume that all the other variables continue on the same path as assumed under the baseline projections, which of course might not necessarily be the case if there are interaction effects between aging and variables that countervail the negative impacts of aging on growth (see Bloom et al. 2016). All else being equal, demographic changes are projected to account for about a third of the decline in GDP growth between 2020 and 2050. Projections using the LTGM show that the decline in population growth accounts for about 19 percent of the slowdown in headline growth, while the decline in the growth rate of working-age to total population ratio accounts for 12 percent (see Table 2.1). This means that, altogether, demographic changes are projected to account for 31 percent of the decline in GDP growth between 2020 and 2050. It also means that in an alternative scenario where both population growth and the growth rate of working-age to total population ratio were to remain constant at their 2020 level until 2050, the long run GDP growth rate would be 1.1 percent higher (2.9 percent instead of 1.8 percent in 2050). Figure 2.17 and Figure 2.18 show that under the assumption that the population growth rate remains constant at 1.3 percent per year, headline GDP growth will still decline from 2020 to 2050, but only to 2.5 percent instead of 1.8 percent. Similarly, Figure 2.19 and Figure 2.20 show that under the assumption that the ratio of persons in working age to the overall population will remain unchanged, headline GDP growth of 2.3 percent will be achieved in 2050. A Silver Lining: Productive and Inclusive Aging for Malaysia 43 Chapter 2: Context TABLE 2.1: Projected headline GDP growth and drivers of growth Variables Contribution (%) Share (%) Population growth -0.63 19 Growth in ratio of persons in working age to the overall population -0.41 12 TFP growth -0.39 12 Human capital growth -0.07 2 Public Investment (falling effectiveness) -1.43 43 Private capital to output ratio (falling effectiveness) -0.40 12 Baseline -2.60 100 Source: Authors’ calculations based on LTGM. Certain structural changes in the economy are likely to limit the impacts of aging on growth. In many countries the female labor force participation has risen concurrently with a decline in fertility rate (see Canning et al. 2009). This has been documented for the Republic of Korea (see Lee and Chung 2008), as well as for Australia, Canada, and the United States (see Brown and Guttman 2017). The female labor force participation rate has also been rising in Malaysia, though the World Bank (2019b) ascribes this more to the expansion of pre-school education and other policy choices, rather than to the country’s declining fertility rate. Another variable that might not remain constant is human capital. In fact, this is expected to increase further as the average educational attainment in Malaysia continues to rise (see World Bank 2016). Finally, an increase in savings due to a larger share of the population living longer could lead to a rise in private investment (see below). If any of these interaction effects come to pass, the negative impact of aging on Malaysia’s long-term economic growth prospects will likely be less than projected in Table 2.1. FIGURE 2.17: Baseline and alternative FIGURE 2.18: Baseline and alternative projections for population growth rate, projections for headline GDP growth, Percentage Percentage 1.4 5 1.2 4 1.0 Population growth (%) GDP growth (%) 3 0.8 0.6 2 0.4 1 0.2 0.0 0 2040 2042 2044 2046 2048 2050 2040 2042 2044 2046 2048 2050 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 Baseline Constant population growth Baseline GDP growth Constant population growth Source: Authors’ calculations based on UN WPP. Source: Authors’ calculations based on LTGM. 44 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context FIGURE 2.19: Baseline and alternative FIGURE 2.20: Baseline and alternative projections for ratio of persons in working age to projections for headline GDP growth, Percentage overall population, Percentage 0.1 5 0.0 4 Ratio of persons in working age to -0.1 overall population (%) -0.2 GDP growth (%) 3 -0.3 -0.4 2 -0.5 -0.6 1 -0.7 -0.8 0 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 Baseline Constant growth of ratio of persons Baseline Constant growth of ratio of persons in working age to overall population in working age to overall population Source: Authors’ calculations based on UN WPP. Source: Authors’ calculations based on LTGM. A Silver Lining: Productive and Inclusive Aging for Malaysia 45 Chapter 2: Context Impacts on Public and Private Savings The fiscal effects of aging will likely be consequential for Malaysia. There is little doubt that aging will put more pressure on the already limited fiscal space in Malaysia, on both the revenue and expenditure sides. On the revenue side, a shrinking working-age population and lower economic growth rate will lead to a narrower tax base. This has particularly stark implications for Malaysia given that the country already collects less revenue (as a share of GDP) than many comparator economies. On the expenditure side, the reliance on a defined-contribution provident fund places Malaysia in a relatively unusual situation in which— outside of the civil service pension system—institutions providing old-age income security pose no fiscal risks. Nevertheless, an aging population will necessitate accelerated spending on health care, aged care and civil service pensions. According to projections by the International Monetary Fund (IMF), pension spending in Malaysia will rise to 4.6 percent of GDP by 2030 and to 5.6 percent by 2050 (see IMF 2010 and Figure 2.21). Moreover, Rannan-Eliya et al. (2013) estimate that aging will account for up to one-third of Malaysia’s expected increase in health care-related spending between 2010 and 2030 (see Figure 2.22). In addition to that, aging will likely require higher spending on tax-financed social assistance for older persons (see Section 3.2) and aged care (see Section 3.3), though in both cases from a very low base. FIGURE 2.21: Projected spending on pension, FIGURE 2.22: Projected health spending, Percentage of GDP Percentage of GDP 6 5 5 4 Spending on pensions (% of GDP) Spending on health (% of GDP) 4 3 3 2 2 1 1 0 0 2020 2030 2050 2011 Aging Other 2030 Source: IMF (2010). Source: Authors’ calculations based on Rannan-Eliya et al. (2013). With a shrinking tax base and rising public expenditures related to aging, the fiscal deficit is likely to widen leading to higher long-term debt. In 2020, the COVID-19 pandemic is putting pressure on an already limited fiscal space (see World Bank 2020), and this pressure will likely be further reinforced by aging. Moreover, many of the policy interventions required to mitigate the negative impacts of aging on economic growth, such as raising the quality of education, would require not only better spending but also more outlays of public funds. This does not mean that these policy interventions are not worth pursuing—on the contrary. But it does mean that without proactive policy efforts on the revenue, as well as on the expenditure side, public debt is likely to rise with aging making the debt sustainability objective more and more difficult to achieve. 46 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context In contrast to public savings, aging will likely lead to a rise in private savings. Theoretically, the effects of aging on private savings are not straightforward because aging increases some drivers of savings and lowers others. On the one hand, aging may lead to a decline in aggregate saving because a smaller fraction of the population is earning wages (compositional effect). On the other hand, because of longevity people start to save more for the future, which may lead to a rise in saving (behavioral effect). As such, the dynamics of saving is then dependent on the comparative size of the composition and behavioral effects. Empirically, there is compelling evidence from various countries in East Asia to suggest the behavioral effect is larger than the compositional effect (see Schult 2004, Kinugasa and Mason 2007, Li, Zhang and Zhang 2007 and World Bank 2013). Thus, empirically there is a clear positive association between aging and private savings. Long-term simulations of the dynamics of savings in Malaysia confirm that aging will likely be associated with higher private saving. Using estimated coefficients of the determinants of savings from Abdullah and Ahmad Khan (2010) and following an approach that is methodologically similar to Hevia and Loayza (2013), an extension of the LTGM can be used to simulate long-term aggregate savings rates for Malaysia. The simulations show that aging is projected to lead to a higher long-term national savings rate in Malaysia, which in turn boosts investment and thus long-term growth and real GDP per capita (see Figure 2.23 and Figure 2.24). The actual simulated path of the long-term savings rate rises above 50 percent of GDP by 2050 (“LTGM Savings Extensions Path”). Given that such an outcome appears unlikely, an alternative scenario considers the average path until 2030 and keeps the savings rate constant thereafter (“LTGM Savings Extensions Average Path by 2030”). This alternative scenario still leads to higher headline economic growth than in the baseline simulations. FIGURE 2.23: Baseline and alternative FIGURE 2.24: Baseline and alternative projections for savings rates, Percentage of GDP projections for GDP per capita, 2010 US$ 50 35,000 30,000 GDP per capita (constant 2010 US$) 40 Savings rate (% of GDP) 25,000 30 20,000 15,000 20 10,000 10 5,000 0 0 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 Baseline Baseline LTGM savings extensions path LTGM savings extensions path LTGM savings extensions average path by 2030 LTGM savings extensions average path by 2030 Source: Authors’ calculations based on LTGM. Source: Authors’ calculations based on LTGM. The projection that aging will lead to an increase in savings is consistent with recent studies on the effects of aging on long-term interest rate and monetary policy. Recent studies that document the impact of population aging on monetary policy and its transmission through the economy show that aging most likely leads to a decline in real interest rates (see Eggertson, Mehrotra and Robbins 2019 and A Silver Lining: Productive and Inclusive Aging for Malaysia 47 Chapter 2: Context Carvalho, Ferrero and Nechio 2016). This typically implies a lower natural rate of interest (NRI).10 In fact, Bielecki, Brzoza-Brzezina and Kolasa (2018) show that aging in the euro area—that is, in the monetary union of 19 member states of the European Union—will likely drive down the NRI from 4 percent to 0.4 percent by 2030. The increase in savings due to aging leads to an increase in the supply of loanable funds and thus a lower interest rate. This has monetary policy implications: A lower NRI in any given country would mean less room for the central bank to lower the policy rate, which is usually done to boost employment and nudge inflation toward its target. This suggests that population aging is likely to create challenges for monetary policy makers in the future. 2.3 Socioeconomic Context With Malaysia’s transition to an aging society in 2020, it is vital to address its rapid pace of aging in order to support continued economic growth and enhance social protection. To support an appropriate policy response, there is a need for in-depth understanding about the economic activities and poverty incidence among older persons. It is also necessary to understand how their well-being differs from that of non-older persons in Malaysia, with further disaggregation based on individual and household characteristics including age group, gender, and geographical location. Based on such evidence, policymakers can make short-term and long-term decisions to protect all layers of society based on specific needs and capabilities. This section examines the economic well-being of older Malaysians in terms of employment, income, and poverty, as well as considers the average income (including pensions and social assistance) needed by older persons to fulfill their basic needs and maintain a decent standard of living post-retirement. Household Composition and Income Co-residence between older and non-older persons remains common in Malaysia; as of 2016, only 19 percent of households with at least one older person had no younger members in the household. For older persons, co-residing can be beneficial as it makes it easier to pool resources, thus reducing the likelihood of falling into poverty. Co-residence also facilitates childcare or aged care while larger households, including older persons co-residing with younger persons, can reduce expenditure per capita by sharing communal resources such as housing and utilities. In this report, co-residence refers to older persons living with at least one working-age adult age 15 to 64. Around 23 percent of all households in Malaysia have at least one older person, and of these, 80 percent have at least one older person and at least one working- age adult. In comparison, about 9 percent of older-person households consist of one older person only, and approximately 10 percent have two or more older persons in the same household but no persons below the age of 65. Households comprising older persons and children age 14 or below but no working-age adults are rare (see Table 2.2). Approximately 56 percent of households with older persons are headed by a working-age adult.11 Of the 44 percent of households with older persons that are headed by older persons, a little less than one-half (18.9 percent of all households with older persons) have no household members who are not 10 The NRI is defined as the interest rate at full employment, which drives the conduct of monetary policy: the policy rate is set below the NRI to incentivize spending and investment and raise employment/output and inflation and vice-versa. A lower NRI increases the probability of hitting the Zero Lower Bound, leading to liquidity trap and rendering monetary traditional policy tools ineffective. 11 In the Household Income and Expenditure Survey, the household head is defined as any household members considered as the head of household by other members. The head of household must be an income recipient and age 15 years and over. 48 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context older persons (see Table 2.3). An almost equal number of households headed by older persons contain older persons and working-age adults but no children. A much smaller number of households headed by older persons (5.8 percent of all households that include older persons) are three-generation households that contain older persons, working-age persons, and children. The large majority of three- generation households are headed by a working-age adult. TABLE 2.2: Composition of households with at least one older person, Percentage One older Two older Three or more Household type Total person persons older persons Older persons and working-age adults 38.7 10.6 0.3 49.6 Older persons, children, and working-age adults 23.6 7.0 0.1 30.7 Older persons only 8.7 9.9 0.2 18.9 Older persons and children 0.3 0.4 0.0 0.7 Total 71.3 28.0 0.7 100.0 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). TABLE 2.3: Composition of households with at least one older person by age of household head, Percentage Non-older Older-person Household type Total person head head Older persons and working-age adult 31.5 18.1 49.6 Older persons, children, and working-age adult 25.0 5.8 30.7 Older persons only 0 18.9 18.9 Older persons and children 0 0.7 0.7 Total 56.5 43.5 100.0 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Income receipt rapidly declines around age 60, especially with regard to income from economic activities such as paid work and self-employment. Figure 2.25 and Figure 2.26 show that the proportion of older men and women who receive income from paid employment or self-employment begins to drop rapidly around age 60. Compared with those age 55 to 59, the proportion of men age 60 to 64 with paid employment income is 28.4 percentage points lower. The proportion of men with paid employment income further declines from 15 percent to 1.8 percent between age 65 to 69 and 80 to 84, while in parallel, the percentage of men with self-employment income declines from 37.6 percent to 22.5 percent. This is consistent with the higher rates of poverty found among the oldest age groups (see Figure 2.34 and Figure 2.35 below). For older women, the patterns of decline are less pronounced due to the lower proportion of women with paid employment and self-employment incomes across all age groups. Despite the significant A Silver Lining: Productive and Inclusive Aging for Malaysia 49 Chapter 2: Context FIGURE 2.25: Older men receiving income by FIGURE 2.26: Older women receiving income by income source and age group, Percentage income source and age group, Percentage 100 100 80 80 Percentage of older women Percentage of older men 60 60 40 40 20 20 0 0 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ Age group Age group Paid employment income Self-employment income Paid employment income Self-employment income Asset income Public transfers Asset income Public transfers Private transfers Receives any income Private transfers Receives any income Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Note: Asset income includes interest, dividends, and net proceeds from Note: Asset income includes interest, dividends, and net proceeds from the rental of land or buildings. It does not include imputed rent for owner- the rental of land or buildings. It does not include imputed rent for owner- occupied housing. occupied housing. increase in the overall female labor force participation rate, from 46.8 percent in 2010 to 55.2 percent in 2018, female labor force participation remains low when compared to 80.4 percent of working-age men (see World Bank 2019b). For both genders and for all age groups from 50 to 54 and beyond, asset income and transfer income are more common than paid employment or self-employment income and the receipt of these income types also declines slower with age.12 The receipt of private transfer income declines more rapidly than public transfers. This may be because of older persons moving in with younger family members in their later years, and thus any financial support provided by the younger members is no longer counted as a private interhousehold transfer. A similar mechanism may be at work with public transfers, as the younger family member may be recorded as the official recipient of public transfers that benefit the entire co-resident household. Finally, it is noteworthy that by age 70 to 74 approximately one-third of men and three-fourths of women have no sources of income at all (see Section 3.1 for more details on the employment and earnings pattern of older persons and Section 3.2 for a discussion of income security of older persons). 12 In this analysis, asset income comprises income received as returns to assets owned, including interest received on bank deposits and loans, dividends related to investment in company shares or unit trusts, income from rental of land or buildings, or royalties received. It excludes imputed rent for owner-occupied housing, which may be interpreted as the in-kind return on investment from purchasing the home and is conventionally included in measures of gross household income. Transfer income encompasses both private transfers (transfers made between households or between charities and households) and public transfers (transfers made between the Government and households), and transfers both within Malaysia and between Malaysia and another country. 50 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context Incidence and Depth of Poverty Using the B20 per capita income threshold as the poverty line, in 2016 the poverty rate was 21 percent for households with older persons and 19.7 percent for households without older persons (see Table 2.4). To investigate poverty among older persons, it is useful to consider both households that include older persons and households that are headed by older persons. For this report, the poverty lines are set at the annual pre-transfer per capita income thresholds for the bottom 20 percent (B20) and the bottom 40 percent (B40) of the population.13 Based on the B20 poverty line, older persons and people living in households with them are slightly more likely to be poor (with a poverty rate of 21 percent) than those in households that do not have members age 65 years or above (poverty rate of 19.7 percent). The difference in poverty rates between households with older persons and households without older persons is slightly larger when using the B40 per capita income threshold, at 42.3 percent and 39.3 percent, respectively. Higher poverty rates for households with older persons are attributable in part to the lower household income per capita associated with the rapid decrease in the share of persons receiving any income beyond age 60. TABLE 2.4: Poverty rates for those living in households with and without older persons, Percentage Household type B20 poverty line B40 poverty line No older persons present 19.7 39.3 Older persons present 21.0 42.3 All households 20.0 40.0 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Note: Differences in the poverty rates are statistically significant at the 99-percent confidence level. The poverty incidence curves in Figure 2.27 show that households with older persons have higher poverty incidence regardless of the specific choice of poverty line. This is evident from the fact that the poverty incidence curve for households with older persons is consistently above that for households without older persons. The figure also shows that the difference in poverty rates—measured by the vertical distance between the two curves—is relatively small, ranging from nearly zero for very low poverty lines to about 3 percentage points at a high poverty line of RM10,000 per capita per year (equivalent to RM3,333 per month for a family of four). Thus, the finding that households with older persons are slightly more likely to be poor is robust to the choice of the poverty line. The shape of the poverty incidence curve also implies that there is a sizable number of households with older persons who have incomes that are just slightly above any given poverty line and therefore vulnerable to falling into poverty. Households that are headed by older persons have slightly higher poverty rates than households headed by non-older persons (see Table 2.5). Based on the B20 per capita income threshold, 21.7 percent of persons living in households headed by older persons are poor compared to 19.9 percent of those in households headed by non-older persons. Using the B40 per capita income threshold as the poverty line, 13 The focus on pre-transfer per capita income is relatively unusual for Malaysia but follows the approach suggested in The State of Social Safety Nets 2018 (see World Bank 2018) and the B40 poverty line in World Bank (2020). It is also worth noting that Malaysia’s poverty line income (PLI) based on the 2005 methodology is not used because it is so low that almost no one is poor by that definition. Therefore, differences in poverty rates are practically non-existent, giving little guidance for policy. It is not possible to apply the newer 2019 PLI methodology because details for calculating household-specific PLIs are not yet available. A Silver Lining: Productive and Inclusive Aging for Malaysia 51 Chapter 2: Context FIGURE 2.27: Poverty incidence curves for households with and without older persons 80 60 Cumulative population (%) 40 20 0 2,500 5,000 10,000 20,000 Annual pre-transfer income (2016 MYR per capita) Households with older persons Households without older persons Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). 52 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context TABLE 2.5: Poverty rates for households headed by older and non-older persons, Percentage Household type B20 poverty line B40 poverty line Head non-older person 19.9 39.8 Head older persons 21.7 43.3 All households 20.0 40.0 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Note: Differences in the poverty rates are statistically significant at the 99-percent confidence level. the poverty rates are 43.3 percent for households headed by older persons and 39.8 percent for others. Poverty rates for households headed by older persons are also marginally higher than the poverty rate for all households with older persons. This means that households headed by older persons are slightly more likely to be in the B20 or B40 than households with older persons that are headed by non-older persons. Among households with an older person, poverty is more strongly associated with the presence of children in the household than with that of older persons. Households composed of only older persons and households comprising older persons co-residing with working-age adults but no children have below- average poverty rates of about 12 percent at the B20 poverty line and 30 percent at the B40 poverty line (see Table 2.6). In contrast, poverty rates are much higher among three-generation households (comprising older persons, working-age adults and children). For these households, poverty rates are almost three times higher at the B20 poverty line and almost double at the B40 line.14 This is most likely attributable to many older persons bringing in at least some income that helps support the household’s needs, as opposed to children who generally do not earn any income at all. It could also be an indication that being in a three- generation household—which is becoming less common in Malaysia but still accounts for almost one in three households with older persons (see Box 2.2)—could be an indicator of economic stress.15 TABLE 2.6: Poverty rates by household composition, Percentage Household type B20 poverty line B40 poverty line Older persons and working-age adult 11.7 29.6 Older persons, children, and working-age adult 30.6 55.5 Older persons only 11.6 29.9 Older persons and children 33.1 56.9 Total 21.0 42.3 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). 14 Poverty rates are also elevated among households composed of only older persons and children. This household configuration is relatively rare, accounting for less than 1 percent of households with older persons (see Table 2.4). 15 The pattern that poverty in Malaysia is associated with the presence of children in the household is also confirmed by cross country comparisons using the OECD relative poverty approach that sets the poverty line at 50 percent of national median income and three age groups, children age up to 17 years, working-age adults age 18-65 years, and older persons age 66 and above). These comparisons show that in Malaysia, the relative poverty rate among older persons is slightly higher than that of working-age adults, but relatively low compared to child poverty rates. Furthermore, the poverty rate among older persons is lower in Malaysia than in most upper middle- or high- income comparator countries. A Silver Lining: Productive and Inclusive Aging for Malaysia 53 Chapter 2: Context BOX 2.2 Trends in co-residence, living arrangements and poverty among older persons Patterns in living arrangements among older persons have changed gradually yet distinctly during the last decade. Figure 2.28 shows that this is most evident among households in which older persons co-reside with working-age adults. Among households that include an older person, the share of three-generation households declined from 41.1 percent in 2004 to 30.7 percent in 2016. This was mostly offset by an increase in the share of households comprising only older persons and working-age adults but no children, which increased from 41.5 percent to 49.6 percent over the same period. This is likely due to younger working-age people staying in school longer and starting their own families at a later age. In addition, there was a smaller but still significant increase in the percentage of households composed of only older persons, from 16.4 percent to 18.9 percent of all households with older persons. FIGURE 2.28: Household composition among FIGURE 2.29: Older persons living in households households with at least one older person, with only older persons by age group, Percentage Percentage 100 30 80 25 Household composition (%) Household composition (%) 60 20 40 15 20 10 0 5 2004 2007 2009 2012 2014 2016 Older person, children, and working-age adult 0 Older person and working-age adult 2004 2007 2009 2012 2014 2016 Older person and children Older person only 65-69 70-74 75-79 80-84 85+ Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2004–2016 (DOSM). Amenities Survey 2004–2016 (DOSM). 54 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context The prevalence of older persons living alone or only with other older persons is highest for those in their 70s. Approximately one in four older persons age 75 to 79 live in households composed solely of older persons, a rate that has increased slightly since 2004 (see Figure 2.29). Households comprising only older persons are slightly less common for those age 65 to 69, most likely because many of them live with spouses who are younger than 65. For those age 80 and older, the lower prevalence of older person-only households most likely occurs because of declining income with age and the limited availability of care for older persons in Malaysia. Digging deeper, Figure 2.30 shows that older persons in rural areas are more likely to live in households of only older people than those in urban areas. In rural areas, the percentage of older persons in households with only older persons has stayed fairly constant between 23 percent to 26 percent from 2004 to 2016. In contrast, the rate in urban areas has increased from 15 percent to 22 percent over the same period. This most likely reflects both increasing urbanization and ability for older urban dwellers to live on their own in urban areas rather than living with their children or moving back to the kampung they may have left at a younger age. It may also be due in part to a shrinking pool of working-age persons residing in rural areas as more of them move to urban areas for work. Looking at gender differences, older women are slightly more likely than older men to live in households composed of only older persons, with the prevalence increasing gradually over time for both women and men (see Figure 2.31). One reason for this gender difference is women’s greater longevity. FIGURE 2.30: Older persons living in households FIGURE 2.31: Older persons living in households with only older persons by location, Percentage with only older persons by gender, Percentage 30 30 25 25 20 20 Older persons (%) Older persons (%) 15 15 10 10 5 5 0 0 2004 2007 2009 2012 2014 2016 2004 2007 2009 2012 2014 2016 Urban Rural Men Women Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2004–2016 (DOSM). Amenities Survey 2004–2016 (DOSM). A Silver Lining: Productive and Inclusive Aging for Malaysia 55 Chapter 2: Context BOX 2.2 (continued) 56 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context Also during the last decade, overall poverty rates have declined rapidly and even more so for households that include older persons. Figure 2.32 and Figure 2.33 compare poverty trends from 2009 to 2016 between households with older persons and those without. In this case, the poverty lines are set at the B20 and B40 thresholds for pre-transfer income per capita in 2009 and “anchored” in real terms, and therefore act as absolute poverty lines that are adjusted only for inflation, not for income growth. In 2009, poverty rates were significantly higher among households with older persons— by 3.8 percentage points at the B20 poverty line and 9.2 percentage points at the B40 line. Since then, the incidence of poverty has fallen significantly for both types of households, but particularly so for households with older persons for which it dropped from 23.1 to 4.8 percent according to the “anchored” B20 poverty line and from 57.6 to 17.6 percent per the “anchored” B40 line. As a consequence, the difference in poverty rates across household types also narrowed considerably. FIGURE 2.32: B20 poverty rates among FIGURE 2.33: B40 poverty rates among households with and without older persons, households with and without older persons, Percentage Percentage 50 50 45 45 40 40 35 35 B20 poverty rates (%) B40 poverty rates (%) 30 30 25 25 20 20 15 15 10 10 5 5 0 0 2008 2010 2012 2014 2016 2008 2010 2012 2014 2016 Households with older persons Households with older persons Households without older persons Households without older persons Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2009–2016 (DOSM). Amenities Survey 2009–2016 (DOSM). A Silver Lining: Productive and Inclusive Aging for Malaysia 57 Chapter 2: Context The poverty incidence for older persons rises with age. The higher poverty rates among persons at a more advanced age can be explained by decreasing income with age, as older persons become less involved in paid work and self-employment and have increasing dependence on asset income and public or private transfers. However, this is mitigated to some degree by the fact that most older persons live with a working-age adult and therefore their poverty status depends more on the incomes of the working-age members than on their own incomes. Based on the B20 income threshold, 14 percent of men and women between age 65 and 69 are poor, with the poverty rate rising to 19 to 20 percent by age 85 (see Figure 2.34). When using the B40 income threshold, poverty rates increase with age, from 34 percent for men and women age 65 to 69 to 40 percent among those age 85 or older (see Figure 2.35). The marginally higher poverty rates among men age 75 to 85 years could be attributable to a greater absolute and relative decline in male labor force participation. It bears noting that up until age 85, the poverty rates for older persons are lower than the rates for the overall population. FIGURE 2.34: B20 poverty rates among older FIGURE 2.35: B40 poverty rates among older persons by age group and gender, Percentage persons by age group and gender, Percentage 25 45 B20 poverty rates (%) B40 poverty rates (%) 20 40 15 35 10 30 65-69 70-74 75-79 80-84 85+ 65-69 70-74 75-79 80-84 85+ Age group Age group Women Men Women Men Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Further disaggregation of households headed by older persons by urban-rural location and ethnicity shows higher rates of poverty among rural households as well as among the Bumiputera. Based on the B20 income threshold, 14.3 percent of those in households headed by older persons in urban areas are poor as compared to 34.1 percent in rural areas (see Table 2.7). Using the B40 income threshold, the difference is even more pronounced; in urban areas, 33.6 percent of households headed by older persons live in poverty, as compared to 59.9 among those in rural areas.16 In addition, Table 2.8 shows that 29.2 percent of those in households headed by an older Bumiputera person are poor according to the B20 income threshold, compared to 7.1 percent of Chinese and 13.9 percent of Indians.17 16 Because the B20 and B40 poverty lines used in this analysis are not adjusted for urban/rural price differences, these estimates underestimate urban poverty and overestimate rural poverty by an unknown extent. 17 See Section 3.2 for a comprehensive discussion of the challenges of and policy recommendations to improve income security for older persons of all backgrounds in Malaysia. 58 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 2: Context The average poverty gap is about the same between households with and without older persons. While poverty rates express the percentages of persons or households living below a specified poverty line, poverty gaps combine information about the incidence of poverty (what percentage of the population is poor) with information about the depth of poverty (on average, how far below the poverty line are the poor). Poverty incidence is slightly higher among households with older persons (see Table 2.4 above). In addition, Table 2.9 shows that the poverty gap index is also slightly higher in households with older persons. This is largely because of the higher incidence of poverty—that is, because a higher proportion of households with older persons are below the poverty line—not because they are further below the poverty line than households without an older person. TABLE 2.7: Poverty rates for households headed by older persons by location, Percentage Location B20 poverty line B40 poverty line Urban 14.3 33.6 Rural 34.1 59.9 All households 20.0 40.0 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). TABLE 2.8: Poverty rates for households headed by older persons by ethnicity, Percentage Ethnic group B20 poverty line B40 poverty line Bumiputera 29.2 54.1 Chinese 7.1 22.2 Indian 13.9 34.5 Others 36.0 58.7 Total 21.7 23.3 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). TABLE 2.9: Poverty gaps for households with and without older persons, Percentage Household type B20 poverty line B40 poverty line No older persons present 5.9 14.3 Older persons present 6.3 15.2 All households 6.0 14.5 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Note: Differences in the poverty gaps are statistically significant at the 95-percent confidence level. A Silver Lining: Productive and Inclusive Aging for Malaysia 59 CHAPTER 3 Policy Areas 60 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas 3.1 Employment Labor Market Participation As Malaysians live longer, the right policy environment must be put in place to allow older workers— defined informally as those age 50 and above—to work longer and more productively so that they are able to have sufficient incomes at older ages (see Section 3.2). The better use of the productive potential of older workers can also increase Malaysia’s long-term economic growth. According to the Long-term Growth Model (LTGM) introduced in Section 2.2, gradually raising the minimum retirement age from 60 to 65 by 2040 would increase headline GDP growth by 0.3 percentage points per year. In turn, this would lead to a higher income per capita (US$22,311 vs. US$20,658 in 2040 in the baseline scenario). Figure 3.1 and Figure 3.2 depict a scenario where the working-age population gradually changes from 15 to 59 toward 15 to 64 by 2040. This redefinition leads to an increase in the working-age to total population ratio and a considerable boost in its rate of growth, under the assumption that with healthy aging, older workers remain productive and relatively unscathed by disabilities and the complications of NCDs. In this alternative scenario, the ratio of persons of working-age to the overall population stays positive until 2040. Thereafter, growth in this ratio returns to a negative path, but remains slightly higher than in the baseline scenario. As a result, the increase in the minimum retirement age leads to a better use of Malaysia’s productive potential and thus to a modest but sustained boost in the long-term growth path of both headline and per capita GDP.18 FIGURE 3.1: Baseline and alternative projections FIGURE 3.2: Baseline and alternative for growth in working age to population ratio, projections for GDP per capita, 2010 US$ Percentage 0.6 25,000 Growth in working age to population ratio (%) 0.4 GDP per capita (constant 2010 US$) 20,000 0.2 15,000 0.0 -0.2 10,000 -0.4 5,000 -0.6 -0.8 0 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 Baseline Baseline Retirement age to 64, phased in by 2040 Retirement age to 64, phased in by 2040 Source: Authors’ calculations based on LTGM. Source: Authors’ calculations based on LTGM. 18 The population data used in this report is split into age cohorts of 5 years each, including from age 59 to 64. Thus, the retirement age policy simulation models an increase in the minimum retirement age from 59 to 64, instead off from 60 to 65. Quantitatively, this does not make a difference since what matters for growth is the rate of the change of the of the working-age to total population ratio. It is also worth pointing out that in an alternative scenario in which the higher minimum retirement age is phased in by 2025 instead of 2040, the dynamics would be qualitatively similar, but the long-term growth rate would be boosted by about 0.2 percentage points. A Silver Lining: Productive and Inclusive Aging for Malaysia 61 Chapter 3: Policy Areas Effectively extending productive working lives will require a sound understanding of relevant policy parameters, as well as the patterns of labor force participation, employment, and earnings of older workers. A Malaysian who retires at 55 can expect to live for another 24.6 years. This is the result of a steady increase in life expectancy over the last decades, as well as only very limited changes in relevant policies. Between 1955 to 1960 and 2015 to 2020, life expectancy at birth increased from 58.5 years to 75.9 years. Over the same period, life expectancy at age 55 increased from 73.2 years to 79.6 years. Further extensions in life expectancy are projected for the coming decades. In parallel, the main policy parameters have seen no, or only very gradual, changes. The EPF minimum (full) withdrawal age has been 55 years since the 1950s. The mandatory retirement age for civil servants was gradually increased, but only from 55 to a modest 60, and while a minimum (not mandatory) retirement age for private sector workers was introduced in 2012 this was again set at a very modest level of age 60 (see Figure 3.3). The increased divergence between policy parameters and life expectancy raise questions regarding the income of older persons (discussed in Section 3.2) and their care needs (covered in Section 3.3), as well as the employment and productivity of older workers (discussed in this section). In this context, it is important to note that there is no evidence that increased employment among late-career workers negatively affects the employment prospects of younger workers (see Box 3.1). FIGURE 3.3: Life expectancy and age-related labor and pensions policy parameters, Years 90 85 83 79 80 80 77 81 75 73 73 75 76 Age 70 73 65 60 59 60 55 55 50 1950 - 1955 1955 - 1960 1960 - 1965 1965 - 1970 1970 - 1975 1975 - 1980 1980 - 1985 1985 - 1990 2050 - 2055 1990 - 1995 1995 - 2000 2000 - 2005 2005 - 2010 2010 - 2015 2015 - 2020 2020 - 2025 2025 - 2030 2030 - 2035 2035 - 2040 2040 - 2045 2045 - 2050 Life expectancy at birth Life expectancy at age 55 EPF full withdrawal age Statutory minimum retirement age Civil servant compulsory retirement age Source: Authors’ calculations based on DOSM, EPF and KWAP. Compared to other upper middle- and high-income countries, the employment rate of persons age 50 to 74 in Malaysia is relatively low, partly due to low levels of female labor force participation. In Malaysia, the employment rate—calculated as the share of persons employed at a given age—for persons age 50 to 74 stands at 45.9 percent.19 This rate is higher than Turkey’s rate of 34.6 percent, but lower than the rate in most other upper middle- and high-income countries. In fact, it is about 17 percentage points lower than the Republic of Korea’s employment rate of that age group, which is 62.9 percent. The 19 Often, the employment rate is defined as employment as a share of the labor force (which includes all persons who are employed or unemployed but not those not actively looking for work or not available for work). In this report, the employment rate is instead defined as employment as a share of the population overall or within a specific age group. 62 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas differences between Malaysia and comparator countries are to a large extent driven by Malaysia’s relatively low employment rates for persons between age 50 and 64, particularly among women (see Table 3.1 and Figure 3.7 below). In fact, the female labor force participation rate in Malaysia is low relative to the male labor force participation rate at all ages, and the country has one of the highest gender gaps in labor force participation compared to its ASEAN peers and other comparator countries (see Figure 3.9 below and World Bank 2019b). This contributes to the low employment rates for persons age 50 and 64. In contrast, for persons beyond age 64, Malaysia’s employment rate is close to the average across comparator countries. Table 3.1 also shows that the 25-year gap between the normal retirement age and the life expectancy at 55 is wider in Malaysia than many other countries. For example, in Germany the gap is about 18 years, in Singapore and the United Kingdom it is about 20 years, and in Japan—which has one of the highest life expectancies in the world—it is about 22 years. TABLE 3.1: Employment rate of persons age 50 to 74 by country Normal Life expectancy Employment rate by age group (%) Country retirement age at age 55 50-54 55-64 65-69 70-74 50-74 Male Female Male Female Turkey 50.1 35.3 20.0 12.1 34.6 51.0 48.0 79.4 83.5 Malaysia 66.3 45.2 25.7 17.9 45.9 55.0 55.0 78.5 80.7 United Kingdom 82.0 65.1 21.3 10.8 52.3 65.0 62.7 82.0 84.7 Canada 81.3 62.9 25.3 13.2 53.1 65.0 65.0 82.7 86.0 Australia 79.2 63.8 28.5 13.0 53.2 65.0 65.0 83.6 86.8 Mexico 69.2 55.3 37.6 29.2 53.8 65.0 65.0 78.7 81.5 United States 77.0 63.1 31.9 18.9 54.1 66.0 66.0 80.9 84.3 Singapore 79.3 66.8 26.8 54.4 64.0 64.0 82.9 86.6 Thailand 83.0 67.7 25.2 56.1 55.0 55.0 79.6 83.7 Germany 86.2 71.4 17.0 7.7 57.1 65.5 65.5 81.2 85.1 Chile 74.1 65.6 39.9 26.3 58.4 65.0 65.0 81.2 84.6 Japan 85.4 75.2 46.6 30.2 61.8 65.0 64.0 83.4 88.8 Republic of Korea 77.7 66.8 46.2 34.3 62.9 61.0 61.0 81.9 87.0 Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM), OECD, ILOSTAT, and UN WPP. Note: The normal retirement age is defined as the age at which a person with an uninterrupted career starting at age 22 can retire without any reduction in their pension. For Malaysia, the EPF minimum withdrawal age is listed. For Singapore and Thailand, the employment rates for persons 50 and older and 70 and older are listed instead of the employment rates for persons 50 to 74 and 70 to 74. Arguably, the low employment rate among persons age 50 and above in Malaysia is at least partly due to the relatively low minimum retirement age and EPF minimum withdrawal age. While Malaysia has no statutory retirement age beyond which workers would be prohibited from working, for many employers and employees the minimum retirement age of 60 appears to act as a strong signal. According to a survey conducted by ILMIA in 2018 presented in the National Strategic Development Plan on Ageing Population (ILMIA 2019), 34 percent of persons age 40 and above report that reaching the retirement age would be the A Silver Lining: Productive and Inclusive Aging for Malaysia 63 Chapter 3: Policy Areas BOX 3.1 The lump of labor fallacy The idea that increased labor market participation of older workers will negatively affect the employment prospects of younger people is known as the lump of labor fallacy. While intuitively appealing, there is no solid evidence that increased employment among older workers negatively affects the employment prospects of younger workers—quite the opposite. The lump of labor fallacy is based on the assumption that workers of different ages are substitutes and that there is a fixed number of jobs in an economy. If this was indeed the case, every additional older worker with a job might translate into one fewer younger worker who is employed (see Gruber, Milligan and Wise 2010). The evidence, however, shows otherwise. Contrary to intuition, studies show that a higher labor market participation of older workers tends to increase that of younger workers. More specifically, the labor market participation of older workers can stimulate the economy, creating higher aggregate demand for labor, including that 64 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.4: Employment rates of men age 20- FIGURE 3.5: Employment rates of women age 24 and 55-64 by district, Percentage 20-24 and 55-64 by district, Percentage 100 100 Employment rate of women age 20-24 (%) Employment rate of men age 20-24 (%) 80 80 60 60 40 40 20 20 0 0 0 20 40 60 80 100 0 20 40 60 80 100 Employment rate of men age 55-64 (%) Employment rate of women age 55-64 (%) Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Note: Agriculture sector workers excluded (see Box 3.3). Note: Agriculture sector workers excluded (see Box 3.3). supplied by younger workers. Further, job types differ between older and younger workers, which means that they are not substitutes but complements (see World Bank 2016). An international analysis of the relationship between the employment rate of persons age 55 to 64 and those age 20 to 24 reveals that in all but one country, increases in the employment of older workers had at worst a neutral association, and in most cases, a positive association with the employment rate of young workers (see Gruber, Milligan and Wise 2010). Several country-specific studies, for instance on China, confirm that a higher employment rate of older workers has no negative impact on the employment prospects of younger workers (see Zhang and Zhao 2012 and Munnell and Wu 2013). Higher employment rates among older workers are associated with higher employment rates among younger workers in Malaysia, consistent with the lump of labor fallacy. Based on employment rates by district in Malaysia, Figure 3.4 shows a positive correlation between the employment rates of men age 55 to 64 and 20 to 24, with a correlation coefficient of 0.34. Similarly, Figure 3.5 shows a positive correlation between the employment rates of women age 55 to 64 and 20 to 24, albeit with a smaller correlation coefficient of 0.22. This supports the conclusion that increased labor market participation of older workers is not likely to come at the expense of the employment prospects of younger people in Malaysia, for the aforementioned reasons. A Silver Lining: Productive and Inclusive Aging for Malaysia 65 Chapter 3: Policy Areas primary reason for them to leave the labor force for good. The EPF’s minimum withdrawal age of 55 years similarly acts as a signal to retire. In addition, it provides older workers in formal, stable, and relatively well- paying jobs with the means to leave the labor force when they turn 55 (see Section 3.2). Many countries have increased or are in the process of increasing their normal retirement ages. While efforts to increase Malaysia’s minimum retirement age or minimum EPF withdrawal age might face political economy challenges, experience from pension reforms in different European countries show how the provision of information and the adoption of concepts from behavioral economics can at least partly mitigate some of these challenges (see Boeri and Tabellini 2012 for Italy, Fontoura Gouveia 2017 for Portugal, and Parlevliet 2017 for the Netherlands). Some countries, including Denmark, Italy, and the Netherlands, even aim to increase the normal retirement age beyond 70 years (see OECD 2019). Moreover, a number of OECD countries have made provisions to automatically link the normal retirement age to life expectancy, reducing the political economy challenges of repeated discretionary policy changes (see OECD 2019). Among countries’ objectives for increasing the retirement age is to encourage longer working lives and to increase the financial sustainability of pension systems. To achieve these objectives, alongside gradually increasing the minimum retirement age, some countries have implemented policies to increase the employment prospects of older workers. For instance, Singapore has put in place the Retirement and Re-employment Act. This act requires employers to offer opportunities for eligible employees who turn 62 to continue employment in the company until age 67. If the company is unable to provide a reemployment opportunity for the worker, it will be required to offer the worker an employment assistance payment equivalent to up to 3.5 months of his or her salary, up to a maximum of SGD13,000.20 Beyond age 40, some workers in Malaysia exit the labor market, a process that happens even faster beyond age 55. Figure 3.6 shows that employment rates are generally highest among workers age 25 to 50 and very gradually start to decline after age 40. In 2016, 76.6 percent of those age 40 were employed. In the same year, the employment rate was 68.4 percent, 36.9 percent and 19.3 percent for those age 50, 60 and 70, respectively. For those age 80, the employment rate stood at 7.2 percent. As is evident from these figures, relatively few persons exit the labor market between age 40 and 50, but the pace of decline in the employment rate increases after age 50. In fact, it is most rapid between age 55 and 59, supporting the notion that the EPF minimum withdrawal age of 55 years acts as a signal to retire. A comparison of employment rates by age between 2007 and 2016 reveals a largely similar pattern, with the exception of a marked increase of employment rates for persons between age 56 and 59—which suggests that this change in employment rates was driven by policies affecting the relevant cohorts differently rather than by age effects (see Box 3.2). This pattern suggests that the introduction of the minimum retirement age in 2012 led to an increase in the employment rate for persons age 56 to 59 between 2007 and 2016. Women exit the labor market at much younger ages than men. According to Figure 3.7, the gap in the employment rates of men and women is large at all ages. For instance, in 2016, 58.7 percent of all women age 40 and 17.9 percent of all women age 60 were employed, while this was the case for 95.8 percent of all men age 40 and 59.7 percent of all men age 60. This resulted in a gender gap in employment rates of 37.1 percent and 41.8 percent at age 40 and age 60, respectively. Generally speaking, gender gaps in employment rates narrowed between 2007 and 2016 as female labor force participation rates increased steadily over time (see World Bank 2019b). Nevertheless, pronounced gaps persist across all age groups, indicating distinct employment patterns between men and women. In addition to lower employment rates of women of all age groups, there are also noteworthy differences between three distinct age groups. First, 20 Similarly, other countries have also introduced promotional policies to increase the employment prospects of older workers. In Japan, employers have been given subsidies to hire older workers since 2003, and small or medium-sized companies that provide employees with the opportunity to work until age 70 are provided with variable financial aid, depending on the number of employees covered. In the Republic of Korea, starting 2011, a new system of subsidies aimed at retirement-age extension and retiree re-employment has been adopted in selected occupations, replacing a previous employment grant for older persons (see World Bank 2016). 66 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.6: Employment rate by age, Percentage FIGURE 3.7: Employment rate by gender and age, Percentage 100 100 90 90 80 80 70 70 Employment rate (%) Employment rate (%) 60 60 50 50 40 40 30 30 20 20 10 10 0 0 40 45 50 55 60 65 70 75 80 40 45 50 55 60 65 70 75 80 Age Age 2007 2016 Gap 2016 Men 2007 Women 2007 Men 2016 Women 2016 Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2007 and 2016 (DOSM). Amenities Survey 2007 and 2016 (DOSM). between roughly age 40 and 50, there is a relatively stable and significant gap in employment rates between men and women. Second, between roughly age 50 and 60 the gap is largest as women begin to retire earlier than men. Third, starting at roughly age 60 the gap gradually narrows as the remaining men and women exit the labor market. Data from the Labour Force Survey indicate that housework and family responsibilities— which can be defined to include household chores and caregiving—are the main reason cited by women, including women age 50 and older, for being outside of the labor force (see World Bank 2019b). In contrast, the main reason cited by men age 50 and older is old age. Low levels of female labor force participation throughout working age, combined with women’s early labor market exit leads to a large gender gap in employment rates at older ages. The average retirement age reflects the actual age that people leave employment, as opposed to the legally defined normal retirement age. 21 Figure 3.8 shows that in Malaysia the average retirement age is 67.1 for men whereas among women it is 61. As a result, the difference in the average retirement ages between men in women is 6.1 years. This is substantially larger compared to OECD countries, where the difference is only 1.7 years on average. Compared to other upper middle- and high-income countries, Malaysia’s average retirement age is very low for women and about standard for men. Hence, at 33 percentage points, the gender gap in the employment rate of workers age 55 to 64 is higher in Malaysia than in most comparator countries, including Singapore and Thailand, which have gender gaps of 22 and 25.5 percentage points respectively (see Figure 3.9). Among Malaysia’s three main ethnic groups, Indian men on average retire earliest followed by Chinese men and then by Bumiputera men. Figure 3.10 compares employment rates of men by ethnicity and age while Figure 3.11 does the same for women. Figure 3.10 shows that the employment rate of Indian men has the steepest decline, while the employment rate of Chinese men undergoes a more gradual decline. 21 The average age of retirement (or average effective age of retirement) is calculated as a weighted average of the declines in the participation rate over a 5-year period for people in each 5-year age group between the ages of 40 and 80 at the beginning of the period and those aged 5 years older at the end of the period (OECD 2019). For Malaysia, this is calculated for 4-year intervals due to the availability of data. A Silver Lining: Productive and Inclusive Aging for Malaysia 67 Chapter 3: Policy Areas BOX 3.2 Distinguishing between age and cohort effects To understand changes in the employment patterns of older workers over time, it is important to distinguish between age and cohort effects. An age effect is a change in variable values that is experienced by all birth cohorts as each cohort grows older, independent of the time period. In contrast, a cohort effect is a change which affects populations born at a particular point in time, but is not affected by the process of persons getting older (see Blanchard et al. 1977). Distinguishing between age and cohort effects is not straightforward without access to panel data (see Bell and Jones 2013), but understanding the two concepts can nevertheless be useful to draw some inference from repeated cross-sectional data. This section utilizes repeated cross-sectional data collected from surveys of different individuals in different years. DOSM’s Household Income and Basic Amenities Survey is conducted twice in five years and involves interviews with respondents that are randomly selected in each period to form a nationally representative sample. Hence, for any given year, it is not possible to identify the precise, quantitative contribution of age and cohort effects to the employment rate for different age groups or similar variables. Nevertheless, if similar patterns are observed again and again over several years this suggests the presence of an age effect, while patterns that change frequently over time suggest that there is likely a cohort effect. Presenting data for more than one year reveals, to some degree, patterns that can be attributed to a worker’s life cycle. For instance, a downward trend in employment after the age of 40—which becomes more pronounced after the age of 50—is observed in Figure 3.7 for both men and women and using data for both 2007 and 2016. Additional data not presented in this report show very similar patterns for the years 2010, 2012 and 2014. The combination of this evidence suggests that the general decrease in employment rates as workers get older is not specific to any one cohort but due to an age effect. In contrast, data for 2007 and 2016 also show that while the employment rates of men between age 40 and 55 remained more or less constant, the employment rates of women in this age group increased between 2007 and 2016. This implies that there is likely a cohort effect at play. In other words, women born in later years are comparatively more likely to be employed between age 40 and 55 than those born earlier. 68 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas The employment rate of Bumiputera sharply falls at around 60 but only slowly declines beyond that age. According to the figure, the difference in employment rates between age 59 and 61 is 20.1 percentage points for Bumiputera men, 12.5 percentage points for Chinese men, and 24.7 percentage points for Indian men. In addition, the figure also shows that Bumiputera men age 66 have the highest employment rate, at 46.2 percent, followed by the Chinese men at 42.9 percent, and Indian men at 17.8 percent. FIGURE 3.8: Average age of retirement by gender FIGURE 3.9: Gender gap in employment of and country, Years workers age 55-64 by country, Percentage points 74 45 72 40 70 Gender gap in employment of 35 Average age of retirement people aged 55-64 (p.p.) 68 30 66 25 64 20 62 15 60 58 10 56 5 54 0 Malaysia Germany United Kingdom OECD 2018 Canada Australia Turkey United States Chile Mexico Japan Republic of Korea Canada Germany United Kingdom Australia United States Japan Thailand Republic of Korea Singapore Turkey Malaysia Chile Mexico Male Female Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM), Amenities Survey 2016 (DOSM) and OECD. ILOSTAT and OECD. FIGURE 3.10: Employment rate of men by FIGURE 3.11: Employment rate of women by ethnicity and age, Percentage ethnicity and age, Percentage 100 100 80 80 Employment rate of women (%) Employment rate of men (%) 60 60 40 40 20 20 0 0 40 45 50 55 60 65 70 75 80 40 45 50 55 60 65 70 75 80 Age Age Bumiputera Chinese Indian Bumiputera Chinese Indian Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). A Silver Lining: Productive and Inclusive Aging for Malaysia 69 Chapter 3: Policy Areas There are only small differences in the employment rates of older women between ethnicities, with Indian women having the lowest average employment rate. The employment rates of Bumiputera, Chinese, and Indian women age 40 are 57.9 percent, 61.8 percent, and 56.8 percent respectively (Figure 3.11). Among those age 50, the rates are 48.2 percent, 41.1 percent, and 41.3 percent, respectively. The employment rates of Bumiputera, Chinese, and Indian women age 60 are markedly lower at 18 percent, 18.9 percent, and 14.8 percent, respectively. Across all three ethnicities, the employment rates for women age 70 years are even lower, at 8.9 percent, 3.5 percent, and 2.1 percent, respectively. Beyond age 70, very few Chinese and even fewer Indian women are employed, and employment rates among Bumiputera women continue to fall steadily. FIGURE 3.12: Employment rate of urban and FIGURE 3.13: Employment rate of urban and rural men by age, Percentage rural women by age, Percentage 100 100 80 80 Employment rate of women (%) Employment rate of men (%) 60 60 40 40 20 20 0 0 -20 -20 -40 -40 40 45 50 55 60 65 70 75 80 40 45 50 55 60 65 70 75 80 Age Age Gap Urban Rural Gap Urban Rural Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Older workers from rural areas are more likely to work compared to those from urban areas, particularly beyond the minimum retirement age. Figure 3.12 shows that between age 40 and 55, the employment rates of urban and rural men are both around 94 percent. The difference in the employment rates between urban and rural men start to widen at age 56, with the employment rate of rural men exceeding that of urban men by 4.5 percentage points. This gap increases with age. The employment rate of rural men age 60 years exceeds that of urban men by 18.8 percentage points and the gap averages 28.6 percentage points between age 61 and 80. For women, differences between urban and rural areas are less pronounced (at least among workers age 50 and above) and also exhibit different patterns. Among women in their 40s and 50s, rural women are generally less likely to be employed than urban women, but the reverse is the case for women age 60 or above. According to Figure 3.13, the employment rate of urban women age 40 years exceeds that of rural women by about 11.3 percentage points. In contrast, among women age 61 to 80, rural women are on average 8.2 percentage points more likely to work than urban women. 70 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas Productive Characteristics and Types of Jobs Today’s older workers have relatively low education, but the sustained expansion of educational opportunities since the formation of today’s Malaysia means that this is gradually changing. Since the formation of today’s Malaysia in 1963, educational attainment has been gradually increasing. In 2016, more than 70 percent of persons age 70 and above had at most primary education while the same was true for only about 25 percent of persons in their 40s (see Figure 3.14 and Figure 3.15). This has two implications. First, to improve the potential productivity of today’s older workers, dedicated skills-building initiatives that consider the target group’s relatively low average level of education are a necessity. The need to provide upskilling and reskilling opportunities for older workers has also become more pertinent against the backdrop of rapid technological advancement and the COVID-19 pandemic, which has accelerated changes in skills demand already underway beforehand, such as an increasing demand for digital and socioemotional skills (see World Bank 2020). Second, since today’s younger workers are tomorrow’s older workers, higher levels of educational attainment among comparatively younger cohorts will gradually increase the average level of education among older persons. This will make it more and more imperative to harness the full productive potential of all workers, including older workers. FIGURE 3.14: Share of men by educational FIGURE 3.15: Share of women by educational attainment and age, Percentage attainment and age, Percentage 100 100 90 90 80 80 70 70 Share of women (%) Share of men (%) 60 60 50 50 40 40 30 30 20 20 10 10 0 0 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-80 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-80 Age group Age group No certi cate Primary Secondary Tertiary No certi cate Primary Secondary Tertiary Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Older workers with higher levels of educational attainment are more likely to be employed in high- skilled occupations. Figure 3.16 and Figure 3.17 show that while most workers age 60 to 80 are employed in mid-skilled occupations, those with tertiary education are more likely to be employed in high-skilled occupations. Specifically, 78.1 percent of men and 84.7 percent of women age 60 to 80 with tertiary education are employed in high-skilled occupations. At the same time, sizable shares of men and women in the relevant A Silver Lining: Productive and Inclusive Aging for Malaysia 71 Chapter 3: Policy Areas age range with secondary education are also employed in high-skilled occupations, at 32.6 percent and 23 percent respectively. It is worth noting that skilled employment is usually less physically demanding (see Acemoglu and Autor 2011 and Dicarlo et al. 2016). Thus, it may be relatively more compatible with the physical abilities of older workers. While a relatively large share of older workers are engaged in agriculture, the services sector has become the largest employer of workers age 40 to 80. Figure 3.18 and Figure 3.19 show that the share of men and women age 60 to 80 working in the agriculture sector—a very distinct sector, see Box 3.3—is higher than the share of men and women age 40 to 59 working in the sector. At the same time, the figures also show that the services sector has become the largest employer of workers age 40 to 80, especially among women.22 This corresponds to the structural transformation of Malaysia in the last three decades from an agriculture-driven to a services-driven economy (Abdur Rahman and Schmillen forthcoming). In fact, the share of men and women age 60 to 80 employed in the agriculture sector has been declining in the last decade. The share of men age 60 to 80 employed in the sector declined from 47 percent in 2007 to 35.9 percent in 2016 and that of women from 46 percent in 2007 to 28 percent in 2016. 22 The increasing employment share of the service sector among older workers can be attributed to both demand- and supply-side factors. According to ILMIA (2019), employers in the accommodation and food service activities, education, and human health and social work activities are all comparatively receptive to the hiring of older workers. In parallel, again according to ILMIA (2019) older workers who are interested in self-employment cite accommodation and food services, agriculture, and education as preferred sectors, while those who are interested in salaried employment cite accommodation and food service activities and information technology as preferred sectors. 72 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.16: Share of older men employed FIGURE 3.17: Share of older women employed by educational attainment and skill level, by educational attainment and skill level, Percentage Percentage Tertiary Tertiary Secondary Secondary Primary Primary No certi cate No certi cate 0 20 40 60 80 100 0 20 40 60 80 100 Share of men age 60-80 employed (%) Share of women age 60-80 employed (%) Low-skilled Mid-skilled High-skilled Low-skilled Mid-skilled High-skilled Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). FIGURE 3.18: Share of men employed by sector FIGURE 3.19: Share of women employed by and age, Percentage sector and age, Percentage 100 100 90 90 80 80 Share of women employed (%) Share of men employed (%) 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-80 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-80 Age group Age group Agriculture Mining Manufacturing Agriculture Mining Manufacturing Construction Services Construction Services Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). A Silver Lining: Productive and Inclusive Aging for Malaysia 73 Chapter 3: Policy Areas BOX 3.3 The agricultural labor market Malaysia’s agricultural labor market is very distinct. According to variables describing the structure of employment, Malaysia’s agricultural sector (which encompasses the plantation sector) is very distinct from other sectors. First, the share of self-employed workers is high, at 71.4 percent compared to 17.6 percent in other sectors. Second, the share of workers who are not Malaysian citizens is high. Data from the 2017 Labour Force Survey indicates that 37.5 percent of workers in the agriculture sector are noncitizens, compared to 12.7 percent in other sectors. Third, the agriculture sector employs a higher share of workers age 55 to 64 than all other sectors except for services. Fourth and finally, workers in the agriculture sector are also more likely to work in rural areas, and to have lower earnings. Put together, these distinguishing features set the agriculture sector apart. In addition, they also mean that an analysis of the types of jobs for older workers either needs to carefully consider the interplay between agricultural work and other characteristic elements of their jobs—such as the high rates of self-employment—or exclude the agriculture sector. This report largely follows the second approach. The agriculture sector requires its own set of policies in the face of an aging workforce and a changing economy. The distinct nature of the sector, and particularly the large share of older workers in the sector suggests that general labor market policies and programs adequate for the non-agricultural sectors are not always appropriate for the agriculture sector. Instead, at least some dedicated policies and programs might be needed to address employment- and aging-related issues in the agriculture sector. While a dedicated analysis of the agricultural labor market in Malaysia and its relationship to aging is beyond the scope of this report, it is worth noting that Malaysia is not the only country where the agricultural labor market is very distinct. Malaysia can learn from other countries’ experience in supporting older workers during their transformation out of agriculture (see World Bank 2019a). For example, the French government bought out older farmers and granted them secure lifelong pensions before proceeding to transform the rural farming system through grants and loans to youth entrepreneurs, a process that was instrumental in the consolidation of farms. The Republic of Korea’s government applied a similar strategy, through the provision of retirement packages and the subsidization of monthly contributions by farmers and fishers to the national pension system. The payments allow older farmers to retire, while releasing the land to more productive uses. World Bank (2019a) suggests that it is also worthwhile for Malaysia to consider such strategies, with the objective to protect the livelihoods of older farmers while allowing for the modernization and increased productivity of the agriculture sector. 74 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas As workers age, self-employment becomes more and more prevalent. In Malaysia, as in other countries in the region, the percentage of workers that are self-employed increases significantly with age (see World Bank 2016). Excluding agricultural workers, who are more likely to be self-employed throughout their working lives, Figure 3.20 shows that as workers become older, the shares of both men and women who are employees decline, while the shares of men and women who are self-employed increase, especially markedly just beyond the minimum retirement age of 60. For instance, at age 40 to 44, 81 percent of both men and women are employees, at age 55 to 59 that is the case for 72.5 percent of men and 67.9 percent of women, and at age 60 to 64 it applies to 56.9 percent of men and 50 percent of women. This is consistent with the data collected by ILMIA (2019), which suggest that 53.3 percent of respondents age 40 and above who are interested in working in older ages plan to be self-employed. While no quantitative evidence on the precise magnitude of flows from dependent employment into self-employment among older workers is available, qualitative evidence indicates that many older workers in Malaysia indeed opt for self-employment after they have reached the minimum retirement age, partly due to the prospect of more flexible working hours (ILMIA 2019). FIGURE 3.20: Share of workers by status in FIGURE 3.21: Share of part-time workers by employment, gender and age, Percentage gender and age, Percentage 100 10 90 Share of employed in part-time work (%) 80 8 70 Share employed (%) 60 6 50 40 4 30 20 10 2 0 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-80 0 Age group 40-44 45-49 50-54 55-59 60-64 Age group Male employees Male self-employed Female employees Female self-employed Male Female Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Amenities Survey 2016 (DOSM). Note: Part-time work defined as work of less than 30 hours a week. Note: Agriculture sector workers excluded (see Box 3.3). Agriculture sector workers excluded (see Box 3.3). Part-time work also becomes relatively more common as workers age. Within the group of non- agricultural workers from age 40 to 44, 0.9 percent of men and 4.9 percent of women work part-time, that is 30 hours or less per week. On the other hand, this is the case for 4.5 percent of employed men and 9.6 percent of employed women age 60 to 64 (see Figure 3.21). This pattern is consistent with studies on older workers in the United States (see Penner, Perun and Steuerle 2002), Sweden (see Wadensjö 2006), and Italy (see Trucchi, Fornero and Rossi 2018) that have also found that many older workers are willing to continue working beyond the normal retirement age instead of retiring if they have the option of working less hours. Even among part-time workers, average weekly hours are lower for older workers, particularly among men, while a comparatively lower share of older part-time workers would be willing to work longer hours. Figure A Silver Lining: Productive and Inclusive Aging for Malaysia 75 Chapter 3: Policy Areas 3.22 shows that the average number of working hours per week of part-time workers stands at 20.3 for men age 40 to 44 while it is 18.3 for men age 60 to 64. The average number of working hours per week of women who are part-time workers is 18.5, and is only slightly different for different age groups, although it is lowest for women age 60 to 64, at 18.1. In addition, Figure 3.23 shows that even though the share of part-time workers willing to work longer hours is relatively high across all age groups, it also declines with age. The relevant shares for part-time workers age 60 to 64 are 69.4 percent for men and 58.3 percent for women as compared to 94.5 percent for men and 87.1 percent for women among part-time workers age 40 to 44.23 FIGURE 3.22: Average weekly working hours of FIGURE 3.23: Part-time workers willing to work part-time workers by age, Number longer hours by age, Percentage 22 100 90 21 Share of part-time workers willing to 80 Average hours of work per week 20 70 work longer hours (%) 60 19 50 18 40 17 30 20 16 10 15 0 40-44 45-49 50-54 55-59 60-64 40-44 45-49 50-54 55-59 60-64 Age group Age group Male Female Male Female Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Note: Part-time work is defined as work of less than 30 hours a week. Note: Part-time work defined as work of less than 30 hours a week. Agriculture sector workers excluded (see Box 3.3). Agriculture sector workers excluded (see Box 3.3). Working at older ages, particularly if it is voluntary, is very often associated with greater life satisfaction and slower cognitive decline. Using data on workers in Europe and the United States, Nikolova and Graham (2014) find that workers age 46 to 65, as well as 66 and older who work full-time or voluntarily work part-time have higher levels of life satisfaction and subjective well-being than retirees. Some of the possible reasons for the positive association between working at older ages and subjective well-being are that employment provides social contacts and interactions, personal growth, autonomy, and a sense of purpose (see Nikolova and Graham 2014). Similarly, there are studies covering the United States (see Calvo, Haverstick and Sass 2009 and Bender 2012) and Germany (see Bonsang and Klein 2012) that find that involuntary retirement caused by constraints such as lack of employment options for older workers decreases these workers’ subjective well-being. In another strand of research, working at older ages has been found to be associated with a lower risk of cognitive decline, including with regard to memory and mental health, as well as with higher average physical functioning (see Rohwedder and Willis 2010; Bonsang, Adam and Perelman 2012; Wickrama et al. 2013). 23 Excluding the agriculture sector (see Box 3.3), 1,867 out of 3,167 part-time workers in the Labour Force Survey 2017 responded to the question on willingness to work longer hours. 76 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas On average, workers’ cognitive capabilities change, rather than decline, as they age. Changes in physical and cognitive capabilities that happen as people age, particularly if they continue working, do not necessarily affect productivity in an adverse way. Studies show that even though certain basic and higher- level cognitive functions—such as the speed of information processing and episodic memory—decline with age, other functions like semantic memory, language, and speech, improve with age (see World Bank 2015). Changes in capabilities can also be accommodated by making adjustments at the workplace in ways that enable firms to maintain, or even improve the productivity of older workers. In parallel, research has shown major improvements in cognitive functioning from generation to generation, a phenomenon known as the Flynn effect. Projections show that by 2034 the United Kingdom will have aged chronologically but due to the Flynn effect will in fact have become cognitively younger (see Skirbekk et al. 2013). A Silver Lining: Productive and Inclusive Aging for Malaysia 77 Chapter 3: Policy Areas Many older workers are able to adapt to new technology while in turn, technology holds important potential in improving the productivity of an aging workforce. According to Gordo and Skirbekk (2013), many older workers are able to adapt to new technology irrespective of their level of educational attainment. In turn, there are at least three ways in which technology can potentially improve the productivity of older workers (see ADB 2018). First, improved health and longevity resulting from advancements in medical science and technology can extend productive working lives. Second, technology can transform work and workplaces to be less physically demanding and allows for flexible and remote work, which may better suit the physical capabilities and preferences of older workers. Third, innovative technology for improved delivery of education and skills training, as well as labor market matching technologies can increase the efficiency of job placements. Adaptive technologies at the workplace can augment physical strength as well as visual, hearing, dexterous, and cognitive capacities of older workers. In Malaysia, the relationship between age and employment income is inversely U-shaped, with an earnings premium for more educated older workers. The average monthly employment income of men age 40 to 44 is RM4,587. It increases to RM4,854 for men age 45 to 49, and to RM4,884 for men age 50 to 54, before starting to decline. The average monthly employment income of men age 60 to 64 is RM4,003, and decreases to RM2,941 for men age 75 to 80. Controlling for the fact that older workers are on average relatively less educated, an inverse U-shaped relationship between age and income is observed for both men and women (see Figure 3.24). There is also a large gap between the earnings of men and women of almost all ages (see World Bank 2019b). Further, Figure 3.25 shows that there is some evidence of a seniority wage system in the sense that older workers earn higher incomes than younger workers with the same level of education. This evidence is particularly pronounced among more highly educated workers and among men. Men age 50 to 59 with at least a secondary education earn over 10 percent more than men age 40 to FIGURE 3.24: Monthly employment income FIGURE 3.25: Difference in average monthly by gender and age conditional on workers’ employment income between workers aged 40-49 education, RM and 50-59, Percentage 4,500 25 4,000 20 3,500 15 Predicted monthly income (RM) Differemce in average monthly 3,000 10 employment income (%) 2,500 5 2,000 0 -5 1,500 -10 1,000 -15 500 -20 0 15 20 25 30 35 40 45 50 55 60 65 -25 Age No certi cate Primary Secondary Tertiary Male Female Male Female Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Note: Monthly income predicted based on a regression of the natural log Note: Agriculture sector workers excluded (see Box 3.3). of employment income on age, age-squared, and education of workers excluding the agriculture sector (see Box 3.3). 78 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas 49 with the same level of education. Differences are even more pronounced for men with tertiary education but less so for women and nonexistent or even reversed for workers with no or only primary education. The tentative evidence of a seniority wage system for at least some workers in Malaysia is important, as such a system may adversely affect the retention of older workers if it implies that their wages exceed their productivity. This has prompted some countries to implement policies to reduce seniority-based wages or entitlements (see OECD 2019). Women, workers in rural areas, and less educated workers accumulate lower levels of earnings over their careers. “Synthetic” estimates of lifetime earnings show that women, workers in rural areas, and less educated workers on average accumulate less employment income between age 20 to 59 compared to men, those in urban areas, and those with more education.24 Figure 3.26 shows that, on average, men from urban areas can expect to earn about 1.7 times more than men from rural areas over their careers, while women from urban areas can expect to earn about twice more than their rural counterparts. Similarly, Figure 3.27 shows that both men and women with lower levels of educational attainment tend to accumulate less income over their careers. In both cases, men are more likely to earn more than women, both due to the higher labor force participation rate among men and due to the persistent gender wage gap (World Bank 2019b). The earnings accumulated over a workers’ career directly affects this workers’ ability to have an adequate income during old age, an issue discussed in detail in Section 3.2. FIGURE 3.26: Average accumulated employment FIGURE 3.27: Average accumulated employment income from age 20-54/59 by location and income from age 20-54/59 by education and gender, RM gender, RM 2,000,000 3,000,000 2,500,000 Accumulated employment income (RM) Accumulated employment income (RM) 1,500,000 2,000,000 1,500,000 1,000,000 1,000,000 500,000 500,000 0 No certi cate Primary Secondary Tertiary No certi cate Primary Secondary Tertiary 0 Urban Rural Urban Rural Male Female Male Female Age 54 Age 59 Age 54 Age 59 Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on Household Income and Basic Amenities Survey 2016 (DOSM). Amenities Survey 2016 (DOSM). Note: Agriculture sector workers excluded (see Box 3.3). Note: Agriculture sector workers excluded (see Box 3.3). 24 These estimates are obtained using the working population’s annual earnings at each age and summing up their age-specific average earnings for the years multiplied by the probability of employment (see Cheeseman Day and Newburger 2002 and Schmillen and Stüber 2014). A Silver Lining: Productive and Inclusive Aging for Malaysia 79 Chapter 3: Policy Areas 3.2 Income Security for Older Persons Contributory Social Insurance Over the past 70 years, Malaysia’s institutional framework for providing income security for older persons—including through retirement savings—has remained largely unchanged, in spite of rapid aging. Historically, families in Malaysia often financed the needs of older persons no longer able to work. Rapid aging and growth in the number of older persons have increased the importance of retirement savings arrangements to smooth incomes over the life cycle (see Chapter 1). As Malaysia’s population continues to age and urbanize and as co-residence declines, such savings arrangements in concert with broader social insurance policies must be adjusted to support the country’s increasing number of older persons. Social insurance policies are part of a country’s overall social protection system which also encompasses social assistance and active labor market programs (ALMPs). Contributory social insurance programs encompass old-age and disability pensions and unemployment insurance; non- contributory social assistance programs include cash transfers, school feeding and targeted food assistance; and ALMPs include skills-building initiatives, employment services and wage subsidies. Together, these programs can facilitate the achievement of three objectives: First, they can increase resilience for the vulnerable by providing protection against the impact of declines in well-being. The COVID-19 pandemic shows how volatility increases in an integrated world, thus necessitating adaptive social protection. Second, they can provide equity for the poor through protection against destitution and the promotion of equal opportunity. Third, they can create opportunities for all through measures to build human capital in children and adults, as well as enable men and women to engage in more productive employment (see World Bank 2012). FIGURE 3.28: Stylized overview of Malaysia’s social protection system Social assistance programs Social insurance programs Labor market programs Publicly nances skills-building BSH/BR1M unconditional cash transfers Private sector retirement savings by EPF programs by various ministries Core social assistance bene ts and Levy- nanced skills-building services by JKM Public sector pensions programs by HRDF Voluntary, supplementary pensions Employment services by Education assistance savings arrangements JobsMalaysia and others Disability, survivorship and Malaysians@Work wage subsidies Fuel and other subsidies unemployment insurance by SOCSO (announced) Zakat Other programs Other programs Other programs Source: Authors based on World Bank (2015) 80 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas In Malaysia, a number of different ministries and agencies are responsible for implementing social protection programs, under the coordination of the Malaysia Social Protection Council (MySPC). The implementation of social protection programs faces inherent challenges related to coherence across agencies and programs. In this context, the relaunch of MySPC in 2020 to provide overall coordination has been an important development. Though the effectiveness of MySPC is still largely untested, MySPC could enable Malaysia to move from a set of social protection programs (see Figure 3.28) toward a more coherent social protection system. Social insurance in Malaysia includes the Employees Provident Fund (EPF), the Social Security Organization (SOCSO), the Public Service Pension Fund (Kumpulan Wang Persaraan or KWAP), the Armed Forces Board Fund (Lembaga Tabung Angkatan Tentera or LTAT), and Private Retirement Schemes (PRS). Insurance companies also offer a number of occupational pension schemes and retirement products. EPF is Malaysia’s mandatory private sector retirement savings scheme, while KWAP provides pensions to civil servants, and LTAT provides pensions to the members of the armed forces. The PRS are voluntary, supplementary pensions savings arrangements. SOCSO provides benefits for people with disabilities, survivorship and some health procedures. It also manages the employment insurance system (EIS), introduced in 2018. This report focuses largely on EPF as the retirement savings institution with the largest coverage but also covers SOCSO. Conversely, a discussion of the very specific challenges of ensuring old-age income security for civil servants and members of the armed forces, while safeguarding fiscal sustainability, is beyond the scope of this report. EPF is a contractual savings institution established for the purposes of ensuring that citizens save for old age and other contingencies including housing, education and select medical procedures. It is a defined-contribution provident fund with a regular minimum contribution rate of 11 percent for employees and 12 to 13 percent for employers.25 There is no ceiling on wages subject to such contributions. Seventy percent of contributions are placed in Account 1, which is dedicated to old-age benefits, while the remaining 30 percent goes into Account 2, which can be used for home purchase or loans, tuition for tertiary education, and payment of the costs of specified treatments for major medical procedures and illnesses. The self-employed can contribute voluntarily to EPF. Balances are offered as lump-sums upon reaching a minimum withdrawal age and retirees can choose to make periodic withdrawals, though annuity options are not provided. Thanks to its strong governance structure, diversified investment strategy, and operational effectiveness, EPF has grown from a small pension fund set up in 1951 to one the largest pension funds in the developing world (see World Bank 2017a). EPF encourages voluntary contributions and has a relatively modest voluntary matching contribution scheme (i-Saraan) for eligible workers. EPF members who are self-employed and do not earn a regular income can make voluntary contributions toward retirement (up to RM60,000 per year), and receive financial support from the Government. The Government provides a 15 percent matching contribution for business owners, freelancers, professional accountants, doctors, lawyers, insurance/unit trust/property agents, and other eligible workers up to a maximum of RM250 per year. Selected Grab drivers (so-called Gold- and Platinum-tier drivers) who register with EPF’s i-Saraan scheme also receive an additional 5 percent matching contribution, up to a maximum of RM80 per year. 25 The statutory contribution rate for employees has been lowered from 11 percent to 7 percent from April to December 2020, as part of the Government’s response to COVID-19 (see World Bank 2020). A Silver Lining: Productive and Inclusive Aging for Malaysia 81 Chapter 3: Policy Areas In Malaysia, while the majority of the labor force is covered by some form of retirement savings scheme, coverage is still far from universal. In 2019, 11.9 percent of the labor force consisted of civil servants or members of the armed forces, while 48.9 percent contributed to EPF. This implies that 39.2 percent of the labor force were neither covered by EPF nor pension arrangements for civil servants or members of the armed forces (see Figure 3.29). Although workers contributing to EPF are technically “covered” by social insurance arrangements, many move in and out of standard employment. Thus, for many of them, their contribution histories tend to be inadequate to support their retirement (see Figure 3.31 and Figure 3.32 below). The relatively high and persistent coverage gap might be at least in part be due to the high prevalence of non-standard forms of work coupled with the only modest success of the i-Saraan scheme. For instance, according to the Labour Force Survey, the share of own-account workers (that is, self- employed persons without paid employees) as a proportion of all workers is relatively high and growing, increasing from 16.4 percent in 2010 to 18.1 percent in 2019 (see Figure 3.30). This translates into 2.7 million own-account workers in 2019. At the same time, there were only 137,000 EPF members who contributed to i-Saraan in 2019. SOCSO provides employment injury insurance, invalidity, survivorship and unemployment insurance to contributing members. Contributions are mandatory for formal private sector workers. Employers contribute 1.75 percent of wages and employees 0.5 percent of wages to the employment injury and invalidity scheme. The wage base has a limit of RM4,000 per month. The employment injury scheme covers commuting accidents, industrial accidents, and occupational diseases. The benefits include a medical benefit, temporary disability benefit, permanent disability benefit, constant attendance allowance, vocational rehabilitation, a return to work program, a dependents’ benefit, a funeral benefit, and an education benefit. In addition, the invalidity scheme covers incapacity to work or death by causes not related to employment. Under this scheme, there are an invalidity pension, an invalidity grant, an attendance allowance, a survivors’ pension, a funeral benefit, facilities for physical or vocational rehabilitation and dialysis, and an education benefit. FIGURE 3.29: Coverage of retirement savings FIGURE 3.30: Share of own-account workers institutions, Share of labor force among all workers by gender, Percentage 25 20 Share of own-account workers (%) 39.2% 15 48.9% 10 11.9% 5 0 EPF contributors 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Civil Servants and Armed Forces Uncovered Men Women Source: Authors’ calculations based on EPF, KWAP and DOSM. Source: World Bank (2019b). 82 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas On top of the employment injury and invalidity schemes, SOCSO has supported unemployed workers through the EIS since 2018. This has an employer contribution rate of 0.2 percent of wages and an employee contribution rate also of 0.2 percent and provides unemployment insurance coverage and ALMPs. In 2017, SOCSO covered about 43 percent of the labor force and expanded its social security protection scheme to the self-employed. In 2017, SOCSO had 16.5 million registered employees of which 6.8 million were considered active. It had about 569,000 beneficiaries, almost half of which were survivors. The criteria for required contributions in SOCSO is not aligned with the EPF and this partially explains the differences in coverage between the institutions. As an important recent initiative, in 2017 the employment injury scheme began providing benefits to self-employed taxi, Uber and Grab drivers. The contribution rate under this initiative was set at 1.25 percent of earnings per month and taxi, Uber and Grab drivers were given the option to select one of four salary options. The intention in 2017 was that by the year 2020, the scheme would be expanded to other informal sectors such as fishermen, farmers, hawkers, artists and others. Occupational schemes, in particular the pension scheme for the civil service, are an important part of the landscape of retirement savings and old-age income protection. The civil service pension scheme is a defined-benefit occupational retirement benefit program. As the most important provider of annuitized benefits, it provides benefits both to retired civil servants and their dependents. The scheme provides relatively high replacement rates and an annuitized retirement benefit for almost 10 percent of the labor force. The scheme is non-contributory and benefits largely are paid out of the current budget. The pension benefit is 1/600 of the last drawn salary multiplied by the months of pensionable service (up to a maximum of 60 percent of the last salary after 30 years of service). For instance, a civil servant with a last salary of RM5,000 and 20 years of pensionable service will receive a pension of RM2,000. The last salary is used as the basis for pension determination, a practice which many countries are moving away from to improve predictability and fairness. Retirees also receive a lump-sum gratuity which is equal to 7.5 percent times the number of months of pensionable service times the last salary. For instance, the civil servant with a last salary of RM5,000 and 20 years of pensionable service will receive a gratuity of RM90,000. Pension rights vest on completing three years of service, and early retirement on completing ten years. Retirement is compulsory for public servants at age 60 though many qualify for retirement at an earlier age. There is a minimum pension of MYR820 per month for those with 25 years of service. There are concerns that the civil service pension scheme may act as an obstacle to labor mobility and with regard to its sustainability and the predictability of benefits. There is no transferability of accrued rights between the civil service pension scheme and EPF though many civil servants opt to make contributions to EPF for a 10-year period before vesting and moving to the civil service scheme. Moreover, workers who defer their pension by moving to the private sector face significant portability losses. Many countries over the past few decades have undertaken measures either to harmonize the provisions between civil servant schemes and national social security schemes or to merge such schemes and establish a supplemental scheme for civil servants. In addition to the issue of labor mobility, there are concerns regarding the civil service pension scheme’s sustainability (see Section 2.2 and studies cited therein) as well as the predictability of its benefits. Therefore, it would be useful to further review the long-term fiscal cost attributable to the scheme as well as the anticipated levels of income replacement in retirement. However, such a review and the development of specific policy recommendations regarding the civil service pension scheme are beyond the scope of this report. A Silver Lining: Productive and Inclusive Aging for Malaysia 83 Chapter 3: Policy Areas Complementing EPF, the civil service pension scheme and SOCSO, Private Retirement Schemes (PRS) serve an important function by enabling Malaysians to save additional funds for retirement. PRS are voluntary, supplementary, and tax-advantaged savings vehicles established in 2012 to improve the adequacy of pensions savings. In March 2020, there were eight PRS providers with assets of about RM3.5 million and 455,000 members (or about 2.8 percent of the labor force). Among the primary criteria to evaluate the performance of Malaysia’s system of contributory retirement savings and to identify potential adjustments are the adequacy, coverage, and affordability of the system (see World Bank 2008). An adequate system with high coverage is one that provides benefits sufficient to prevent old-age poverty to the full breadth of the population, in addition to providing a reliable means to smooth lifetime consumption for the vast majority of the population. An affordable system is one that is within the financing capacity of individuals and society and that does not unduly displace other social or economic imperatives or have untenable fiscal consequences. Adequacy and coverage of Malaysia’s system of contributory retirement savings are major challenges. The distribution of balances for active EPF contributors at age 54 suggests that a majority of retirees will receive very low benefits in retirement from their EPF accounts (see Figure 3.31). Over half of members at age 54 have balances of under RM150,000 and almost three quarters have balances under RM250,000. Translated into an indexed annuity, almost three quarters of workers retiring with active EPF accounts will have a monthly benefit of less than RM1,050, only slightly more than the PLI of RM980 before its recent revision (see World Bank forthcoming). Moreover, over 40 percent of EPF retirees would have an annuitized income of less than RM420 per month, only slightly higher than the Department of Social Welfare’s (Jabatan Kebajikan Masyarakat or JKM) Financial Assistance for Older Persons’ (Bantuan Orang Tua or BOT) benefit of RM350 per month (see below). FIGURE 3.31: Active EPF members age 54 and FIGURE 3.32: Simulated replacement rates by estimated pensions, Number and RM EPF minimum withdrawal age, Percentage 25,000 4,500 140 4,000 Estimated monthly pension (RM) Active EPF members age 54 20,000 3,500 120 3,000 Replacement rates (%) 15,000 2,500 100 2,000 10,000 1,500 80 5,000 1,000 500 60 0 0 40 less than 50 50-100 100-150 150-200 200-250 250-500 500-1,000 1,000+ 20 50 52 54 56 58 60 62 64 66 68 70 72 74 EPF account balances (RM'000) Age Number of active members age 54 Indexed annuity (amount per month) 2% Real Return 3% Real Return Source: Authors’ calculations based on EPF. Source: Authors’ calculations based on EPF. 84 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas While the real rates of return on individual balances have been substantial, the low minimum withdrawal age has been one of the key reasons for low balances. The minimum withdrawal age is 50 for Account 2 and 55 for Account 1, which is very low by international and regional standards and also lower than the minimum retirement age of 60 (see Section 3.1). Along with the fact that many EPF members engage in intermittent formal work, this means that the contribution rate would have to be inordinately high to support a meaningful replacement rate as a percentage of final salaries. Life expectancy at age 55 is 24.5 years. Thus, many workers will spend longer in retirement than as active contributors. Increasing the minimum withdrawal age from 55 to 65 could almost double the effective replacement rate for individuals who work continuously from age 30 (see Figure 3.32). This is due to three effects: individuals could contribute for an additional 10 years, substantially increasing balances at retirement; they would reap an additional 10 years of dividends; and the payout period would be substantially shortened.26 Another reason for low benefits is that many workers do not contribute to EPF for part or all of their working lives. Malaysia’s coverage of retirement savings institutions is slightly below expectations when benchmarked against other countries (see Figure 3.33).27 On average, countries with a higher GDP per capita have higher coverage by retirement savings institutions. This is primarily due to the positive correlation between GDP per capita and the proportion of workers in “standard” employment relationships— characterized by long-term, full-time wage employment with a single employer—which are most amenable to contributory retirement savings schemes. The pattern that countries with a higher GDP per capita have higher coverage also reflects that many upper middle- and high-income countries have been able to extend the coverage of retirement savings institutions to the self-employed. Malaysia falls slightly short of the global benchmark; it would need to increase coverage by about one percentage point to align with it. To achieve more adequate benefits, it would also need workers to contribute longer. Workers from low-income households have both lower rates of coverage with EPF and lower balances at age 54 (see Figure 3.34). Those with lower incomes are less likely to be in “standard” employment relationships covered by EPF. Average contributions per person are also relatively low among the B40, resulting in low balances. The low coverage and adequacy amongst the B40 suggest that measures are needed to increase coverage amongst workers outside of “standard” employment relationships. It also suggests that a substantial portion of older persons will need public assistance in addition to EPF payouts. Women also have both lower rates of EPF coverage and lower balances. There are two reasons why a disproportionate number of women are uncovered by EPF and other forms of social insurance. First, as already mentioned in Section 2.2 and Section 3.1, the female labor force participation rate in Malaysia is comparatively low. Second, many of the women who participate in the labor market are self-employed or engaged as unpaid family workers or in the informal sector and thus, outside of “standard” employment relationships. In addition, women who are covered by social insurance, on average, have lower EPF balances than men, partly due to the prevailing gender wage gap (see World Bank 2019b). As a result, the average EPF balance at age 54 is RM177,000 for women and RM233,000 for men.28 Taking a broader view on how well-protected women are by pension systems, Chlon-Dominczak (2016) ranks eight countries in East Asia according to (i) the pension coverage, (ii) the expected years in wage employment, (iii) the projected pension benefit for female average earners, (iv) the difference between the level of men’s and women’s pension benefits at average wage levels (v) the degree of redistribution toward retirees who earned lower wages, and (vi) access to and level of survivor benefits. According to this ranking, women are in an overall relatively 26 Anecdotal evidence suggests that the frequent underreporting of earnings by employers also contributes to low balances. 27 The number may overstate coverage for Malaysia as it is based on the number of people who made at least one contribution in the 12 months to December 31 2019. The number of people who make consistent contributions is likely to be lower as many EPF members engage in intermittent formal work. 28 In 2020, a new rule was enacted enabling a male EPF member to divert 2 percent of his monthly earnings (from the total 11 percent employee contribution rate) into the EPF account(s) of his lawful wife or wives. This measure is meant to increase retirement savings among women. A Silver Lining: Productive and Inclusive Aging for Malaysia 85 Chapter 3: Policy Areas FIGURE 3.33: Labor force covered by retirement FIGURE 3.34: Working age individuals savings institutions, Percentage contributing to EPF and average contributions per capita per month, Percentage and RM 100 600 50 Labor force covered by retirement savings institutions (%) 90 500 40 Average EPF contributions 80 per capita per month (RM) Working age individuals contributing to EPF (%) 70 400 30 60 Malaysia 300 50 20 200 40 10 30 100 20 0 0 10 1 2 3 4 5 6 7 8 9 10 0 Household income decile 0 50,000 100,000 150,000 Average monthly EPF contributions per capita (RM) GDP per capita (2017 US$ adjusted by purchasing power parity) Share of working age individuals contributing (%) Source: Authors’ calculations based on World Bank Pensions Database and Source: Authors’ calculations based on Household Income and Basic World Development Indicators. Amenities Survey 2016 (DOSM). good situation in Mongolia, China and Thailand. In contrast, Malaysia takes the seventh place overall, and is placed last in terms of coverage, expected years in wage employment, and projected pension benefits. EPF balances are offered as lump sums and not as annuities or phased withdrawals limiting consumption smoothing during retirement or protection against longevity risks. Retirees can voluntarily leave their balances in their EPF accounts and while some do, most must bear the full burden of investment and longevity risks in retirement. Rabi et al. (2019) estimate that more than two thirds of EPF members withdraw their full balances at retirement. This is a material weakness that exposes retirees to substantial risk, even those who may have accumulated substantial balances for retirement. Globally, most of the countries that historically had provident funds providing lump sum benefits have since transitioned either to phased withdrawals or provide annuitized benefits. In the short term, the COVID-19 crisis may exacerbate the vulnerability of older persons while temporarily reducing savings for retirement among the active labor force. Temporarily, retirement savings will be reduced through declining EPF contribution rates, withdrawals from EPF accounts, and declines in asset prices. In fact, reduced contribution rates to and early withdrawal modalities from EPF accounts formed an important part of Malaysia’s response to COVID-19. EPF members were given the option to reduce their contribution from 11 percent to 7 percent of wages from April to December 2020. In parallel, EPF members were also allowed to withdraw RM500 per month from April 2020 to March 2021. These measures were intended to increase households’ liquidity during the crisis, and are estimated to amount to RM50 billion. Fortunately, retirement savings are intended to meet long-term needs and it is expected that most people with adequate savings will be able to withstand these short-term adjustments, just as they have in earlier economic crises. 86 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas The combined employer/employee EPF contribution rate of 23 to 24 percent (plus 3.5 percent of monthly wages up to RM4,000 for SOCSO) is relatively high, posing a challenge to the affordability of Malaysia’s contributory retirement savings schemes. Malaysia has understandably kept the contribution rate relatively high with the aim to achieve improved adequacy in the face of a very low minimum withdrawal age. However, this has created challenges with regard to affordability. Potentially, it might also deter the formalization of some employment relationships and negatively impact Malaysia’s international competitiveness. EPF contributions, accumulations and distributions are largely exempt from taxation. Employers’ and employees’ contributions to the EPF are deductible expenses for purposes of corporate income taxes up to a maximum of RM4,000 per year. The benefits of tax deductibility of contributions for personal income tax most likely benefits individuals with higher incomes, as income tax rates are highest for these individuals. An individual can deduct RM9,000 per year from his or her income taxes both for him- or herself and a dependent spouse. There is additional tax deductibility of EPF and life insurance contributions of up to RM7,000 per year, for SOCSO contributions of up to RM250 per year, as well for other dependents based on other criteria. EPF dividend yields and lump-sum benefit distributions are not subject to personal taxation. Overall, this tax treatment provides an effective incentive and thereby improves the effective affordability of the EPF scheme for workers. However, this comes at the expense of the generation of tax revenues. Non-Contributory Social Assistance As most older persons in Malaysia have either no pension or an inadequate EPF balance, it is essential that non-contributory social assistance is strengthened to protect older persons against poverty and destitution. Over time, Malaysia’s social assistance programs have experienced a considerable degree of evolution and proliferation. The country’s largest social assistance program is the Cost of Living Aid Program (Bantuan Sara Hidup or BSH). BSH is an unconditional cash transfer program designed to benefit the B40. More closely related to the protection of poor and vulnerable older persons, JKM—under the Ministry of Women, Family and Community Development (MWFCD)—provides a core set of social assistance benefits and welfare services. In total, it implements ten programs intended to benefit specific vulnerable groups, including children, the disabled, and the chronically ill. Importantly, JKM provides cash assistance for poor and vulnerable older persons through BOT. To qualify, individuals must be age 60 or above and have no income or family which they can rely upon for support. JKM also provides substantial social assistance to poor and vulnerable older persons through its Public Assistance program directed at various vulnerable groups facing financial difficulties (Bantuan Am or BA). About three times as many older persons benefit from BOT than from BA (see Hamid 2020). For BOT beneficiary selection, JKM utilizes a combination of income and categorical targeting. For instance, BOT provides beneficiaries with a cash transfer of RM350 per month and is targeted at persons over 60 with no fixed income to sustain their livelihood and no family support. Enrolment into JKM programs is managed by eBantuan, the JKM’s information system and beneficiary registry. All information provided by the prospective beneficiary is verified through a home visit. To qualify for BOT like for most JKM programs, household incomes must be below the PLI of RM980 before its recent revision, though there has been a discussion that the program qualification requirements or benefit levels might be adjusted to account for the new PLI. In addition to JKM’s outreach efforts and proactive applications by prospective beneficiaries, authorities at the district level can also make recommendations for inclusion into BOT and other JKM programs. A Silver Lining: Productive and Inclusive Aging for Malaysia 87 Chapter 3: Policy Areas Overall, there were almost 500,000 beneficiaries of JKM programs in 2018, with a total expenditure on these programs of about RM1.7 billion. In 2018, the total expenditure on all 10 JKM programs stood at 26.7 percent of the allocation for BSH in the same year. As shown in Figure 3.35 and Figure 3.36, there were about 135,000 beneficiaries of BOT in 2018, equivalent to 27.6 percent of the total number of JKM beneficiaries. In the same year, Government expenditure on BOT amounted to about RM554 million, equivalent to 32.5 percent of JKM’s total expenditure on social assistance. Besides BOT, BA has in recent years been the program that has had the largest number of beneficiaries and accounted for the highest share of expenditures. In 2018, the total number of beneficiaries of BA stood at 106,000, with some individuals potentially being beneficiaries of multiple programs. In the same year, expenditures on BA amounted to RM266 million. The number of older persons receiving social assistance benefits through BOT almost quintupled in the period from 2008 to 2012, but since then, the number has remained roughly constant. Figure 3.35 and Figure 3.36 show that the number of JKM beneficiaries increased from 221,602 in 2008 to 486,858 in 2018. Over the same period, there was also a marked increase in total expenditure. The most significant increase in coverage and spending occurred between 2008 and 2012. Since then, coverage and spending have been mostly flat, except in 2018, when expenditure increased by close to RM200 million. In terms of individual programs, the number of BOT beneficiaries saw a particularly significant increase—by more than 120,000—between 2008 and 2012 (from 31,000 to 152,000). As a result, BOT has become the largest JKM program with expenditures of RM554 million in 2018 as compared to RM78 million in 2008, possibly reflecting the fact that most older persons in Malaysia have either no pension or an inadequate EPF distribution. While BOT and other JKM programs are highly progressive, coverage among older persons is very low, even among the poorest and most vulnerable due to their relatively small budget allocation. Figure 3.37 shows that 34.4 percent of households with members age 65 and older in the bottom decile according to their income per capita receive BOT or some other form of assistance from JKM, with this share falling with each successive income decile. In total, 19.2 percent of B40 households with older persons are FIGURE 3.35: Beneficiaries of JKM social FIGURE 3.36: Expenditures on JKM social assistance programs, Number (’000) assistance programs, RM (million) 600 1,800 1,600 500 Expenditures on JKM social assistance 1,400 Number of beneficiaries (’000) 400 1,200 programs (RM million) 1,000 300 800 200 600 400 100 200 0 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 BOT BA Others BOT BA Others Source: JKM. Source: JKM. 88 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.37: Share of households with FIGURE 3.38: Illustration of the tradeoffs in older persons covered by JKM programs and policy parameters for social pension beneficiary incidence, Percentage 40 Share of older persons receiving JKM transfers Bene t level + 35 Size of target population and beneficiary incidence (%) 30 25 20 15 10 5 Fiscal cost 0 1 2 3 4 5 6 7 8 9 10 Income deciles Share of households with older persons receiving JKM transfers (%) Bene ciary incidence (%) Source: Authors’ calculations based on Household Income and Basic Source: Authors. Amenities Survey 2016 (DOSM). covered. The low level of coverage is unsurprising given the programs’ relatively small budget allocation and narrow eligibility criteria. The average monthly transfer per person under the JKM programs is RM292, or about 70 percent of the average post-transfer income for JKM beneficiaries in the B20. To assess the performance of the JKM programs, it is illuminating to analyze their beneficiary incidence as a yardstick of the share of beneficiaries that are in a certain income bracket. According to Figure 3.37, JKM programs are progressive, with the bulk of social assistance being channeled to the B20. In fact, about 26.6 percent of all households with older persons who are covered by JKM programs are in the bottom 10 percent of the income distribution. That said, about 28.8 percent of covered households are not even in the bottom 40 percent of the income distribution. This suggests that there is still potential to transform these programs so that older persons most in need make up an even higher share of beneficiaries. The need for strengthening social assistance also follows from the significant and potentially growing vulnerability of the current and future older persons in Malaysia. Section 2.3 shows that the extent of vulnerability of older persons is already significant, even though poverty rates among households with older persons are only slightly higher than among the population as a whole. Several factors, including demographics, will shape how the vulnerability of older persons will evolve. As outlined in Section 2.1, Malaysia’s old-age dependency ratio is rapidly growing. As the ratio of older persons to the working-age population increases and families accordingly struggle to cope, this can potentially increase the financial vulnerability of older persons. In addition, household incomes are becoming more volatile both globally and in Malaysia, as evidenced by the COVID-19 crisis. As household incomes become more volatile, older persons that cannot work or have no other source of income become increasingly vulnerable. In addition to the gaps in coverage and adequacy of contributory retirement savings arrangements and existing social assistance programs for older persons, the significant and potentially growing vulnerability of current and future older persons provides a rationale for strengthening social assistance, as well as for prioritizing the poorest households. A Silver Lining: Productive and Inclusive Aging for Malaysia 89 Chapter 3: Policy Areas There are a range of issues and options to consider when strengthening social assistance for older persons. In recent years, more and more countries have introduced broad non-contributory social assistance programs for older persons—often labelled “social pensions”. World Bank (2016) notes that the only countries in East Asia and Pacific with substantial pension coverage among poorer households are those with broad social pensions, such as China, Samoa, Thailand, and Timor-Leste, or with special treatment and subsidization of certain groups, such as Mongolia. If Malaysia is to follow this international trend, it will need to weigh the size of the target population, the benefit level, and the fiscal costs (see Figure 3.38). In addition, it will need to consider the benefit level and coverage for other contributory and non-contributory old-age income security schemes. There are four (not necessarily mutually exclusive) basic types of social pensions to consider: • A universal social pension covers all persons above a certain age, usually with a flat benefit. It has the least effect on the incentives for work because the benefit is provided to all, regardless of economic circumstances or sources of retirement income. A universal benefit is also the easiest to administer. On the other hand, a universal benefit is fiscally the costliest so that achieving universality often necessitates reducing the amount of the benefit, or raising the eligibility age, or both. One approach that some countries (for instance Thailand) use to mitigate the cost is to have tiered benefits for different sub- groups among older persons, escalating the flat payment as people age. A universal social pension might also be seen as inequitable by some as even older persons with substantial pensions or other income receive a tax-financed social assistance benefit. • A pensions-tested social pension covers older persons neither receiving a sufficient civil service pension nor having a sufficient balance with EPF to guarantee old-age income security. The rationale behind this type of benefit is that older persons whose livelihoods are already ensured through 90 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas contributory retirement savings schemes or other mechanisms do not need to receive a tax-financed social pension. A well-designed benefit would also incorporate some tapering, to avoid a harsh cliff for those with EPF balances just above the relevant threshold. Although a pensions-tested benefit is not as costly as a universal benefit, a large share of older persons would likely still need to be covered given that most older persons have either no pension or an inadequate EPF balance. • A means-tested social pension targeted at older persons in poor or vulnerable households. In a sense, the current BOT program is a means-tested social pension targeted at very poor older persons who have no other source of support from family or other care providers. However, targeting is very narrow indeed as only 4 percent of persons 60 or older in Malaysia receive BOT. In contrast, in many countries means testing is used to target a much broader group of poor, as well vulnerable older persons that can easily fall into poverty in the face of hardship. • A supplemental social pension is provided to older persons in poor or vulnerable households qualifying for other forms of social assistance. The rationale behind this approach is that it can ensure the livelihoods of older persons as a particularly vulnerable group and minimizes the administrative burden as compared to the establishment of an entirely new pensions-tested or means-tested social pension. At the same time, this option requires a strong implementation performance of the social assistance program that it is to supplement. Rough cost projections suggest that the introduction of a social pension in Malaysia would only be fiscally sustainable if there was a limit to the size of the benefit or the size of the target population. For the purposes of this report, rough projections compare the fiscal costs of four possible options of a social pension that differ with regard to the benefit amount (RM350 per month as with BOT vs. RM769 per month to cover 30 percent of the median wage according to EPF data), eligibility age (60 and above vs. 65 and above) and the targeting (4 percent of the relevant age segment as with BOT vs. 40 percent as with BSH vs. universal).29 Of course various other parameters could be considered. In addition, the cost projections do not consider a pensions-tested social pension, explicitly consider the administrative costs of different options, or make any kind of statements with regard to whether a social pension should be introduced as an entirely new program or as a supplemental social pension. Nevertheless, a number of illuminating insights emerge (see Figure 3.39): • As a benchmark, covering 4 percent of all persons age 60 and above and providing a benefit of RM350 per month as with BOT requires about 0.2 percent of fiscal revenues in 2020 (or 0.04 percent of GDP). Henceforth, costs would modestly rise in line with aging to 0.8 percent of projected fiscal revenues in 2050. This option would be fiscally sustainable but leave most poor and vulnerable older persons without income protection. • A social pension that would extend the coverage of the current benefit of RM350 per month to 40 percent of persons age 65 and above would increase fiscal costs to about 1.6 percent of fiscal revenues in 2020 (or 0.2 percent of GDP). Henceforth, costs would rise in line with aging to 4.7 percent of projected fiscal revenues in 2050. This option would likely be fiscally sustainable, but both coverage and adequacy would be somewhat incomplete. • A universal benefit of RM350 per month for all persons age 65 and above would cost about 4 percent of projected fiscal revenues in 2020 (or 0.6 percent of GDP). Costs would grow substantially thereafter, reaching 11.8 percent of projected fiscal revenue in 2050. This appears unaffordable. 29 According to EPF administrative data, the median wage for covered workers at end 2018 was RM2,354 per month. A Silver Lining: Productive and Inclusive Aging for Malaysia 91 Chapter 3: Policy Areas • A social pension covering 40 percent of all persons age 65 and above and providing a benefit of 30 percent of the median wage (MYR769 per month) would lead to initial costs of 3.5 percent of fiscal revenues in 2020 (or 0.5 percent of GDP), rising to 10.4 percent in 2050. This option also appears unaffordable at least in the short term, especially given the particularly constrained fiscal environment caused by the COVID-19 crisis. Thus, a universal social pension or one that provides a significantly higher benefit than BOT might be more of a medium-term ambition. Before introducing a social pension, it would also be helpful to review the experience of some of Malaysia’s regional comparators. Various countries across East Asia and beyond have introduced social pensions, but there are significant cross-country differences in the objectives, implementation arrangements and adequacy—for instance as a percentage of the average per capita income of the B20—of these pensions (see Figure 3.40). For instance, Mongolia pays dedicated non-contributory social assistance benefits to less than 1 percent of older persons because the country’s contributory pensions scheme is generous and almost universal. The Philippines uses its social registry and proxy means-testing capability to target dedicated social assistance to the poorest and most vulnerable older persons. The benefit level is low but still meaningful for the target population. New Zealand has a social pension scheme which is universal and provides substantial benefit levels at considerable cost. However, it is worth noting that New Zealand does not have a mandatory contributory retirement savings scheme. Instead, it uses a framework of voluntary saving schemes to supplement tax-financed social assistance, in particular the default opt-in KiwiSaver scheme which covers around three quarters of the working-age population. In contrast, Australia requires employers to pay pension contributions and has a means-tested basic pension which covers around 60 percent of the older population, with around 40 percentage points of those receiving the full flat benefit and the benefit tapering as people’s tested income approaches the threshold. FIGURE 3.39: Simulated fiscal cost of social FIGURE 3.40: Adequacy of social pensions for pension, Percentage of revenue B20, Percentage of average income 15 RM769/Month South Africa Brazil RM350/Month Cost of social pension (% of revenue) Mauritius Costa Rica Paraguay 10 Lithuania Panama Chile Mexico Romania Latvia 5 Bulgaria Cape Verde Rwanda Poland Turkey Colombia Sri Lanka 0 Nepal 2020 2025 2030 2035 2040 2045 2050 Bangladesh Tajikistan Slovak Republic Current RM350/month - Bottom 40% - Age 60+ Current RM350/month - Bottom 40% - Age 65+ 0 20 40 60 80 100 120 140 Current RM350/month - Universal - Age 65+ 30% of median (RM769/month) - Bottom 40% - Age 65+ Adequacy of social pension for B20 (% of average income) Source: Authors’ calculations based on Household Income and Basic Source: Authors’ calculations based on IMF, EPF, and UN WPP. Amenities Survey 2016 (DOSM) and World bank ASPIRE database. 92 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas 3.3 Aged Care Demand, Provision and Financing Rapid aging accompanied by a changing socioeconomic context poses a significant challenge to existing aged care arrangements in Malaysia. Like in many other countries in East Asia, familial provision of aged care—both social care and health care—has long played a critical role in Malaysia whereby frail older persons get support and care from their spouse, children, and relatives. However, this traditional arrangement is weakening as the typical household structure changes from large and multi-generation families toward smaller nuclear ones. The average household size in Malaysia has declined from 5.2 in 1980 to 4.1 in 2016 (see KRI 2018). In parallel, the proportion of households with only older persons has increased from 16.5 percent in 2004 to 19 percent in 2016 (see Section 2.3). While co-residence remains common, the share of three-generation households declined from 41.1 percent in 2004 to 30.7 percent in 2016. The decline in co-residence would affect households’ ability to informally provide care to older household members. In contrast, the formal provision of aged care is still at an early stage of development. While home and community-based aged care services tend to have much lower unit costs, are generally preferred by almost all older persons and their families, and are typically able to provide adequate care to almost all older persons other than those with the most intensive care needs, public provision is weighted toward an institutional care model which provides a very small proportion of the most destitute with basic care services. Coverage of private institutional care is low and uneven with only relatively few exploratory attempts to establish viable, promising business models. Home and community-based aged care are at a small scale and confined mainly to urban areas. Care needs will rise steeply due to rapid aging and increasing rates of individual functional limitations and dependencies among older persons. As detailed in Section 2.1, aging will accelerate in the coming decades, which has implications for the demand for aged care. First, the number of older persons will continuously rise from its base of 2.4 million in 2020. This rise will speed up over time, with a projected annual net increase of 100,000 older persons in 2030 and 210,000 in 2040 (see Figure 3.41). Second, the proportion of older persons age 75 and above among all older persons will significantly increase. Third, the prevalence of individual functional limitations and dependencies will increase as people get older, especially the oldest old who have a much higher prevalence of frailty and dementia. Altogether this will present a growing challenge to the social and health sectors in Malaysia. The prevalence of individual functional limitations and dependencies is rising in part because the prevalence of NCDs is higher among older than among younger persons. ADLs correspond to essential elements of self-care such as feeding, toileting, mobility, dressing, and cleaning. Hence ADLs functional limitations imply the need for substantive care services. IADLs are instrumental to enabling independent living—such as food preparation, shopping, housekeeping, and managing finances—which may be addressed by lower intensity home and community-based services. The number of persons with ADLs functional limitations is projected to increase from 770,000 in 2020 to 1,160,000 in 2030 and to 1,630,000 in 2040. In parallel, the number of persons with IADLs dependencies is projected to increase by about one million per decade (see Figure 3.42). Old persons tend to have a high prevalence of NCDs—a major cause of preventable disability worldwide (Richards, et.al 2016), underscoring the criticality of health promotion and health management for healthy aging. In Malaysia, 41.5 percent of those age 60 and above had diabetes in 2019, according to the National Health and Morbidity Survey 2019. A Silver Lining: Productive and Inclusive Aging for Malaysia 93 Chapter 3: Policy Areas FIGURE 3.41: Projected net annual increase FIGURE 3.42: Projected annual net new persons in older persons by age group, Number and with ADLs limitations and IADLs dependencies, Percentage Number 14,000 16 140,000 Net new persons with functional limitations to ADLs 14 Older persons (% of total population) 12,000 Net increase in older persons (’000) 120,000 12 10,000 100,000 and dependency in IADLs 10 8,000 8 80,000 6,000 6 60,000 4,000 4 2,000 40,000 2 0 0 20,000 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 0 2020 2025 2030 2035 2040 2045 2050 Age 55-64 Age 65-74 Age 75-84 Age 85+ % Age 55-64 % Age 65-74 % Age 75-84 % Age 85+ Functional limitation to ADLs Dependency in IADLs Source: Authors’ calculations based on UN WPP and National Health and Source: Authors’ calculations based on UN WPP and National Health and Morbidity Survey 2018. Morbidity Survey 2018. The rise in functional limitations and dependencies is occurring against the backdrop of already significant unmet care needs. According to the National Health and Morbidity Survey 2018, 30.8 percent of those age 60 and above are malnourished. On top of that, the same survey also found that 30.8 percent of those in the same age group have poor social support. An earlier study using the 2003-2005 National Survey on Mental Health and Quality of Life of Older Malaysians painted a similar picture, showing that 18 percent of functionally disabled older persons suffered from unmet needs (see Momtaz et al. 2012). Moreover, the prevalence of dementia, a critical disability due to the intensity of care required, is estimated at 8.5 percent of those age 60 and over (or approximately 270,000 persons), according to the National Health and Morbidity Survey 2018. An aged care system includes a set of building blocks that together enable the delivery of care services. Typical building blocks of an inclusive aged care system include needs assessment, aged care planning, financing, human resources, quality standards and assurance, an information system, and monitoring and evaluation (see Figure 3.43). Each building block is an interlinked sub-system that involves specific requirements as well as program design, financing and management aspects. Since the early 1990s, there has been progress in establishing an inclusive aged care system. The Government has formulated a range of policies, plans and laws for strengthening aged care (see Appendix D for details). With regard to legislation, in 1993 the Care Centers Act (Act 506) placed the regulation of private aged care centers (and other care centers, equivalent to care homes) under JKM while in 1998 the Private Healthcare Facilities and Services Act (Act 586) placed the regulation of private nursing homes under MOH. The Private Aged Healthcare Facilities and Services Act (Act 802) was passed in 2018 to address the perception that regulatory stringency required by the Care Centers was too low and that the Private 94 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.43: Building blocks of an inclusive aged TABLE 3.2: Types of aged care provision in care system Malaysia Private Monitoring Needs Type of care Public providers and NGO and evaluation assessment providers Institutional Care homes (RSK Care homes Information Aged care care and RWT) (RS) system Older planning Nursing homes Nursing persons- (RE) homes centered approach Retirement villages Standards Financing and quality (capital, Home and Activity centers Daycare assurance services) community- (PAWE) centers based care Home help Home and Human resources and transport community- services (UPWE) based care Source: Authors. Source: Authors. Healthcare Facilities and Services Act requirements were too high. After a five-year transition period, this Act places the responsibility of regulating private aged care facilities under MOH. In parallel to the legislative agenda, the Government has provided institutional care through care homes (Rumah Seri Kenangan or RSK and Rumah Warga Tua or RWT) and nursing homes (Rumah Ehsan or RE). It has also introduced home help services, activity centers (Pusat Aktiviti Warga Emas or PAWE), and transport services (Unit Penyayang Warga Emas or UPWE), as well as promoted health awareness and provided health care services for older persons. These initiatives are important for developing an aged care system. However, the system is still fragmented, because both the Federal Government and state Governments share responsibilities in financing and providing care for older persons while District and Local Authorities have responsibilities for policy coordination and implementation. Policies or regulations do not clearly define the respective roles and responsibilities of different levels of Government. Malaysia’s aged care system caters to beneficiaries with a range of care needs and limitations and includes a range of service delivery models (see Table 3.2): • Publicly owned, financed and operated. This model covers care services targeted at the destitute directly operated by public operators such as RSK for those without ADLs functional limitations, and at RE for those with such limitations or who are chronically ill. In addition, short-term care of up to seven days to allow caregivers respite is provided on-site at one RSK. Public health care services for older persons are provided directly by MOH. • Purely privately owned and financed and operated for profit. This model includes a wide range of for-profit private aged care facilities providing regulated and unregulated services. Private daycare centers, care homes, and nursing homes are regulated, and operators require a license from the A Silver Lining: Productive and Inclusive Aging for Malaysia 95 Chapter 3: Policy Areas relevant authorities (though there is also a large number of unlicensed providers, see Table 3.3 below). In contrast, some aged care services are unregulated including home live-in or daily care, mobile care, community care, and retirement villages. • Private not-for-profit and mixed models. In this model, private provision is financed through public grants, private donations, or a mixture of the two, and delivered by nongovernmental organizations (NGOs). Examples of public financing of not-for-profit providers include care homes for older persons (Rumah Sejahtera or RS), PAWE, UPWE, and home help services which are run by NGOs with the help of volunteers (who receive a modest allowance). There are also some NGOs including foundations and charitable and religious organizations that rely purely on donations to run daycare centers, care homes and nursing homes. Institutional aged care covers only a very small portion of older persons. An estimated 731,000 older persons have ADLs functional limitations. In contrast, there are about 2,745 residents in RSK, RE and RS—or less than 0.4 percent of the estimated older persons with ADLs functional limitations. Private residential care homes have more residents than public ones, but still cover less than 5 percent of the potential demand. While precise data on the number of beneficiaries of private residential care homes are unavailable, estimates by informed stakeholders suggest that these homes provide for about 30,000 older persons. This is more than 10 times the total residential care provided by public providers but still covers just 5 percent of estimated older persons with ADLs functional limitations (see Table 3.3). TABLE 3.3: Number of aged care facilities and beneficiaries Care facility Care type Number Beneficiaries RSK Institutional 17 facilities 1,681 RE Institutional 2 facilities 207 RS Institutional 63 facilities 857 Private residential Institutional 320 licensed and ≈ 30,000 care homes ≈ 1,400 unlicensed homes Private nursing homes 30 Institutional 24 licensed homes N/A PAWE Home/community-based 88 centers N/A UPWE Home/community-based 9 vehicles 5,512 Home help services Home/community-based 2,150 volunteers 5,892 Private daycare centers Home/community-based 6 licensed centers N/A Source: Authors’ calculations based on JKM and estimates by informed stakeholders. Most recent available data. 30 Private residential care homes include for-profit and not-for-profit providers, while private nursing homes include hospices but excludes psychiatric centers. 96 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.44: RSK residents, Number FIGURE 3.45: Licensed private residential aged care centers by state, Number and Number per population age 65 and above 2,000 70 30 60 25 Licensed aged care centers Licensed aged care centers 50 per 100,000 age 65+ 1,500 20 40 15 RSK residents 30 1,000 10 20 10 5 0 0 500 Melaka Negeri Sembilan Perak Selangor Johor Pahang Kedah Kuala Lumpur Perlis Penang Sarawak Sabah Terengganu Kelantan Labuan Putrajaya 0 2016 2017 2018 Licensed aged care centers West Malaysia Sabah and Sarawak Licensed aged care centers per 100,000 age 65+ Source: JKM. Source: Authors’ calculations based on JKM. The utilization of public residential aged care facilities is low and falling. The capacity of public residential aged care facilities in Malaysia (excluding Sabah and Sarawak) is approximately 5,000 according to correspondence with MWFCD, but the actual number of residents is less than half of that, indicating public resources are not fully utilized. Potentially, this may be because of strict eligibility criteria or because only basic services are provided.31 In turn, it raises questions on how to improve the efficiency, quality and effectiveness of care provision for destitute older persons. These questions become even more pressing given that from 2016 to 2018, the number of RSK residents declined from 1,915 to 1,681 (see Figure 3.44). In parallel, their share among the population of older persons declined as well. While this decline might at least partly be due to a declining rate of destitution amongst older persons (a key eligibility criteria for placement in an RSK), there might also be evolving perceptions regarding quality of RSK relative to private options. There has been an expansion of formal home and community-based aged care services, but these services still cover less than 1 percent of older persons with ADLs functional limitations. Home help services mainly provide home visits and food services and cover less than 6,000 older persons, largely in urban areas. Similarly, in 2018 transportation services helped 5,512 older persons access medical providers. In addition, there is a limited number of private for-profit or not-for-profit daycare centers that provide daycare services. The number of PAWE has also expanded substantially from 59 in 2018 to 88 in 2019, with expansion to a total of 133 centers planned for 2020. PAWE largely implement activities like physical exercises, recreation, health screenings, and education targeted toward older persons without ADLs functional limitations. While these activities are helpful to promoting productive and inclusive aging, in their current form PAWE are not daycare centers that provide care services and support to older persons with 31 A site visit to an RSK showed that services provided are in line with basic requirements, including personal care and protection, counselling, recreation, medical treatment, occupational therapy, and physiotherapy. However, the site visit also showed that only basic services were provided, leaving room to improve the scope, level and quality of aged care services at the public care homes and nursing homes. A Silver Lining: Productive and Inclusive Aging for Malaysia 97 Chapter 3: Policy Areas ADLs functional limitations. According to the qualitative fieldwork conducted for this report, many older persons without significant ADLs functional limitations see home and community-based aged care services as more affordable and preferred alternatives to institutionalized care (see Interview quote 3.1). INTERVIEW QUOTE 3.1 “It is pathetic if I have to live in old folks’ home. Because I do not know people there. I don’t like it! I visited the old folk’s home before—the environment is pitiful, and the cost is high as well.” Source: 80-year old female, care recipient Geographic disparities present another challenge to aged care provision in Malaysia. Today, there are significant imbalances in the supply of aged care services both between states and between urban and rural areas. Figure 3.45 depicts the number of licensed private residential aged care centers by state and their distribution relative to the population of older persons across states. Significant disparities emerge. For example, Melaka has 26 private residential aged care centers per 100,000 older persons, while Penang, Sarawak, Sabah, Terengganu, Kelantan, and the federal territories of Labuan and Putrajaya all have less than five. For care provision, differences in the local context—in addition to differences in norms and values between different ethnicities or other population groups that might also influence aged care demand in Malaysia—can have important implications for aged care planning and provision. It would be worthwhile to carefully examine these dimensions with more local data and information from localities, if available. As Malaysia’s population ages, aged care provision will need to reach the large and growing “missing middle.” The small coverage of both public and private aged care services suggests that most aged care needs are unmet or met informally through home care provided either by hired domestic helpers or family members, often with economic and personal consequences such as less female labor market participation (see Interview quote 3.2, Interview quote 3.3, and Interview quote 3.4). Today, destitute older persons are targeted by public care homes and nursing homes, and well-off households can buy aged care services from the private sector. In contrast, most B40 households struggle to receive aged care services. On the one hand, they cannot benefit from public provision of services limited to the destitute. On the other hand, they are unable to afford private sector prices (see below). As a result, many low-income households rely on informal care provision by family members. Indeed, among those age 60 and above, 7.5 percent report experiencing neglect in the past 12 months, according to the National Health and Morbidity Survey 2018. Public financing for aged care services also poses challenges—partly because it is very limited. Although the precise amount of the Federal Government’s expenditure on aged care services is unknown due to the co-mingling of some expenditures, total public spending is unlikely to have been much higher than RM61.4 million in 2020 (see Figure 3.46), excluding institutional care in Sabah and Sarawak. Nevertheless, this INTERVIEW QUOTE 3.2 “I need to resign from my work and look for another job which has suitable time so that I can take care of my mother. I can’t have a permanent job, as I might need to take care of my mother anytime. I feel worried in terms of financial aspects, however I get to cover the expenses with my father’s retirement funds, and my brother also helps me with that.” Source: 42-year old female, informal caregiver 98 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas INTERVIEW QUOTE 3.3 “I go out less now, as I need to take care of [my father]. We have to make a change, as this is our commitment, it is an obligation in terms of mental, physical, and financial.” Source: 40-year old male, informal caregiver INTERVIEW QUOTE 3.4 “It is not about my children don’t want to take care of me in the future, however they already get married, and have their own family. They need to take care of kids and go out for work. They have a lot of things to worry about. If we really need them to take care of us, it will become another burden for them.” Source: 69-year old female, independent older person was an increase from RM46.8 million in 2015. In addition to financing from the Federal Government, State Governments also finance some aged care services. For example, Perak State allocates RM2 million toward social well-being initiatives for women, families, and older persons. Given the limited federal financing of aged care services and even more limited financing by State Governments, total public spending on aged care programs is estimated to account for only 0.01 percent of Malaysia’s GDP while private spending on aged care is estimated to be 0.06 percent of GDP. This means that Malaysia spends a much lower share of its GDP on aged care services than OECD countries (see Figure 3.47). FIGURE 3.46: Public spending on aged care by FIGURE 3.47: Spending on aged care by country, year, RM (million) 32 Percentage of GDP 70 4.0 3.5 60 Public spending on aged care (RM million) Spending on aged care (% of GDP) 3.0 50 2.5 40 2.0 1.5 30 1.0 20 0.5 10 0 Netherlands Norway France Japan OECD17 Germany Israel Spain Republic of Korea Singapore Malaysia United Kingdom 0 2015 2016 2017 2018 2019 2020 (b) RE RSK Home Help PAWE UPWE Source: Authors’ calculations based on JKM and MOF. Source: Authors’ calculations based on JKM, MOF, and OECD. Most recent Note: Data on spending on RSK in Sabah and Sarawak unavailable. available year. Note: Data on Spending on RSK in Sabah and Sarawak unavailable. 32 Estimating total public financing for aged care is challenging because expenditures for directly provided services such as for RSK and RE are co-mingled under multiple categories (such as salaries and services and supplies) and cannot be easily separated out from the budget. For operational expenditures, these specific-object discrete- object (SODO) categories include emoluments, services and supplies, assets, transfers and fixed payments, and other expenditures. For development expenditures, the relevant SODO categories are services and supplies, assets, and transfers and fixed payments. Partly as a result of this co-mingling, the usability of public financing data for monitoring and evaluation purposes and the establishment of linkages between spending and outcomes is challenging. A Silver Lining: Productive and Inclusive Aging for Malaysia 99 Chapter 3: Policy Areas Another challenge of public financing of aged care services is that this remains heavily skewed toward institutional care rather than home or community-based care (see Figure 3.46). More than four-fifths (82 percent) of public financing for aged care services is allocated to RSK and RE which serve less than 2,000 older persons throughout Malaysia, although this is a welcome decrease as RSK and RE received almost nine-tenths of public financing in 2015. However, as spending data for RSK in Sabah and Sarawak have not been available for purposes of this report, the proportion of overall public spending allocated to institutional care is likely to be even higher. The moderation in the skewness of public financing toward institutional care from 2015 to 2020 was driven by an increase in spending on home and community-based care services (such as for PAWE, home help services, and UPWE), which rose from RM5.3 million in 2015 to a budgeted RM11.2 million in 2020. Also of note is that the home and community-based care services are commissioned from NGOs rather than directly provided by the government. Compared with private sector prices, unit operational costs for RSK and RE appear relatively high. Using a set of simple assumptions, the implied monthly operational costs per resident of public residential aged care institutions with dormitory-style accommodation in 2018 is estimated at approximately RM3,760 and RM2,232 for RE and RSK respectively. By way of comparison, prices for private aged care homes range from RM1,000 to RM2,000 in Kedah (for a twin room) to RM2,000 to RM3,000 in Putrajaya (for a triple shared room) per month according to a list of care centers compiled by the Association of Residential Aged Care Operators of Malaysia (AgeCope).33 Quality Assurance and Human Resources Malaysia’s aged care quality standards focus on infrastructure but do not set appropriate service standards. The setting of appropriate quality standards encompassing both infrastructure and services is important to ensure the quality of aged care services and value for money. Malaysia’s relevant laws such as the Care Centers Act, the Private Health Facilities and Services Act, and the Private Aged Healthcare Facilities and Services Act have specific legal schedules on the processes for licensing, approval, and enforcement for residential and daycare facilities but do not set quality standards themselves. With reference to the relevant laws, infrastructure standards have been introduced for care homes, nursing homes and daycare centers. However, service standards are not available which makes it challenging to guide, monitor and ensure the quality of aged care services. COVID-19 infections in several aged care homes in Malaysia demonstrate the importance of service standards, in particular health and safety standards as part of a broader response to protect older persons. In response to COVID-19, Malaysia has implemented a range of proactive measures to improve infection prevention and control in aged care homes to protect frail older persons: MOH adopted an interim Recommendation for the Prevention of COVID-19 Transmission while JKM provided cash grants to private aged care homes. Furthermore, MOH announced an initiative to test all care home staff and residents— including in unregistered care homes—as part of measures intended to secure the safety of Malaysia’s most vulnerable population and promote social and economic recovery. While these measures present a breakthrough in treating licensed and unlicensed aged care facilities equally, more public resources are needed (i) to make personal protective equipment available to caregivers, (ii) to strengthen coordination between the social sector and the health sector, as well as between residential facilities, health agencies and hospitals, and (iii) to build capacity for the medium-term management of health and safety in aged care homes (see Box 3.4). 33 Prices as compiled by AgeCope typically include food, lodging, care, and some consumables but not adult diapers or medication. 100 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas BOX 3.4 COVID-19 and the protection of older persons The COVID-19 pandemic has revealed numerous weak points of the aged care systems throughout the world. While hitting all population segments worldwide, the pandemic threat is particularly high in the case of older persons, and responses have shown disproportionately lesser impact, resulting in dramatic consequences in terms of loss of lives, opportunities and inclusion (see Rutkowski 2020). The residency arrangements and care needs of vulnerable populations present particular challenges in the face of the pandemic. Dependent people staying in care homes generally live in close and sometimes crowded quarters, facilitating the spread of COVID-19. That is why many countries including Malaysia have enacted various measures to control infections and protect older persons dependent on long-term care. These measures span from prevention and control measures, to management of human and financial resources, to reporting and coordination with public health authorities. They include policies such as daily symptom screening, social distancing requirements, visitor restrictions, detection and quarantine procedures, enhanced disinfection regimes, staff training and protection, reporting procedures, clear leadership structures, awareness campaigns, and the adoption of other national or local guidelines and preparedness plans (see Wang et al. 2020). Health and safety measures need to be further strengthened in Malaysia, especially those for personal protection equipment and coordination between social and health sectors. As of May 25, 2020, 83.8 percent of 10,890 staff and residents from 267 aged care homes had been tested for COVID-19 with 0.2 percent of tests coming back positive. Many care homes and nursing homes are reportedly still short of personal protective equipment for staff and caregivers (see Hasmuk et. al. 2020). Therefore, more public resources are needed to make personal protective equipment available to caregivers, to strengthen coordination between the social sector and the health sector and between residential facilities, health agencies and hospitals, and to build capacity for the medium-term management of health and safety in aged care homes. Testing also needs to be further scaled up to cover staff and residents in all aged care facilities. Measures for managing the workload of and guaranteeing sick leave for caregivers should also be continued to safeguard persons in care homes. In parallel, the coordination between care homes and hospitals should be better managed to facilitate a safe transition of patients to ensure the continuum of care services. In this context, modern information technology can be used as an effective tool to promote care integration, as well as to address emergency and psychological counseling care needs (see Rutkowski 2020). For instance, telehealth systems can be used to monitor vital signs and provide emergency and medical services while avoiding physical contact. Similarly, using teleconferencing and other communication systems can connect older persons to psychological counseling services and lessen isolation. A Silver Lining: Productive and Inclusive Aging for Malaysia 101 Chapter 3: Policy Areas BOX 3.4 (continued) The COVID-19 pandemic offers a wake-up call for countries including Malaysia to rethink how to reform, regulate and govern aged care systems to ensure older persons live in a healthy and safe environment. The COVID-19 pandemic has exacerbated some of the challenges of an aged care system overly reliant on institutional care. Accordingly, it has heightened the importance for Malaysia to introduce and expand formal home- and community- based aged care services. In addition, it has also highlighted the importance of strengthening the Government’s stewardship capacity in areas such as needs assessment, quality standards, financing, human resources, regulation and governance, monitoring and evaluation. This will enable the Government to better meet the increasing needs of and provide more protection to older persons. The quantity and quality of human resources are another crucial constraint for Malaysia as it develops and expands aged care services (see Interview quote 3.5). Following the definition of the International Labour Organization (2018), care workers are those who, for profit or pay, provide face-to-face services that develop the human capabilities of care recipients in a relatively dependent position, such as older persons. In the health and social work sectors, care workers encompass (i) personal care workers such as health care assistants and home-based personal care workers, (ii) health professionals such as doctors, (iii) health associate professionals such as medical technicians, and (iv) domestic workers who mind children, care for older persons or persons with disabilities, cook, clean, etc. Based on this classification, estimates for the size and composition of different types of aged care workers in Malaysia can be computed through a two-step procedure: First, care workers can be identified using the International Standard Classification of Occupations and the International Standard Industrial Classification at the two-digit level mapped to the Malaysia Standard Classification of Occupations (MASCO) and the Malaysia Standard Industrial Classification (MSIC), respectively. Second, a range of occupations and sectors not related to aged care (such as the education sector, as well as midwifery professionals and childcare professionals) can be excluded.34 INTERVIEW QUOTE 3.5 “Maybe people in authority are not aware of the increasing aging population. Or they don’t have the funds to start training centers for caregivers. It is very hard to find and hire a local Malaysian as a caregiver. Even when we post a vacancy for caregiver positions, the response is poor.” Source: 46-year old male, nursing home operator There were 388,000 aged care workers in 2017—a slight decline as compared to 2011. During the same period, the share of aged care workers among total employment also declined, from 2.6 percent to 2.1 percent. Figure 3.48 and Figure 3.49 show that the decline in the overall number of aged care workers was primarily due to a decline in the number of domestic workers. This number dropped from 226,000 in 2011 to 106,000 in 2017. In contrast, the number of health workers (encompassing both health professionals and health associate professionals) increased from 180,000 to 256,000 while that of personal care workers was 26,000 in both 2011 and 2017. 34 Estimates for the size and composition of different types of aged care workers in Malaysia derived from the two-step procedure are by their very nature an approximation. For instance, the available data do not make it possible to differentiate between those domestic workers who are engaged in aged care and those who are not; nor can it be used to differentiate between various health professionals who are more likely (e.g. gerontologists) or less likely (e.g. pediatricians) to be aged care workers. Hence, information regarding the number of aged care workers in this report likely represents upper-bound estimates. 102 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.48: Aged care workers by occupation, FIGURE 3.49: Aged care workers by occupation, Number (’000) Percentage of total employment 3 400 Aged care workers (% of employment) Aged care workers (’000) 300 2 200 1 100 0 0 2011 2012 2013 2014 2015 2016 2017 2011 2012 2013 2014 2015 2016 2017 Domestic workers Health associate professionals Domestic workers Health associate professionals Health professionals Personal care workers Health professionals Personal care workers Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Nearly 40 percent of personal care workers and domestic workers engaged in aged care are at least 40 years old and above, and the large majority of aged care workers are women. Figure 3.50 shows that about 67.4 percent of aged care workers are between age 20 and 39. Accordingly, 32.6 percent of aged care workers are age 40 and above. Among personal care workers and domestic workers, nearly 40 percent are at least 40 years old. Also, the large majority of aged care workers are women. According to Figure 3.51, this is the case for 55.8 percent, 77.4 percent, and 93.6 percent of personal care workers, health workers, and domestic workers, respectively. FIGURE 3.50: Aged care workers by occupation FIGURE 3.51: Aged care workers by occupation and age group, Percentage and gender, Percentage 100 100 80 80 Aged care workers (%) Aged care workers (%) 60 60 40 40 20 20 0 0 Personal care Health Health Domestic Personal care Health Health Domestic workers professionals associate workers workers professionals associate workers professionals professionals 15-19 20-29 30-39 40-49 50-59 60-64 Male Female Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). A Silver Lining: Productive and Inclusive Aging for Malaysia 103 Chapter 3: Policy Areas Many aged care workers and family members of older persons providing aged care services might lack the required skills and support. Figure 3.52 shows that 83.2 percent of personal care workers have at most secondary education, and 74.8 percent of domestic workers have at most primary education—raising concerns regarding whether these workers possess the necessary skills to provide quality aged care services. In addition, semi-structured interviews conducted for this report suggest that even trained nurses need support when they transition to become aged care workers (see Interview quote 3.6). The interviews also suggest that there is a lack of training available to enable family members caring for older persons to provide the appropriate care (see Interview quote 3.7). After stepping into their roles, some family members quickly realize that aged care is multi-faceted and requires skills in areas such as physiotherapy, diet and nutrition, and emotion management. The semi-structured interviews also suggest that some aged care workers experience high turnover, mental pressure, and limited opportunities for training and skill development (see Interview quote 3.8 and Interview quote 3.9). INTERVIEW QUOTE 3.6 “I thought the transition would be a natural one. I was wrong. In private nursing or caregiving, we have to rotate the position of the (bedridden) patient to prevent “bed sores”. And all this is done alone, not with the help of a team. If a caregiver is not trained in working solo, they will resent the patient in the long run and the aged care recipient will suffer. You need a certain unique kind of person with a unique set of skills to provide aged care at homes. Personal care-giving is certainly different from traditional nursing.” Source: 28-year old female, wage caregiver INTERVIEW QUOTE 3.7 “After coming back from hospital, I don’t know what (food) to prepare for him, as he has a high index of diabetes. I think they can provide a program which teaches the caregiver on how to prepare the meals, which is a very good program. If the hospital is able to educate us before discharge from the hospital, then, everything will become easier after going back home.” Source: 53-year old female, informal caregiver INTERVIEW QUOTE 3.8 “...not all caregiver can last long being a caregiver. For example, people resign after 2-3 months, because they can’t take care of the emotion and all that. Not everyone is tough to be a caregiver.” Source: 30-year old male, wage caregiver INTERVIEW QUOTE 3.9 “In terms of ways of taking care, if the welfare unit can organize the workshop and invite the relevant speaker to train on how to take care of elderly people or sickly people.” 67-year old male, informal caregiver 104 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas In principle, more demand-driven labor migration policies can help loosen the shortage of aged care workers and also foster broader policy goals. According to Figure 3.53, most personal care workers and health workers engaged in aged care are Malaysian citizens. In contrast, 93.6 percent of domestic workers are non-citizens.35 This means that the decline in the number of domestic workers between 2011 and 2018 might be due to more restrictive immigration policies, rather than a lack of demand for aged care services (see KRI 2018). Conversely, more demand-driven immigration policies could at least in principle help loosen the shortage of aged care workers. In addition, this could also foster other policy goals such as increasing economic opportunities for women. World Bank (2019) finds that one of the main reasons that Malaysian women do not participate in the labor market is due to the need to perform household chores and caregiving while Tan and Gibson (2013) and KRI (2018) argue that the presence of a domestic worker in a household is positively correlated with female labor force participation. Thus, a larger supply of foreign domestic workers might enable more Malaysian women to participate in the labor market. FIGURE 3.52: Aged care workers by occupation FIGURE 3.53: Aged care workers by occupation and education, Percentage and citizenship, Percentage 100 100 80 80 Aged care workers (%) Aged care workers (%) 60 60 40 40 20 20 0 0 Personal care Health Health Domestic Personal care Health Health Domestic workers professionals associate workers workers professionals associate workers professionals professionals No certi cate Primary Secondary Post-secondary Tertiary Malaysian citizen Non-Malaysian citizen Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Source: Authors’ calculations based on Labour Force Survey 2017 (DOSM). Beyond domestic workers, labor migration of professional care workers can potentially contribute to improving the supply of qualified aged care workers. Even though the semi-structured interviews suggest that Malaysia faces a shortage of qualified aged care workers, current labor migration policies do not yet cater for the hiring of these care workers outside of the category of domestic workers (see Interview quote 3.10). Policies are particularly restrictive in the case of male care workers and not only limit supply but also act as a disincentive for internationally mobile workers. Again according to the interviews conducted for this report, many of these workers feel that the job title “domestic worker” or “maid” does not appropriately describe their professional skills and tasks (see Interview quote 3.11). 35 Even this percentage should likely be regarded as a lower bound estimate for the share of domestic workers that are non-citizens, as foreign workers are likely to be underrepresented in data based on household surveys (see World Bank 2020). A Silver Lining: Productive and Inclusive Aging for Malaysia 105 Chapter 3: Policy Areas INTERVIEW QUOTE 3.10 “For some reason, the government doesn’t recognize caregivers at the moment. There is no visa sector given as a caregiver. Immigration only has sector visa for maid, domestic helper and also agriculture. There are only 3-4 levels that are given. How are we to bring in male caregivers from Philippines if they (are) categorized as maids? We need them, Filipinos are trained and very professional caregivers.” Source: 46-year old male, nursing home operator INTERVIEW QUOTE 3.11 “I had to attend an 8-month intensive course where professional medical personnel taught us how to prepare specific combinations of medicines and administer medicine, do IV-tube feeding, draw blood, administer CPR, perform proper wound care and prepare food based of unique diet and nutrition. We (were) also taught the art of physiotherapy and massage. And when we arrived here, we continue with our training periodically when it is offered at the Philippines embassy. Many local Malaysian nurses are not well-versed with palliative care. This is a unique and specialized skill that Filipino caregivers are trained in before given their certification. So, we are not maids. And we should not be given a maid visa but a visa indicating our skill set. Which is professional caregiver.” Source: 35-year old female, migrant wage caregiver Malaysia has some way to go in developing demand- and supply-side policies that would chart and incentivize clear career pathways for care workers. On the demand side, formal accredited training for caregivers at aged care homes is not required by the Care Centres Act while the relevant regulations under the Private Aged Healthcare Facilities and Services Act have not yet been published. Hence, there is little incentive for aged care workers to invest in accredited training. On the supply side, even though aged care is one of the skills areas in the MOHR’s National Occupational Skills Standards (NOSS), the prevalence of relevant qualifications among aged care workers is unclear but likely low. While there are NOSS for Elderly Care Centre Operation, Elderly Care Centre Administration, and Elderly Care Centre Management, only training in Elderly Care Centre Operation is currently offered and at only one accredited center, according to correspondence with the Department of Skills Development (Jabatan Pembangunan Kemahiran or JPK). Whereas the curriculum from NOSS indicates that the training covers care handing, activities for independent and dependent residents, reception of new residents, housekeeping, security, emergencies, and end-of- life issues, it is also unclear if all necessary technical skills are included (see Interview quote 3.12). Finally, there are no dedicated policies for career development of care workers, as well as insufficiency in data and statistical systems for human resources planning, development, and monitoring and evaluation. INTERVIEW QUOTE 3.12 In Malaysia, the training for nurses is not a comprehensive one. There is a lack of focus on caregiving and palliative care. Source: 28-year old female, wage caregiver 106 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas Governance of Private Sector Providers With the acceleration of aging, the private sector and NGOs have started to participate in the provision and financing of aged care services. In recent years, private service providers have been exploring various service delivery models such as home-based services, daycare centers, care and nursing homes, retirement villages, and even using online platforms for home-based care services. The increased private sector activity is a welcome sign for the development of Malaysia’s aged care sector. However, policy and institutional reforms are needed to make business models viable, scalable, and financially sustainable while at the same time assuring high-quality and inclusive provision of care. Various government agencies are involved in the regulation of private aged care service providers. JKM, MOH, local authorities, and the Fire and Rescue Department are all involved in regulating private (for- profit and not-for-profit) residential aged care facilities. The aged care governance framework is currently also in transition following the passing of the Private Aged Healthcare Facilities and Services Act in 2018. New regulations and guidelines which will cover the specific requirements and standards are currently under development. At least until this work is completed, regulations are oftentimes interpreted differently by different local governments and there is also a lack of clarity regarding the regulation of aged care homes run by religious associations. Strengthening policy coordination, improving the licensing process, and leveling the playing field across space and across service providers could lead to large gains in equity and efficiency. Critically, an estimated 1,400 residential aged care centers are unlicensed compared with 320 licensed centers. This is likely at least partly due to an often complicated, prolonged and costly licensing process and a varying complexity of licensing requirements across space (see Box 3.5). Specific difficulties frequently reported by operators include (i) the necessity of obtaining approval from neighbors, (ii) costs involved in complying with standards, particularly fire-related standards, (iii) costs involved in converting residential land titles to commercial land titles, and (iv) the need to obtain approval from multiple ministries and agencies, who may have differing requirements and standards. Difficulties in licensing limit the potential growth of the aged care sector and provide opportunities for rent-seeking. In addition, they encourage many aged care centers to remain unlicensed and therefore potentially weaken the protections of older persons. As already mentioned, this is a critically important issue during a crisis such as the COVID-19 pandemic (see Wang et al 2020). In addition to for-profit private providers, NGOs including civil society organizations, charities, foundations and associations play an important role in aged care policy formulation, provision and financing. There are many NGOs that provide aged care services, financed either through donations or commissioned by the government. For instance, JKM or MOH have earmarked funds for NGOs to run specific activities or programs. In addition, many NGOs have contributed to aged care policy formulation as members of the National Advisory and Consultative Council for Older Persons. NGOs have also been partners in the implementation of the National Policy on Older Persons. Overall, the strong role of NGOs is an important asset as Malaysia develops its aged care sector. When NGOs are commissioned to provide aged care services, payment and management arrangements tend to be rigid and to focus on inputs and processes instead of results. For example, the annual operational grant for a PAWE must be used in accordance with very specific guidelines issued by JKM (see JKM 2018). These guidelines cover areas such as the governance, management, and programs expected from the PAWE. The guidelines state that a registered society is the preferred entity responsible for managing the PAWE as it makes it possible to raise additional funds from other sources while imposing A Silver Lining: Productive and Inclusive Aging for Malaysia 107 Chapter 3: Policy Areas BOX 3.5 Licensing of private residential aged care centers Under the Care Centers Act, the typical process (which can vary by local authority) for applying for a license for a private residential aged care center consists of four steps (see Figure 3.54): • Step 1 – Choose a location: The applicant identifies a location and proposes a name for an aged care center. Local authorities and/or JKM may have varying requirements or guidelines regarding permissible locations. These might include the designation of spatial zones, specific rules on the density of aged care facilities within a particular area, or explicit preferences for the type of housing, such as landed residential lots in residential areas or the conversion of a residential title to commercial property (to be done during Step 3). There might be more flexibility if the applicant is a nonprofit entity. • Step 2 – Apply for JKM approval: The applicant submits an application to the district-level JKM office stating the location and proposed name(s) of the aged care center. The proposed name(s) have to be cleared with the Registrar of Societies or Registrar of Companies (depending on the type of applicant). Once the location and name have been cleared by JKM, JKM will then issue several letters to be used to seek approval from other agencies in Step 3. • Step 3 – Apply for approval from the local authority, the Fire and Rescue Department, and MOH: The applicant seeks approval in parallel from: (i) the local authority, which consults with neighbors to get their no objections (the applicant is given an opportunity to address concerns), decides if rules concerning the automatic convertibility of other types of care centers such as childcare or daycare centers apply, decides if a conversion of the land title from residential to commercial is required (which can require a substantial fee), and determines if technical aspects such as the width of bathroom doors and accessibility are in line with applicable standards; (ii) the Fire and Rescue Department to ensure compliance with the fire code; and (iii) MOH, for enforcement of relevant environmental and sanitary requirements. • Step 4 – Obtain license from JKM: Once all approvals by the local authority, the Fire and Rescue Department, and MOH have been obtained, the applicant can return to JKM to pay a small fee and obtain the license. The overall licensing process can take six months to two years and involve substantial expenditures, particularly to convert a land title from residential to commercial and to renovate the facility to comply with applicable standards. Requirements may also be interpreted differently by the different agencies responsible for regulating aged care centers, thus adding to uncertainty and costs. 108 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas FIGURE 3.54: Typical licensing process for a private residential aged care center Step 1: Choose a location Register name with Registrar Step 2: Apply for JKM approval of Societies or Companies Obtain Step 3a: Local Convert to Step 3b: Fire Renovations to Step 3c: Ministry of neighbor’s Authority approval and commercial Department comply with Health approval permission business licensing land title approval requirements Step 4: Obtain license from JKM Source: Authors. A Silver Lining: Productive and Inclusive Aging for Malaysia 109 Chapter 3: Policy Areas a more formal management. Guidelines also include details on the precise roles of each member of the management committee and the staffing of the PAWE by a part-time administrator and two part-time assistant administrators. Salaries are fixed at RM600 per month for the part-time administrator and RM400 per month for the assistant part-time administrators. An operational grant of RM33,330 per year is applied uniformly to all PAWE, regardless of actual costs or outputs delivered. One-off seed funding is also fixed at RM100,000 (previously RM80,000). Similarly, home help services commissioned from NGOs are paid on a cost reimbursement basis—fixed for each home visit—which results in a heavy administrative burden and focus on inputs instead of results. Detailed monthly reporting of activities is required and frequently done manually. There is room for improving the regulatory oversight of not-for-profit providers and the monitoring and evaluation system for quality improvement. Currently, there are some checks of the services commissioned by the Government, but there is no robust system to hold not-for-profit providers accountable for the quality of services delivered. The wide range of home and community-based services provided in varied settings creates additional challenges for quality monitoring. A more robust monitoring and evaluation system could facilitate regulatory oversight. Ideally, the system should also enable monitoring of the care needs of older persons and produce data that could be used for care planning and quality improvements. More broadly, Malaysia has no unified information system that collects and keeps data on both public and private providers of aged care services and there is only limited data sharing across ministries and agencies. In addition, aggregate information is available only for basic indicators (such as the number of beneficiaries and their demographic profile). Even this information is sometimes only partial, and data collection is frequently paper-based and manual. Monitoring and evaluation systems are also fragmented across programs and between the federal and state levels. Malaysia can learn from international experiences and practices on how to establish an inclusive aged care system to meet the increasing needs of older persons. Both OECD and some developing countries, such as Thailand and China, have introduced a mix of demand- and supply-side interventions and innovations in the area of aged care. This has included service expansion, regulatory measures and financial incentives to promote the development of home- and community-based care services as opposed to residential aged care. This shift at least partly reflects older persons’ changing preferences. In addition, home- and community-based aged care services are on average more cost-effective. Other innovative measures that Malaysia could study include training and wage incentives for human resources development, case management to promote integrated care and continuum of care between the social and health sectors, branding and franchising for service expansion to ensure quality of services, and financing innovations. Finally, in many countries the government has moved toward purchasing aged care services from private providers, outsourcing public aged care facilities to private operators, thereby fostering social and private sector participation. Many countries have also adopted a more results-based financing approach (see Box 3.6). Developing the aged care sector could become a new driver of economic growth in Malaysia. As discussed in Chapter 1, aging in general and aged care more specifically offer opportunities for both a booming silver economy and a more inclusive society. To realize these opportunities, policy actions and institutional reforms are needed to remove institutional barriers. In addition, the introduction of mechanisms that smartly use public financial resources as “seed money” will foster and mobilize social and private investment. This in turn can contribute to making the aged care sector an engine of economic growth and job creation. 110 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 3: Policy Areas BOX 3.6 Developing public-private partnerships (PPP) in the aged care sector A PPP is a long-term contract between a private party and a government entity, for providing a public asset or service, in which the private party bears significant risk and management responsibility, and remuneration is linked to performance. PPPs can be used to provide for both new and existing public assets or services and the private party can either be paid entirely by the users of the services, or a government entity can make some or all of the payments. A PPP usually covers cases where there is a public interest in the provision of the relevant assets or services. A comprehensive PPP framework typically has five components: First, a policy—an articulation of the rationale behind the Government’s intent to use PPPs to deliver public services, and the objectives, scope, and implementing principles of the PPP program. Second, a legal framework—the laws and regulations that underpin the PPP program, enable the Government to enter into PPPs, and set the rules and boundaries for how PPPs are implemented. Third, processes and institutional responsibilities—the steps by which PPP projects are identified, developed, appraised, implemented, and managed, and the roles of different entities in the process. Fourth, a public financial management approach—an approach on how fiscal commitments under PPPs are controlled, reported, and budgeted for, to ensure PPPs provide value for money, without placing undue burden on future generations, and to manage the associated fiscal risk. Fifth, other arrangements—covering how entities other than the government such as auditors, the legislature, and the public participate in the PPP program, and how they are able to hold those responsible for implementing PPPs accountable for their decisions and actions. Many countries use commissioning to purchase aged care services from the private sector or outsource the operations of public aged care facilities to best use available resources, improve efficiency, and maximize outcomes. This type of approach also relies on a partnership between the Government and the private sector, but of a different form than a classic PPP with a long-term contract. There are different models in commissioning aged care services. An approach that is based on strategic planning and outcome-based in the sense of linking resource allocation with meeting assessed needs is best practice. It is also best practice to use evidence to assess needs and make spending decisions, rather than to allocate funding on the basis of historical spending and inputs-based approaches. Regardless of contract type, transparency and a good governance structure have proven to be crucial for maximizing value for money. Source: Authors based on Center for Effective Services (2015) and World Bank (2017b). A Silver Lining: Productive and Inclusive Aging for Malaysia 111 CHAPTER 4 Conclusions and Recommendations 112 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations The year 2020 marks a crucial milestone in Malaysia’s history as it becomes an aging country. Henceforth, the country will age rapidly with the share of the population age 65 and above projected to double from 7 percent to 14 percent within only 24 years. This rapid aging process will have a profound impact on Malaysia’s development trajectory, raising various policy challenges and risks but also offering opportunities. Macroeconomic projections indicate that aging will account for about a third of the decline in Malaysia’s economic growth between 2020 and 2050. Equally important, there will be rising challenges with regard to providing productive and inclusive health, pension, and long-term care systems. Nevertheless, while impending demographic changes are critical and will become even more so in coming decades, demography is not destiny. Through the right mix of policies, the Government of Malaysia can help the country adapt to rapid aging and improve the well-being of all Malaysians. In fact, aging presents not only challenges but also opportunities. For instance, harnessing the productive potential of older workers can help address persistent talent shortages while the aged care sector can become a new driver for economic growth, create job opportunities, improve the quality of social services, and contribute to an improved quality of life. Both the challenges and opportunities created by aging are becoming even more crucial as Malaysia is likely to transition to a high-income economy within the next few years, and the aspirations of its expanding middle class regarding jobs, income, and well-being continue to grow (see World Bank forthcoming). An integrated, interagency policy approach can address challenges and constraints in a systematic and mutually reinforcing way. Reducing or ending some of the barriers to productive and inclusive aging identified in this report will require clear prioritization and finely calibrated policy approaches. For instance, enhanced opportunities for training and lifelong learning can help mitigate the impacts of aging on economic growth; specific parametric changes by EPF can improve the adequacy of EPF members’ retirement income; and selective increases in public financing for aged care services can help address unmet aged care needs. At the same time, many of the recommended policy approaches will address more than one barrier and different approaches will be mutually reinforcing. For instance, an increase in the minimum retirement age to 65 can not only improve employment opportunities for older workers, but also foster old- age income security. In addition to that, more productive and inclusive aging is necessarily a cross-sectoral endeavor; some of the most crucial policy approaches will require cooperation between different ministries and agencies. Maybe most strikingly, strengthening the governance of the aged care sector will necessitate cooperation between MOH, JKM, the Fire and Rescue Department and State and Local Governments, A Silver Lining: Productive and Inclusive Aging for Malaysia 113 Chapter 4: Conclusions and Recommendations among others. Therefore, a systematic and integrated interagency approach to the promotion of productive and inclusive aging—as well as one that adopts a life cycle perspective, mainstreams approaches to address the specific constraints faced by women, and is mindful of political economy considerations—will have the most pronounced and sustainable impact. Ideally, this approach would be guided by an integrated strategy and supervised by a steering committee. The promotion of productive and inclusive aging will require policies that foster the productive employment of all workers, provide minimum income protection to all older persons, and build an inclusive aged care system. General policy directions as well specific short-term, medium-term and long- term policy options are laid out in Table 4.1, with the caveat that this distinction is not always entirely clear- cut and that for some of the long-term policy options to be effective, implantation needs to begin in the short or medium term. With aging, Malaysians will have to work longer. This will need to be reflected in gradual adjustments to the minimum retirement age and EPF minimum withdrawal age, among other policy parameters. In parallel, policies are needed that foster workers’ productive employment—such as enhanced opportunities for training and lifelong learning. In addition, the provision of minimum income protection of older workers will require improvements in the coverage and adequacy of social insurance schemes, as well as the introduction of a modest, broadly targeted tax-financed social pension. Finally, for the aged care sector to become a new driver of economic growth, it will be crucially important to create an enabling market and regulatory environment for private aged care provision, to strengthen the governance of the sector, and to selectively increase public financing in line with available fiscal space. TABLE 4.1. Overview of policy recommendations General Policy Short-term Medium-term Long-term Directions Policy Options Policy Options Policy Options Gradually raise the min. Provide enhanced Develop regulations for retirement age to 65; link opportunities for training older workers’ productive 1 it to life expectancy at and lifelong learning. and flexible employment. Foster the productive retirement thereafter. employment of all Improve women’s Consider piloting workers, including Facilitate the adoption economic opportunities incentives to encourage older workers. of age-management through better childcare the employment of older strategies by companies. and other measures. workers. Adjust the minimum Develop an integrated Introduce a modest, withdrawal age for EPF vision of a system for old- broadly targeted tax- Account 1 and rethink the 2 age income protection. financed social pension. role of Account 2. Provide minimum income protection to Broaden EPF coverage to Explore annuitization Establish a one-stop shop all older persons. more of the self-employed options and longevity for EPF, SOCSO and HRDF and informal sector insurance for EPF contributions. workers. balances. Develop a systematic Upgrade skills of aged Streamline and harmonize and actionable aged care care workers through licensing requirements for strategy based on a solid strengthened training aged care facilities. 3 assessment. arrangements. Build an inclusive aged Strategically reorient Selectively increase public care system. Improve quality standards public financing toward financing to incentivize of aged care homes to home and community- market-oriented care ensure health and safety. based aged care. provision. 114 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations POLICY DIRECTION 1 Foster the productive employment of all workers, including older workers, through a comprehensive set of demand- and supply-side policies. Short-term policy options • Provide enhanced opportunities for training and lifelong learning across the life cycle, including to older workers. Upskilling and reskilling are important to ensure that all workers, including older workers, are able to obtain the skills required to continue working against the backdrop of rapid technological advancement and the COVID-19 pandemic, which has accelerated secular trends in changing skills demand (see World Bank 2020). According to the World Bank’s LTGM, improving the human capital of Malaysia’s workforce can also help mitigate the negative impact of aging on economic growth (see Box 4.1). Older workers who are provided with opportunities to upskill and reskill will be more likely to remain productive for longer, and will be able to transition to jobs that are less physically demanding and more flexible. While there are typically no age limits to existing training programs, World Bank (2017c) shows that employers registered with the Human Resource Development Fund (HRDF) are more likely to train younger workers than older workers. As such, it is worthwhile highlighting the importance of continued and improved access to training for older workers. In turn, this objective could for instance be fostered through dedicated a strategic initiative through HRDF. Any such strategic initiative would benefit from including those without standard employment contracts and taking into account their specific circumstances such as their generally relatively low level of education.36 • Encourage and facilitate the adoption of age-management strategies by companies, encompassing work organization, work equipment, working time policies, and training. Given that some older workers develop physical limitations, many countries have sought to adjust the physical work environment to enhance their productivity (see World Bank 2016). A holistic age-management strategy encompasses work organization, work equipment, working time policies, and training, among other aspects. While age-management strategies are largely the domain of companies, the Government can foster awareness and uptake by issuing guidelines on the types of strategies that can be undertaken, including examples of the types of technologies that can be utilized to maximize the productivity of older workers. Another way to raise awareness is to grant awards or prizes for the implementation of age-management strategies, similar to TalentCorp Malaysia’s Life at Work awards that recognize employers with progressive workplace strategies. Finally, the Government could consider limited financial incentives, such as grants to employers for workplace adjustments, including investments in the relevant technology. Financial incentives could be introduced for a limited period of time to encourage employers to pilot age-friendly workplaces, and could be directly tied to requirements related to the hiring or retaining a predetermined number of older workers for a specified period of time. Medium-term policy options • Develop a regulatory framework for the productive and protected employment of older workers and facilitate part-time and other flexible forms of employment. Japan and Singapore have introduced regulations that require employers to retain employees beyond the legal retirement age. While a mandate might not necessarily be required, Malaysia could consider developing a regulatory framework for the employment of older workers that would define conditions of employment, such as 36 World Bank (2017c) also demonstrates that, on average, HRDF-registered firms are more likely to offer training, and to train a higher share of workers than non-HRDF- registered firms. Therefore, older workers would likely benefit from policies that expand HRDF coverage to sectors where training rates are low and sector-based training bodies are insufficient, as recommended by World Bank (2017c). A Silver Lining: Productive and Inclusive Aging for Malaysia 115 Chapter 4: Conclusions and Recommendations access to social protection benefits and remuneration in a more comprehensive and systematic way than is currently the case. In addition, Malaysia could reinforce measures that facilitate flexible forms of work, including through the pursuit of relevant enabling provisions as part of the planned amendments of the Employment Act, as well as the Employment (Part-time Employees) Regulations. Statutory approaches to flexible work arrangements that could be considered include protecting staff with flexible work arrangements against potential discrimination (see World Bank 2019b). Part-time work is often cited as the most important flexible working arrangement and has been formally regulated in Malaysia since 2010.37 Moving forward, the definition of part-time workers could be broadened and for instance, allow for work from home. • Improve women’s economic opportunities through better availability, quality and affordability of child care, reforms of the legal environment and improved support for parents in line with international legal norms, and policies that address gender norms and attitudes. In addition to investments in human capital, the LTGM also shows that an increase in Malaysia’s female labor force participation rate could boost average GDP growth (see Box 4.1). In addition, not addressing the specific constraints to labor market participation that women face while implementing other policies to foster productive employment might exacerbate existing gender imbalances in Malaysia’s labor market. Against this backdrop, World Bank (2019b) provides a comprehensive analysis and policy recommendations toward better economic opportunities for women in Malaysia. Specific policy recommendations include better availability, quality and affordability of child care, reforms of the legal environment and improved support for parents in line with international legal norms, and policies that address gender norms and attitudes. In addition, it is also important that all other policies to foster productive employment take into account specific constraints of older female workers, such as relevant care burdens. Thus, it would for instance be recommended for training programs to offer support to female beneficiaries in terms of child and aged care needs. Long-term policy options • Gradually increase the minimum retirement age to 65, and thereafter link it to life expectancy at retirement. A higher minimum retirement age is expected to act as a strong signal to both employers and employees, thereby increasing the employment rate at older ages (see Geppert et al. 2019). In turn, this will help mitigate at least some of the impacts of aging on growth. The minimum retirement age should not be abruptly increased. Instead, the transition process should be well-considered, well- communicated and inclusive. For instance, the transition to a minimum retirement age of 65 could happen over a period of ten years and include the provision of transparent and clear information, coupled with inclusive consultations. This could build societal awareness and consensus, and thus at least partly mitigate some of the political economy challenges that are to be expected. After the minimum retirement age reaches 65, it could be linked to life expectancy, which is projected to increase further in the future. Linking the minimum retirement age to life expectancy will remove the need for recurring discretionary policy revisions. Over the medium term, more consistency between the minimum retirement age, the mandatory retirement age for civil servants, the EPF minimum withdrawal age and other social insurance parameters (see below) would be desirable. 37 The Employment Act defines a part-time employee as a person who has entered into an employment contract or a contract of service with an employer, does not earn more than RM2,000 per month, and whose average hours of work per week are more than 30 percent but do not exceed 70 percent of the normal hours of work per week of a full-time worker employed in the same capacity by the same employer or 48 hours. It does not cover employees who are engaged occasionally or whose working hours in one week do not exceed 30 percent of the normal working hours of a full-time employee in one week and employees who are working from their own residence. 116 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations • Consider piloting financial incentives to encourage the employment of older workers, such as targeted, conditional, and time-bound wage subsidies. Wage subsidies are a common instrument for offsetting gaps between the pay and (perceived) productivity of older workers, and can be effective in promoting reemployment when they have strict eligibility criteria and are carefully implemented (see OECD 2019). They can be particularly worthwhile following the labor market crisis created by COVID-19. It may be advisable to limit eligibility to a relatively narrow age group, for example age 55 to 65, and to those who have been unemployed and actively looking for a job for more than a year, as documented through EIS records or similar means. The wage subsidies should ideally also be time-bound. Finally, strong monitoring and evaluation arrangements should ensure its effectiveness in boosting employment among older workers in a cost-effective manner while not hurting other groups of workers. POLICY DIRECTION 2 Provide minimum income protection to all older persons through increased coverage and adequacy of savings and social insurance schemes, as well as a modest, broadly targeted tax- financed social pension. Short-term policy options • Develop an integrated vision of a system for old-age income protection, encompassing strengthened mandatory and voluntary retirement savings schemes and a substantially stronger role for tax-financed social assistance that provides minimum protection to all. Strengthening old-age income protection will require a substantial effort as well as an integrated vision of measures to strengthen mandatory and voluntary retirement savings schemes. This should be coupled with a substantially stronger role for tax-financed social assistance to poor or vulnerable older persons. In other words, the core objective would be to improve the adequacy and coverage of old-age income protection arrangements for all (see Figure 4.1 and Figure 4.2). An integrated vision could unite multiple schemes to address the diverse needs and circumstances of the population both working and in retirement. It could also provide a common framework for the more detailed instruments and incentives that would be needed for the self-employed, informal and gig-economy workers, workers in the agricultural sector, and other specific groups. FIGURE 4.1: Stylized representation of coverage FIGURE 4.2: Stylized representation of coverage and adequacy of old-age income protection based and adequacy of old-age income protection based on current setup on integrated vision KWAP Effective replacement rate Effective replacement rate KWAP PRS Social Increase in PRS EPF coverage pension with expanded and adequacy coverage and bene ts BOT EPF EPF Informal Coverage of older persons Formal Informal Coverage of older persons Formal Source: Authors. Source: Authors. A Silver Lining: Productive and Inclusive Aging for Malaysia 117 Chapter 4: Conclusions and Recommendations BOX 4.1 Mitigating the macroeconomic impacts of aging Policies that increase the female labor force participation rate and improve the quality of human capital can help mitigate the impacts of aging on economic growth in Malaysia. Simulations using the LTGM introduced in Section 2.2 show that policy measures to increase the size of the labor force in Malaysia, including by closing the gap between female and male labor force participation rates, can boost long-term GDP growth. Additionally, term GDP growth can be increased through measures that build the human capital of the labor force such as improvements in the quality of basic education, increases in the educational attainment and health of future generations, and training to reskill and upskill the existing workforce, including older workers. Based on projections using the LTGM, an increase in the female labor force participation rate to the 25th to 75th percentiles of high-income economies would boost average GDP growth between 2020 and 2050 by up to 0.36 percentage points. Despite a substantial increase in recent years, Malaysia’s female labor force participation rate still lags behind high-income comparators and regional peers like Thailand, China and Singapore (see Figure 4.3 and World Bank 2019b). Against this backdrop, three possible reform paths for female labor force participation are worth considering. In the weak, moderate and strong scenario the female labor force participation increases to the 25th, 50 th, and 75th percentiles among high- income countries, respectively. The length of the adjustment period is based on the historical experience of representative high-income countries and adjustment is assumed to continue linearly thereafter (see Figure 4.4). For the weak reform scenario, it takes 27 years for Malaysia to increase the female labor force participation rate from 55 percent to 62 percent, which increases average GDP growth from 2020 to 2050 by 0.14 percentage points. Under the moderate reform scenario, it takes 23 years to increase the participation rate to 69 percent, which increases average GDP growth by 0.31 percentage points. In the strong reform scenario, it takes 27 years to increase the participation rate to 74 percent. This increases average GDP growth by 0.36 percentage points. Raising the human capital of the workforce can also help to mitigate the negative impact of its shrinking size on economic growth. In the LTGM’s baseline scenario, Malaysia’s annual growth rate in human capital—measured as a composite of various indicators such as the quality of basic education and the health of the workforce—is projected to fall from 0.6 percent in 2020 to 0.1 percent by 2050. However, reforms that increase all components of human capital to their 50th percentile in the distribution of high- income economies could practically stabilize the growth rate in human capital at 0.7 percent in 2050 (see Figure 4.5). While these reforms on their own could only partially offset the decline in Malaysia’s long- term growth due to aging (in contrast to findings for other countries by Bloom, Prettner and Strulik 2013 and Lee and Mason 2010), they would nevertheless be expected to have a significant positive long-term impact. A comparison of baseline projections and the alternative scenario—that assumes an increase in all components of human capital to their 50 th percentile in the distribution of high-income economies— demonstrates that in the baseline scenario headline GDP growth would fall to 1.8 percent in 2050. In the alternative scenario, it would be 2.5 percent (see Figure 4.6). 118 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations FIGURE 4.3: Female labor force participation FIGURE 4.4: Baseline and alternative rate by country and country group, Percentage projections for female labor force participation rate, Percentage 80 80 Female labor force particpation (%) Female labor force particpation (%) 60 70 60 40 50 20 40 0 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Philippines Indonesia Malaysia Republic of Korea 25th pctl HI 50th pctl HI Thailand China Singapore Japan 75 pctl HI Vietnam Historical FLFP Baseline th Weak reform Moderate reform Strong reform Source: Authors’ calculations based on World Bank World Development Source: Authors’ calculations based on LTGM. Indicators. Note: 25th pctl. HI refers to the 25th percentile among high-income countries and similarly for 50 th pctl. HI and 75th pctl. HI. FIGURE 4.5: Baseline and alternative projections FIGURE 4.6: Baseline and alternative projections for growth in human capital, Percentage for headline GDP growth rate, Percentage 1.0 4.5 0.8 Human capital growth (%) GDP growth (%) 0.6 3.5 0.4 2.5 0.2 0 1.5 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 Baseline Baseline Quantity Quantity Quantity and quality Quantity and quality Quantity, quality and adult survival Quantity, quality and adult survival Quantity, quality, adult survival and stunting Quantity, quality, adult survival and stunting Source: Authors’ calculations based on LTGM. Source: Authors’ calculations based on LTGM. A Silver Lining: Productive and Inclusive Aging for Malaysia 119 Chapter 4: Conclusions and Recommendations • Broaden EPF coverage, including through a mandate or default opt-in for the self-employed, improved oversight and reporting, cooperation with industry associations, stronger linkages between EPF and SOCSO, and increased tax-financed matching contributions. Several complementary measures could be implemented to increase the proportion of workers actively contributing to EPF. First, the scope of contributors could be broadened to cover the self-employed, farmers and employment insurance beneficiaries. The self-employed could either be required to contribute or be governed by a default option whereby all are registered with EPF when they register with the Companies Commission or the Inland Revenue Board but can opt-out in writing. Second, oversight could be improved, including through a requirement for the registration of all workers as a condition for the granting of employers’ business licenses or Government contracts. Compliance monitoring could also be improved using a risk-based approach that prioritizes types of firms more likely to be noncompliant, as well as select auditing of wage reporting to EPF and the Inland Revenue Board. Third, cooperation with industry associations and other aggregators such as gig economy platforms could be encouraged or mandated. Fourth, linkages between EPF’s i-Saraan scheme and SOCSO’s Self Employment Social Security Scheme could be strengthened. Fifth, the matching contribution in the i-Saraan scheme could be increased to create a stronger incentive. Box 4.2 shows that matching contributions can provide a powerful incentive to informal workers to save for old age but that design features are crucial. Sixth, strengthened seminars, web-based information, and learning tools such as retirement calculators can increase the public awareness of the importance of savings for retirement. Medium-term policy options • Introduce a modest, broadly targeted tax-financed social pension to ensure minimum old-age income protection for all based on a careful study of design options weighing the size of the target population, the benefit level, incentives for EPF participation, and fiscal costs. As discussed in Section 3.2, in principle, a social pension could either be targeted at all those above a certain age, restricted to those with insufficient income or EPF balances, or provided as a supplemental benefit for those in households qualifying for other social assistance. In the short or medium term, one practical option would be to introduce a relatively modest social pension of the same benefit level as BOT of RM350 per month targeted at the B40 age 65 and over and using the BSH implementation system. Projections indicate that the introduction of this benefit would cost about 1.6 percent of projected fiscal revenues in 2020, with increases in line with the growth in the population age 65 and above thereafter.38 Over time, the social pension’s coverage or adequacy could be expanded in line with available fiscal space. For instance, the benefit level could be increased over time to 30 percent of the median wage while the coverage could gradually approach a larger and larger share of the population age 65 and above (though at a considerably higher cost). • Strengthen work injury, disability and survivorship insurance through parametric changes to SOCSO’s relevant schemes, further expansion of protection to informal workers, and the establishment of a one-stop shop for EPF, SOCSO and HRDF contributions. Strengthening protection against disability and death could include the alignment of SOCSO’s retirement age with EPF’s minimum withdrawal age. In addition, all SOCSO benefits could be automatically indexed to inflation and a “no-fault injury coverage” could be adopted, thereby effectively extending coverage of accidents 24/7. Adjustments in other relevant parameters could be studied, including the minimum benefit, constant attendant allowances (paid to SOCSO members so severely incapacitated that they require the constant personal attendance of another person), maximum insurable earnings, and possibly the contribution 38 Without improvements to the targeting performance of BSH, the estimates for coverage and fiscal costs of the social pension are likely to be lower bounds. 120 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations rate (to assure it remains actuarially fair). SOCSO’s expansion of employment injury protection to some informal workers has been an important initiative to increase coverage. This initiative could be further expanded to other groups of informal workers, as intended, and further strengthened by measures to incentivize registration and compliance including auto-enrollment (similar to the recommendations for EPF above). Finally, to reduce the administrative burden of contributions, a one-stop shop could be established which unifies the contribution collections for EPF, SOCSO, and HRDF. Such a one-stop shop could then distribute earmarked contributions to each of the respective institutions. This would reduce the burden on individual employers and make it easier for the self-employed or informal workers to participate. Long-term policy options • Gradually adjust the minimum withdrawal age for EPF Account 1 from 55 to 65, perhaps over a 10- or 20-year period, and consider converting EPF Account 2 to retirement savings. There is a need for measures to increase the contribution histories of workers who are active EPF members to ensure that retirement savings could be sustained for a longer period of time. First, the minimum withdrawal age for EPF Account 1 could be gradually increased from 55 to 65, perhaps over a 10- to 20-year period. This would substantially increase accumulated balances and create incentives for individuals to work longer. In the long run, the age for benefit eligibility might need to be further increased beyond 65, in parallel with Malaysia’s increasing life expectancy. Second, consideration could be given to converting contributions to EPF Account 2 to retirement savings. Before such a step is taken, it would be necessary to evaluate whether there still is a continued rationale for having a mandatory savings program for housing, education and health. Third, the EPF contribution rate could be reduced as the benefit eligibility age increases, provided that there is an observed increase in coverage and adequacy for most workers. Also, a cap on A Silver Lining: Productive and Inclusive Aging for Malaysia 121 Chapter 4: Conclusions and Recommendations BOX 4.2 Expanding pension coverage by offering matching contributions Achieving broad pension coverage and adequate levels of income protection in old age remains an elusive goal for nearly every country. In general, there is a strong relationship between the level of per capita income and participation in formal pension systems. However, there are 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 play a central role in the dynamics of pension coverage and benefit levels. A growing number of countries aim to increase coverage and provide a retirement benefit to informal workers through matching contributions. Those contributions provide tangible incentives for individuals to participate in pension funds, rather than the traditional approach of mandating participation and providing preferential tax treatment for retirement-account contributions. That is especially true for low-income groups and individuals who may not participate in the formal labor market and therefore receive no advantage from tax-based incentives. In principle, matching 122 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations contributions may be provided for public programs or by the sponsors of private occupational plans offered by insurance companies or the like. In practice, nearly all current examples are associated with individual retirement savings accounts. There is consistent empirical evidence that matching participants’ contributions to retirement savings can be effective in increasing participation. The evidence from high-income countries indicates positive but modest effects of matching on participation, with overall effects increasing participation in the range of 5 to 10 percent of potential beneficiaries. The associated finding that a 25 percent match of individual contributions is associated with about a 5 percent increase in participation appears robust. That magnitude is also broadly consistent with results from the Republic of Korea, where a 50 percent match for farmers and fishers increased the probability of their making a pension contribution by 7.4 percent. In New Zealand, a large initial match—a significant element of the KiwiSaver system—elicited enrollment from many people with little or no earnings, providing further evidence of the potential effectiveness of significant matches. The structure of the match—the matching rate, thresholds, and caps—has significant consequences for its effect on the saving rate. The match threshold seems to have a greater impact than the matching rate. Providing a lower match, for example, of 25 percent on contributions, of up to 10 percent of pay (a relatively high share) will induce individuals to save more than a 50 percent match of up to 5 percent of pay, although both formulations may result in similar costs to the organization providing the match. One possible explanation for that result is that matching acts as a signaling device or implicit advice on saving levels. Also notable is the “stickiness” of saving levels, as evidenced by the fact that most people’s contribution levels remain at the default level even when the defaults are subsequently reduced. The international evidence strongly suggests that other features of savings programs and related interventions may have a critical effect on participation and contribution levels. First, evidence from the United States, the United Kingdom, and New Zealand suggests that making participation the default option has two to four times as much impact as the reverse (of course, an automatic enrollment default option works only under specific employment conditions). The evidence regarding schemes from other countries that make participation the default option on contribution efforts is more mixed and at times negative, possibly because of inertia or the low default contribution rate. Second, empirical results for the United States and the United Kingdom and lessons from the German Riester pensions suggest that simplified design affects participation and, perhaps, contribution and saving efforts. Third, social marketing and advocacy can be key. Retirement saving remains an objective that most will embrace but find difficult to implement. In the United States, information sessions and advocacy have been a useful adjunct to the incentives of matching contributions. In Germany, take-up of the match increased after information campaigns generated greater awareness. New Zealand has coupled the introduction of its system with information and advocacy campaigns which are perceived to have had a positive effect. Source: World Bank (2019b) based on Hinz et al. (2013). A Silver Lining: Productive and Inclusive Aging for Malaysia 123 Chapter 4: Conclusions and Recommendations wages subject to mandatory EPF contributions could be established, similar to the way in which a cap has been established for SOCSO contributions. The level of the cap could be relatively high to ensure that retirement savings are adequate. Similarly, EPF contribution rates could be lowered for individuals who have accumulated sufficient balances for both a minimum benefit in absolute terms and a reasonable replacement rate. As with the recommended increase in the minimum retirement age, all initiatives to increase the contribution histories of EPF members would need to be gradual, well-communicated and inclusive. • Transition to phased withdrawals of EPF balances and explore annuitization options, longevity insurance and age-based portfolios. As an intermediate step toward the exploration of annuitization options, phased withdrawals of EPF balances could be mandated to spread income over time. As illustrated in Figure 4.7, phased withdrawal could be based on a calculation that takes the EPF account balance at retirement and divides it by the life expectancy at retirement. Each year, the benefit would be recalculated based on the revised account balance and the life expectancy at the new age. If an individual becomes gravely ill either while working or retired, hardship provisions could permit additional withdrawals. Over the longer-term, age-based portfolios that invest in riskier portfolios for young people while reducing the risk as individuals age could be explored to achieve returns that are improved but remain smooth. Longevity insurance and annuitization options could also be explored as a means of covering risks related to individuals outliving their cash balance and to provide an effective means of sharing longevity risks across retirees. As phased withdrawals offer no insurance against longevity risks, it is also possible for a portion of an individual’s cash balance to be held back until reaching a very high age to ensure that resources are provided for those who live beyond the average life expectancy. It would also be opportune to explore options for the EPF to underwrite the risk of indexed, annuitized benefits while likely drawing upon the expertise of insurance companies to appropriately structure and price such benefit options. Once again, all relevant measures should be introduced gradually as part of a well- considered, well-communicated and coordinated transition process (see Table 4.2). FIGURE 4.7: Simulated EPF balances by age TABLE 4.2: Possible transition phasing under phased withdrawal, RM 50,000 2,000 Change Possible transition phasing 1,000 New Account 1 min. Increase by 6 months every 40,000 Simulated annual change (RM) Simulated EPF balances (RM) withdrawal age year; thus, the increase from 0 age 55 to 65 will take 20 30,000 years. -1,000 Convert Account Five-year transition 20,000 2 to retirement -2,000 savings 10,000 -3,000 Transition to phased Ten-year transition with withdrawals gradual increase in 0 -4,000 proportion of balance 30 40 50 60 70 80 90 with mandatory phased Age withdrawal. Voluntary Once there is a cost-effective Balance - Phased withdrawal at age 55 annuitization system for underwriting risk Annual change - Phased withdrawal at age 55 and covering costs. Source: Authors’ calculations based on EPF Source: Authors. 124 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations POLICY DIRECTION 3 Build an inclusive aged care system with an enabling market and regulatory environment for private not-for-profit provision, strengthened governance, and selectively increased public financing. Short-term policy options • Develop a systematic and actionable aged care strategy based on a comprehensive assessment and diagnostic. To move toward developing an inclusive aged care system, the Government could formulate an actionable aged care strategy that outlines policy priorities, sequences, and roadmaps to help implement specific policy measures. The increasing complexity and scale of aged care needs now and in the next few decades, coupled with changing socioeconomic circumstances and living arrangements, are already severely straining the current aged care model. Building on the analysis of this report and the relevant literature, further in-depth analyses including a comprehensive assessment and diagnostic can inform the development of the strategy. Such in-depth analysis would also allow the Government to draw relevant lessons learned from international experiences and best practices to inform the design and implementation of specific policy measures. Finally, the development of an actionable strategy would benefit from the inclusive involvement of many stakeholders in the sector. • Improve aged care quality standards to ensure the health and safety of residents of aged care homes and the quality of aged care services. Although infrastructure and facility standards for aged care homes are already in place, against the backdrop of the COVID-19 pandemic and its impacts on the residents of aged care homes, there is a need to further develop service standards and strengthen the enforcement of these. Steps in enhancing aged care quality standards and enforcement capacity would include (i) developing a national quality standards framework, (ii) strengthening enforcement efforts, (iii) enhancing the Government’s capacities to write detailed specifications, protocols, and contracts for aged care services, and (iv) providing training for monitoring, inspection, and implementation of aged care quality standards. In many OECD countries, aged care systems rely heavily on inspection and regulation of long-term care providers to ensure quality. Quality standards usually cover facilities (construction, equipment and services), but less often include home-care services. Instead, for home-care services, it is more common to have standards for qualifications and training of staff. For enforcement, a shift has been witnessed from penalties to incentives for aged care providers with good performance. Malaysia could study and potentially follow the OECD countries’ approaches. Medium-term policy options • Streamline and harmonize licensing requirements and processes for existing unlicensed aged care facilities and potential new market entrants. Private providers of aged care and investors into aged care are important stakeholders. Their entry into the regulated aged care market could be facilitated if licensing and accreditation requirements and processes were streamlined and harmonized across the relevant regulatory agencies. For this purpose, a mechanism such as that used to streamline and harmonize the technical requirements and submission processes for private hospitals could be considered. In parallel, pathways toward regularizing the large number of unlicensed private aged care facilities could be considered, in particular against the backdrop of the COVID-19 pandemic and its health and safety implications for residents of aged care homes. The guiding principle would be that regulatory requirements should foster instead of stifle innovations in aged care service delivery. • Strategically reorient public financing toward home and community-based aged care, as well as toward rural and lagging regions. More than four-fifths of public financing for aged care services A Silver Lining: Productive and Inclusive Aging for Malaysia 125 Chapter 4: Conclusions and Recommendations is allocated to RSK and RE which serve less than 2,000 older persons throughout Malaysia. This is even though home and community-based aged care services tend to have much lower unit costs, are generally preferred by almost all older persons and their families, and are typically able to provide adequate care to almost all older persons other than those with the most intensive care needs. Therefore, strategically reorienting public financing toward home and community-based aged care is recommended. In addition, there is also a need to address the imbalance of aged care coverage across states and between rural and urban areas. Long-term policy options • Strengthen arrangements for training and qualifications to upgrade skills of aged care workers. Malaysia is not the only country that faces a shortage of care workers. To address similar challenges, many OECD countries have strengthened arrangements for relevant active labor market policies (ALMPs). Such arrangements would introduce systematic regulatory arrangements that enable (i) the collection of training needs and setting up job qualification standards, (ii) the opening of training programs and increased access to training and skills development, (iii) enhancement of training capacity through increasing public inputs and strengthening technical and vocational education and training, and (iv) adoption of employment and labor market measures to increase job attractiveness for the aged care sector.39 Beyond training and qualifications of local aged care workers, more demand-driven labor migration policies could also contribute to improving the supply of qualified aged care workers in Malaysia. • Selectively increase public financing of aged care services to incentivize a market-oriented and needs assessment-based model of aged care provision and mobilize private and social investment. Given the limited available fiscal space, public financing for aged care services should be increased and allocated strategically. In terms of beneficiary eligibility, it is international best practice to establish transparent criteria based on a combination of needs assessments and income or asset tests. Based on such criteria, public financing could be targeted equitably and efficiently but with sufficient coverage of the “missing middle,” including the B40. The introduction of more systematic and impactful needs assessments will be important to identify who really has physical and economic needs (see Box 4.3). Needs assessments can also be used to develop customized care plans that meet the diverse needs of older persons. For service provision, the Government could commission more private service providers and more directly monitor their performance to ensure service quality. Tentative steps toward increased financing for commissioned or purchased aged care services could be accelerated given that current direct public provision does not appear to be cost-effective. In addition, a results-based commissioning approach, linked to a strengthened monitoring and evaluation framework, could be studied, piloted and scaled up. Innovative measures that could be supported may include the development of “time banks” and “old folks’ partnership” to promote cooperative assistance between younger and older persons and among older persons. The promotion of productive and inclusive aging implies a strong role for the government, possibly entailing additional fiscal expenditures and increasing the need to need to address challenges related to increasing revenue. Even gradual increases in fiscal expenditures may be challenging in the context of the downward trend in government revenues and the fiscal shock of the COVID-19 pandemic. Malaysia’s collected revenues have been on a downward trend since 2012, driven by a confluence of factors, 39 There also remains room to expand the coverage of HRDF to relevant subsectors. While HRDF currently covers private hospital services and health support services, it covers neither residential care activities nor social work activities without accommodation. 126 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations including the under-collection of personal and consumption taxes, the provision of generous tax incentives, and the failure to develop new sources of revenue. World Bank estimates suggest that in the wake of the COVID-19 pandemic the fiscal deficit could widen to as high as 7 percent of GDP (see World Bank 2020). In the medium and long term, there is therefore a need for the government to diversify its revenue base and to increase its revenue collection through more progressive taxation. Thus, the promotion of productive and inclusive aging will require a mixture of public policy measures on both the expenditure and revenue sides.40 In addition to fostering productive employment, providing minimum income protection to older persons, and building an inclusive aged care system, policy efforts in other areas will also be needed to achieve productive and inclusive aging. While this report focuses on the policy areas of employment, income security for older persons, and aged care, a comprehensive analysis of some other relevant policy areas is left for future research. In particular, the report does not specifically discuss Malaysia’s health sector, in spite of the crucial importance of healthy aging. Other relevant policy areas that are left for future research include the fiscal sustainability of the civil service pension system and the interplay between aging and urban planning, as well as housing and mobility. Finally, while the report recognizes that managing aging requires a life cycle perspective, it does not try to comprehensively address all relevant policy areas but focuses on policies either directly affecting older persons or with close linkages or spillovers between older persons and others. Other policy areas that are important from a life cycle perspective, such as children’s nutrition and basic education, are again left for future research. As documented in this report, with the right mix of policies, the Government of Malaysia can make sure that aging is both productive and inclusive. Recognizing that aging is both a challenge and an opportunity, the right mix of policies can enable Malaysia to adapt to rapid aging, improve the well-being of all, and increase the share of persons that experience old age not as a burden but as a period of their life full of independence, social inclusion and productivity (see Interview quote 4.1). INTERVIEW QUOTE 4.1 “One thing that I feel satisfied is I am able to do my own stuff now... I even get to attend [college]... I get to absorb various knowledge from there... I also get to see my kids married, and they stay with their own family now. I am happy because I already don’t have any responsibility, they finished their education... During the weekend, we also have family gathering as well. ...I am still active. I go out of the house daily.” Source: 73-year old male, independent older person 40 To address challenges related to increasing revenue, World Bank (2020) identifies the following measures: First, the progressivity of the personal income tax framework could be increased, with a reassessment of current reliefs and exemptions, as a means both to increase revenue and to redistribute income. Second, capital gains taxes could be expanded, with explorations to assess the feasibility of other forms of taxes on non-earned income. In Malaysia, inequality in wealth is significantly greater than inequality in income, so taxing a share of the gains made from asset price inflation could provide an important new source of public sector revenue. Third, indirect taxation could be broadened by restricting zero rated and exempted items to a more limited set of goods and services. A gradual lifting of the SST exemptions and zero-ratings on selected nonessential items, particularly those not within the B40 consumption basket, could facilitate increased revenue without jeopardizing the purchasing power of lower-income households. A Silver Lining: Productive and Inclusive Aging for Malaysia 127 Chapter 4: Conclusions and Recommendations BOX 4.3 Using needs assessments to determine access to and scope of aged care services An aged care needs assessment is a systematic tool to measure the degree of older persons’ limitations and dependencies, and to identify their unmet care needs. One aspect of an aged care needs assessment is the measurement of an older persons’ functional ability to live independent of care from others. This measurement often uses specific tools to collect individual information covering various dimensions such as ADLs, IADLs, cognition, communication, pain, and depression. This information is then used to generate scores on the degree of an individual’s functional ability. Often, 128 A Silver Lining: Productive and Inclusive Aging for Malaysia Chapter 4: Conclusions and Recommendations these scores will be complemented by an assessment of the older person’s living conditions, including an examination of his or her household composition, financial conditions and living environment (e.g., dwelling conditions and location). In the next step, the results of the two assessments on functional ability and living conditions can then be combined into one measure or matrix to determine eligibility for publicly financed aged care services. It is noteworthy that in addition to Governments, private aged care service providers often also have their own assessment tools, which are mostly used for pricing and admission for customers. Aged care needs assessments often involve screenings and professional assessments as two separate steps. Screenings often use a simple set of questions to be posed to older persons to determine if a professional assessment is required. The screening can be done over the phone or face-to-face and collects basic information on a person’s needs, circumstances and functional ability. If a professional assessment is required, one or a team of trained assessor(s) will be assigned to the case and tasked with using specific tools to carry out a comprehensive assessment, based on the information collected from the screening. For home and community-based care, the assessment tools are often less complex compared to those for institutional care. The latter encompasses similar client information as the former, but at a deeper level, and the assessor(s) will usually comprehensively assess the older person’s physical capabilities, medical conditions, psychosocial factors, cognitive and behavioral factors, physical environmental factors, and restorative needs, if any. Internationally, various assessment tools to measure an older persons’ degree of functional limitations have been developed and used for publicly financed aged care services. For instance, the Resident Assessment Instrument / Minimum Data Set has been used in the United States for the purposes of health status screenings and eligibility determination by nursing homes financed through the public Medicare and Medicaid programs. Some countries have built on the Resident Assessment Instrument / Minimum Data Set to develop their own assessment tools for publicly financed aged care programs. Relevant examples include the Aged Care Funding Instrument in Australia, Easy Care in the United Kingdom, and the Long-Term Care Assessment in Japan. Results from aged care needs assessments can serve to inform public financing decisions and the development of individual older persons’ care plans. The results from aged care needs assessments are helpful for understanding the types of services and the level of care that older persons would need, as well as determining the eligibility and level of public subsidy that they would be entitled to. With this information, the assessor(s), older person, and potentially also the older person’s family can work together to establish a care plan that reflects the older person’s strengths and abilities, areas of difficulty, and the support that will best meet his or her needs and goals. In addition, aggregate information from aged care needs assessments can also inform the Government budgeting process and the allocation of appropriate but not excessive funding to publicly financed aged care services, with the ultimate objective of meeting older persons’ needs in a fiscally sustainable way. Source: Authors based on OECD (2011) and World Bank (2019c). A Silver Lining: Productive and Inclusive Aging for Malaysia 129 References Abdullah, Hussin and Hafizah Hammad Ahmad Khan. 2010. Saving Determinants in Malaysia. Jurnal Ekonomi Malaysia 44: 23–34. Abdur Rahman, Amanina and Achim Schmillen. Forthcoming. From Farms to Factories and Firms – Structural Transformation and Labor Productivity Growth in Malaysia. World Bank: Washington, DC. Accius, Jean, and Joo Yeoun Suh. 2019. The Longevity Economy Outlook: How People Ages 50 and Older Are Fueling Economic Growth, Stimulating Jobs, and Creating Opportunities for All. Washington, DC: AARP Thought Leadership, December 2019. Acemoglu, Daron and David Autor. 2011. Skills, Tasks and Technologies: Implications for Employment and Earnings. In: Handbook of Labor Economics, Volume 4B, edited by Orley Ashenfelter and David Card, 1043–1071. Elsevier: Amsterdam. Asian Development Bank. 2018. Tapping Technology to Maximize the Longevity Dividend in Asia. Asian Development Bank: Manila. Bank of America Merrill Lynch. 2014. The Silver Dollar – Longevity Revolution ref http://www.longfinance.net/ images/reports/pdf/baml_silverdollar_2014.pdf. Bell, Andrew and Kelvyn Jones. 2013. The Impossibility of Separating Age, Period and Cohort Effects. Social Science and Medicine 93: 163–165. Bender, Keith A. 2012. An Analysis of Well-Being in Retirement: The role of Pensions, Health, and “Voluntariness” of Retirement. The Journal of Socio-Economics 41: 424–433. Bielecki, M., M. Brzoza-Brzezina and M. Kolasa. 2018. Demographics, Monetary policy and the Zero Lower Bound. Narodowy Bank Polski Working Paper No. 284. Blanchard, Robert D., James B. Bunker and Martin Wachs. 1977. Distinguishing Aging, Period and Cohort Effects in Longitudinal Studies of Elderly Populations. Socio-Economic Planning Sciences 11: 137–146. Bloom, David, D. Caning and J. Finlay. 2010. Population Aging and Economic Growth in Asia. In: The Economic Consequences of Demographic Change in East Asia, edited by Ito Takatoshi and Andrew Rose, 61-89. University of Chicago Press: Chicago. Bloom, David, David Canning, Günther Fink and Jocelyn Finlay. 2009. Fertility, Female Labor Force Participation, and the Demographic Dividend. Journal of Economic Growth 14: 79–101. Bloom, Davi, and Williamson, Jeffrey. 1998. Demographic Transitions and Economic Miracles in Emerging Asia. World Bank Economic Review 12: 415-455. Bloom, David, Klaus Prettner and Holger Strulik. 2013. Declining Fertility and Economic Well-Being: Do Education and Health Ride to the Rescue. Labor Economics 22: 70–79. Bloom, David, Somnath Chatterli, Paul Kowal, Peter Lloyd-Sherlock, Martin McKee, Bernd Rechel, Larry Rosenberg and James Smith. 2016. Macroeconomic Implications of Population Ageing and Selected Policy Responses. Lancet 385: 649–657. Boeri, Tito and Guido, Tabellini. 2012. Does Information Increase Political Support for Pension Reform? Public Choice 150: 327–362. Bonsang, Eric, Stéphane Adam and Sergio, Perelman. 2012. Does retirement affect cognitive functioning? 31: 490–501. Bonsang, Eric and Tobias J. Klein. 2012. Retirement and Subjective Well-Being. Journal of Economic Behavior and Organization 83, 311–329. Brown, Alexandra and Rochelle Guttmann, 2017. Ageing and Labour Supply in Advanced Economies. Reserve Bank of Australia Bulletin, December 2017. Bussolo, Maurizio; Koettl, Johannes; Sinnott, Emily. 2015. Golden Aging : Prospects for Healthy, Active, and Prosperous Aging in Europe and Central Asia. Washington, DC: World Bank. © World Bank. https:// openknowledge.worldbank.org/handle/10986/22018 License: CC BY 3.0 IGO. 130 A Silver Lining: Productive and Inclusive Aging for Malaysia References Calvo, Esteban, Kelly Harvestick and Steven Sass. 2009. Gradual Retirement, Sense of Control, and Retirees’ Happiness. Research on Aging 31: 112–135. Carvalho, C., A. Ferrero and F. Nechio. 2016. Demographics and Real Interest Rates: Inspecting the Mechanism. European Economic Review 88: 208–226. Center for Effective Services. 2015. Commissioning in Human, Social and Community Services – A Rapid Evidence Review. Center for Effective Services: Dublin. Cheeseman Day, J. and E. C. Newburger. 200). The Big Payoff: Educational Attainment and Synthetic estimates of Work-Life Earnings. U. S. Census Bureau: Washington, DC. Chlon-Dominczak, Agnieszka. The “Gender Pension-Gap” in the Aging Societies of East Asia. World Bank: Washington, DC. Deloitte. 2017. Beyond the Noise: The Megatrends of Tomorrow’s World. Deloitte: Munich. Dicarlo, Emanuele, Salvatore Lo Bello, Sebastian Monroy-Taborda, Ana Maria Oviedo, Maria Laura Sanchez- Puerta and Indhira Santos. 2016. The Skill Content of Occupations across Low and Middle Income Countries: Evidence from Harmonized Data. IZA Discussion Paper No. 10224. Eggertsson, G. B., N. R. Mehrotra and J. A. Robbins. 2019. A Model of Secular Stagnation: Theory and Quantitative Evaluation. American Economic Journal: Macroeconomics 11: 1–48. European Commission. 2015a. Growing the European Silver Economy. Background Paper. 23 February 2015. European Commission. 2015b. The Silver Economy: Final Report—A Study Prepared for the European Commission DG Communications Networks, Content & Technology by Technopolis group, Oxford Economics. Fontoura Gouveia, Ana. 2017. Political Support for Reforms of the Pension System: Two Experiments. Journal of Pension Economics & Finance 16: 371 – 394 Geppert, Christian, Yvan Guillemette, Hermes Morgavi and David Turner. 2019. Labour Supply of Older People in Advanced Economies: The Impact of Changes to Statutory Retirement Ages. OECD Economics Department Working Papers No. 1554. Gordo, Laura Romeu and Vegard Skirbekk. 2013. Skill demand and the comparative advantage of age: Job tasks and earnings from the 1980s to the 2000s in Germany. Labour Economics, 22: 61–69. Gruber, Jonathan, Kevin Milligan and David A. Wise. 2010. Introduction and Summary. In Social Security Programs and Retirement around the World: The Relationship to Youth Employment, edited by Jonathan Gruber and David A. Wise, 1–45. University of Chicago Press: Chicago. Hamid, Tengku Aizan. 2020. Elderly Poverty in Malaysia. Presentation at Network of Malaysian Think Tanks: Alleviating Old-age Poverty in Ageing Malaysia, 18 February 2020. Hasmuk, Kejal, Hakimah Sallehuddin, Maw Pin Tan, Wee Kooi Cheah, Rahimah Ibrahim, Sen Tyng Chai. 2020. The Long Term Care COVID-19 Situation in Malaysia. Last updated 30 May 2020. Hevia, Constantino and Norman Loayza. 2013. Saving and Growth in Sri Lanka. World Bank Policy Research Working Paper 6300. Institute of Labour Market Information and Analysis. 2019. National Strategic Development Plan on Ageing Population: Inclusion and Employment of Malaysia’s Ageing Population. ILMIA: Cyberjaya. International Labor Organization. 2018. Care Work and Care Jobs for the Future of Decent Work. Geneva. International Monetary Fund. 2010. From Stimulus to Consolidation: Revenue and Expenditure Policies in Advanced and Emerging Economies. IMF: Washington, DC. Kinugasa, Tomoko and Andrew Mason. 2007. Why Nations Become Wealthy: The Effects of Adult Longevity on Saving. World Development 35: 1–23. Kinsella, Kevin and Wan He. 2009. An Aging World: 2008. U.S. Government Printing Office International Population Report PS95/09-1. Khazanah Research Institute. 2018. The Unsung Labour: Care Migration in Malaysia. KRI: Kuala Lumpur. Lee, Namchul and Ji-Sun Chung. 2008. Interraction Between Fertility and Female Labor Force in Korea. Journal of Applied Business Research 24. A Silver Lining: Productive and Inclusive Aging for Malaysia 131 References Lee, Ronald and Andrew Mason. 2006. What Is the Demographic Dividend? Finance and Development 43: 16–17. Lee, Ronald, and Andrew Mason. 2010. Some Macroeconomic Aspects of Global Population of Global Population Aging. Demography 47 (1) 151-172. Lee, Ronald and Andrew Mason. 2017. Cost of Aging. Finance and Development 54: 7–9. Li, Hongbin, Jie Zhang and Junsen Zhang. 2007. Effects of Longevity and Dependency Rates on Saving and Growth: Evidence from a Panel of Cross Countries. Journal of Development Economics 84: 138–54. McKinsey. 2015. The Four Global Forces Breaking All the Trends. McKinsey: San Francisco Momtaz, Yadollah, Tengku Aizan Hamid, and Rahimah Ibrahim. 2012. Unmet Needs among Disabled Elderly Malaysians. Social Science & Medicine 75: 859—863. Munnell, Alicia and April Wu. 2013. Will Delayed Retirement by the Baby Boomers Lead to Higher Unemployment among Younger Workers? Center for Retirement Research at Boston College Working Paper 2012–22. Nikoleva, Milena and Carol Graham. 2014. Employment, Late-Life Work, Retirement, and Well-Being in Europe and the United States. IZA Journal of European Labor Studies 3. Organisation for Economic Co-Operation and Development. 2011. Help Wanted? Providing and Paying for Long-Term Care. OECD: Paris. Organisation for Economic Co-Operation and Development. 2019. Ageing and Employment Policies: Working Better with Age. OECD: Paris. Parlevliet, Jante. 2017. What Drives Public Acceptance of Reforms? Longitudinal Evidence from a Dutch Pension Reform. Public Choice 173: 1–23. Penner, Rudolph G., Pamela Perun and C. Eugene Steuerle. 2002. Legal and Institutional Impediments to Partial Retirement and Part-Time Work by Older Workers. The Urban Institute: Washington, DC. Rabi, Ahmad, Norma Mansor, Halimah Awang and Nurul Diyana Kamarulzaman. 2019. Longevity Risk and Social Old-Age Protection in Malaysia. Social Wellbeing Research Centre: Kuala Lumpur. Rannan-Eliya, Ravi, Chamara Anuranga, Adilius Manual, Ruwani Wickramasinghe, Prasadini Perera, Shanaz Saleem, Jayalal Chandrasiri, Indra Pathmanathan and Institute for Health Systems Research Core Team. 2013. Malaysia Healthcare Demand Analysis: Inequalities in Healthcare Demand and Simulation of Trends and Impact of Potential Changes in Healthcare Spending. Institute for Health Systems Research: Kuala Lumpur. Richards, Nicola, Hebe Gouda, Jo Durham, Rasika Rampatige, Anna Rodney and Maxine Whittaker. 2016. Disability, Noncommunicable Disease and Health Information. Bulletin of the World Health Organization;94:230-232. Rohwedder, Susann and Robert J. Willis. 2010. Mental Retirement. Journal of Economic Perspectives 24: 119–138. Rutkowski, Michal. 2020. 3 Ways to Improve COVID-19 Response to Elderly Care and Persons with Disabilities. July 07, 2020. https://blogs.worldbank.org/voices/3-ways-improve-covid-19-response-elderly-care- and-persons-disabilities. Schult, Paul T. 2004. Demographic Determinants of Savings: Estimating and Interpreting the Aggregate Association in Asia. Yale University Economic Growth Center Working Paper 901. Sanderson, W. C. and S. Scherbov. 2019. Prospective Longevity: A New Vision of Population Aging. Harvard University Press: Cambridge, MA. Schmillen, Achim and Heiko Stüber. 2014. Lebensverdienste nach Qualifikation: Bildung lohnt sich ein Leben lang. Institute for Employment Research: Nuremberg. Skirbekk, F. Vegard, Martin Stonawski, Eric Bonsang and Ursula M. Staudinger. 2013. The Flynn Effect and Population Aging. Intelligence 41: 169–177. Tan, Peck-Long and John Gibson. 2013. Impact of Foreign Maids on Female Labor Force Participation in Malaysia. Asian Economic Journal 2: 163–183. 132 A Silver Lining: Productive and Inclusive Aging for Malaysia References Trucchi, Serena, Elsa Fornero and Mariacristina Rossi. 2018. Retirement Rigidities and the Gap between Effective and Desired Labour Supply by Older Workers. IZA Journal of Labor Policy 7: 1–19. United Nations. 2019. World Population Prospects 2019 – Methodology of the United Nations Population Estimates and Projections. United Nations: New York. Wadensjö, Eskil. 2006. Part-Time Pensions and Part-Time Work in Sweden. IZA Discussion Paper 2273. Wang, Dewen, Achim Schmillen, Elena Glinskaya, Amanina Abdur Rahman, Alyssa Farha Jasmin, Linhui Zhu and Maria Iglesia Gomez. 2020. The Elderly Care Response to Covid-19: A Review of International Measures to Protect the Elderly Living in Residential Facilities and Implications for Malaysia. World Bank: Washington, DC. Wickrama, Kandauda (K. A. S.), Catherine W. O’Neal, Kyung H. Kwag and Tae K. Lee. 2013. Is Working Later in Life Good or Bad for Health? An Investigation of Multiple Health Outcomes. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences 68: 807–815. World Bank. 2008. The World Bank Pension Conceptual Framework, World Bank: Washington, DC. World Bank. 2012. World Bank Social Protection and Labor Strategy: Resilience, Equity, and Opportunity. World Bank: Washington, DC. World Bank. 2013. Capital for the Future: Saving and Investment in an Interdependent World. World Bank: Washington, DC. World Bank. 2015. Social Protection for a Productive Malaysia World Bank: Washington, DC. World Bank. 2016. Live Long and Prosper: Aging in East Asia and Pacific. World Bank: Washington, DC. World Bank. 2017a. Case Study on the Employees Provident Fund of Malaysia. World Bank: Washington, DC. World Bank. 2017b. Public-Private Partnerships: Reference Guide Version 3. World Bank: Washington, DC. World Bank. 2017c. Study on the Effectiveness of the Human Resources Development Fund. World Bank: Washington, DC. World Bank. 2018. The State of Social Safety Nets. World Bank: Washington, DC. World Bank. 2019a. Agricultural Transformation and Inclusive Growth: The Malaysian Experience. World Bank: Washington, DC. World Bank. 2019b. Breaking Barriers: Toward better Economic Opportunities for Women in Malaysia. World Bank: Washington, DC. World Bank. 2019c. Guizhou Aged Care System Development Program. World Bank: Washington, DC. World Bank. 2020. Malaysia Economic Monitor June 2020: Surviving the Storm. World Bank: Washington, DC. World Bank. Forthcoming. Aiming High: Navigating the Next Stage of Malaysia’s Development. World Bank: Washington, DC. World Economic Forum. 2017. Four Mega-Trends That Could Change the World by 2030. World Economic Forum: Geneva. World Data Lab. 2019. Silver Economy Spending Power Trends in Aisa. December 5, 2019. https://worlddata. io/blog/silver-economy-asia World Health Organization. 2015. World Report on Ageing and Health. World Health Organization: Geneva. Zarinah Mahari. “Demographic Transition in Malaysia: The Changing Roles of Women,” 2011., 31. Zhang, Chuanchuan and Yaohui Zhao. 2012. The Relationship between Elderly Employment and Youth Employment: Evidence from China. Munich Personal RePEc Archive Working Paper 37221. A Silver Lining: Productive and Inclusive Aging for Malaysia 133 Methodological Annex 134 A Silver Lining: Productive and Inclusive Aging for Malaysia Methodological Annex A. Definitions TABLE A.1: Definitions Term Definition Age-management strategies Human resource strategies that combat age barriers and/or encourage age diversity at the workplace. Aging society Occurs when the share of the population age 65 and above is at 7 percent or more. Aged society Occurs when the share of the population age 65 and above is at 14 percent or more. Contributory social insurance Entitlement to benefits depend on the contributions made by beneficiaries and their employers, for example old-age pension schemes. Employment income Income from paid employment and self-employment activities. Labor force All persons who are employed or unemployed but not those not actively looking for work or not available for work. Late-career workers Workers age 50 years and older. Minimum retirement age The minimum age at which an employer can retire its employees. In Malaysia, the minimum retirement age of a private employee is upon the employee attaining the age of 60. Minimum withdrawal age The minimum age at which EPF members can withdraw their savings. For Account 1, this is set at age 55, and Account 2 at age 50. Non-contributory social Entitlement to benefits require no direct contribution from beneficiaries or their assistance employers, for example cash assistance programs. Normal retirement age The age at which a person with an uninterrupted career starting at age 22 can retire without any reduction in their pension. Old-age dependency ratio The number of persons age 65 and above divided by the working-age population. Part-time work Work of less than 30 hours a week. Poverty line income The minimum level of income deemed adequate for a particular country. In Malaysia, this is determined through a cost of basic needs approach. Super-aged society Occurs when the share of the population age 65 and above is at 20 percent or more. Total dependency ratio The number of persons age zero to 14 and over the age of 65 divided by the working-age population. Unemployment Occurs when an individual who is actively seeking and available for employment is unable find employment. Underemployment Occurs when an individual works in an employment that does not utilize their full skills, experience and availability to work. Working-age population Population between 15 to 64 years old. Young-age dependency ratio The number of persons age zero to 14 divided by the working-age population. Source: Authors. A Silver Lining: Productive and Inclusive Aging for Malaysia 135 Methodological Annex B. Qualitative Research This report relied on qualitative research to gather various stakeholders’ ideas, concerns, and expectations on aging and aged care in Malaysia. Three interrelated questions were addressed through the qualitative research: • What are the views on the roles and responsibilities of the individual older person, their families and communities, religious organizations, NGOs, the private sector, and the Government in providing and financing care for older persons? • What are the views of informal and wage caregivers who provide care to older persons e.g. with regard to their skills, remuneration, working conditions, quality of services, and help wanted from the Government? • What are the experiences, preferences, ideas, concerns, and expectations of older persons with regard to aged care currently received or expected to be received in the future? The study focused on four categories of persons, with representatives from each category selected to cover the diversity of Malaysia’s society—e.g., urban and rural, male and female, and various ethnicities. The four categories are: (i) independent older persons who are currently retired, mobile and able to take responsibility for their own health and personal issues; (ii) aged care recipients who are individuals unable to care for themselves and have another adult looking after them in order for them to have a safe and healthy environment to live in; (iii) informal caregivers who are unpaid individuals who provide physical, emotional and daily care for older persons (typically immediate or extended family members of the older person); and (iv) waged caregivers who are paid professionals that provide aged care or medical care to older persons. Qualitative field data collection was contracted to a research firm which organized and conducted 32 interviews involving 42 respondents, with support and supervision provided by the World Bank. The interview period spanned from May 30, 2020 until July 6, 2020. Due to limitations on physical, face-to- face interactions due to the COVID-19 pandemic, the interviews were conducted via online teleconferencing tools or as audio calls. All interviews were conducted in respondents’ native language and typically lasted 45 to 60 minutes. Select interviews were conducted in pairs of respondents, typically comprising one male and one female respondent. Survey instruments in the form of semi-structured interview questions and desired respondent profiles were prepared by the World Bank with feedback from the research firm. The final sample, given constraints due to COVID-19, comprised individuals from the Klang Valley as Malaysia’s economic center and Kuala Selangor, a small town in a rural setting, stratified by gender to achieve a 1:1 male/ female ratio. The majority of respondents were Malay followed by Chinese individuals. As it was challenging to find Malaysian waged caregivers—likely due to a dominance of foreign workers in this profession, see Section 3.3—the sampling was expanded to include foreign workers (see Table B.1). All interviews were transcribed and, if necessary, translated into English. A qualitative analysis of the transcripts was performed by a consultant proficient in qualitative research methods and documented in a background paper to this report (See French 2020). Throughout all phases of the field work, quality assurance was provided by the research firm and supported by the World Bank team in the form of close support and supervision. Quality assurance included the administration of an online training covering methods, sampling, and thematic guidelines for the moderators. In addition, detailed advice and feedback regarding the selection of participants, field work, initial interviews, and logistics was also provided by the World Bank team. 136 A Silver Lining: Productive and Inclusive Aging for Malaysia Methodological Annex TABLE B.1: Characteristics of respondents of qualitative research Category Sub-category Respondents Locality Ethnicity / Citizenship Older persons Independent older 12 10 Urban 8 Malay persons 2 Rural 4 Chinese Older persons Aged care recipients 12 10 Urban 8 Malay 2 Rural 4 Chinese Caregivers Informal caregivers 10 8 Urban 6 Malay 2 Rural 4 Chinese Caregivers Waged caregivers 8 8 Urban 2 Malay 0 Rural 2 Indian 1 Chinese 3 Noncitizens (Filipino) Source: Authors. A. C. Long-Term Long-Term Growth Growth A.Model Model Long-Term Growth Model 1. This report uses1.the World This Bank’s report the -Term uses Long Bank’s Model World Growth (LTGM) Long-Term to analyze Growth Model the (LTG report uses the World This macroeconomic impacts Bank’s Long-Term Growth of demographic changes, Model (LTGM) to analyze thein macroeconomic macroeconomic impacts of in particular, demographic how growth changes, the working in particular, how age to total growth in th impacts of demographic population ratio impacts changes, in particular, economic growth how ingrowth Malaysia.in the The working LTGM age is anto total extension population of the Solow-Swan population ratio impacts economic growth in Malaysia. The LTGM is an extens impacts ratio growth economic model wheregrowth the in Malaysia. key growth building The model where LTGM blocks include the is an extension key of the Solow-Swan saving, investment building blocks growth and saving, productivity. include modelThe model investment is and prod where developed and maintained the key building by the blocks developed include Macroeconomics saving, investment and and Growth Team productivity. of The the Development model is developed and maintained by the Macroeconomics and Growth Team of the Developm Research and Group at the World maintained by the Bank Group.the Macroeconomics addition In World and Growthto saving, Bank Team of Group. investment Inthe and Development addition productivity, Research to saving, the at Group investment model the productivity, and also takes into World the Bank consideration Group. In addition TFP,tohuman saving, capital, investmentdemographics, and productivity,laborthe force model participation also takes into consideration TFP, human capital, demographics, labor force participation and and a country’s consideration external capital, (FDI environment TFP, human and external demographics,environmentdebt). labor 1 Solving the model requires (FDI force and external participation and debt). data 1 Solving a country’s on model the external three key requires environment parameters: data (FDI andon the labor three key share in production, external debt). Solving the model 41 the depreciation share requires rate in production, of capital and the depreciation data on three the key parameters:initial rate capital of labor the to capital output and in share the ratio, and initial capital production, human capital, the to output ra which are depreciation rate all ofprovided capital andby theinitial which the Penn World are all Table. provided capital to by outputTFPthe growth Penn ratio, and is either World human pulled Table. capital,TFP from growth which the are Penn is all World either provided Table pulled fromorth from other sources. Data fromon other demographic sources. changes Data on and projections demographic (population changes and growth and(population projections working-age gr by the Penn World Table. TFP growth is either pulled from the Penn World Table or from other sources. population ratio) are sourced from the population World ratio) are Bank sourced from Development Human the World BankNetwork. Human Development Data on demographic changes and projections (population growth and working-age population ratio) are Network. 2. from sourced the Worldthe Underlying Bank Human Development simulations Network. in this report is the following base model, reproduced here in an 2. Underlying the simulations in this report is the following base model, re Devadas and abridged manner fromabridged mannerPennings (2019). All from Devadas andsimulations Penningsare run using (2019). an Excel-based All simulations toolkit are run usin Underlying the simulations constructed based on this this report inmodel. The is the model following assumes a base model, specification, Cobb-Douglas reproduced herewhere an public and inthe constructed based on this model. The model assumes a Cobb-Douglas specification abridged manner private from capital Devadas stocks have and Pennings unitary private capital (2019). elasticity stocks substitution. ofhave All simulations unitary Theare run following elasticity is an ofusing theExcel-based production substitution. toolkit function The following is at thetime, prod : constructed based on this model. : The model assumes a Cobb-Douglas specification, where the public and private capital stocks have unitary elasticity of substitution. The following is the production function at time, t: = ( ) − ( ) = ( − ( ) (1) ) (1) Each Each firm takes firm takes technology technology (TFP), Each Atfirm and (TFP), and public takes public services technology St(TFP), services as and as given, given, thatpublic is,thattheseis, these services are are externalities externalities to the to the firm. as given, that is, these are e firm. is is the the private private capital capital stock, stock, h is the t L private t is is effective effective labor, labor, which which can can be be further further decomposed decomposed capital stock, is effective labor, which can be further h into into t , human , human capital decomposed per worker capital per worker and L and , the number t, the number per worker of workers. of workers. and 1 − and β are – β number , the and are ofprivate private workers. capital capital −and and labor and labor income are income shares. private shares. capital and labor i 3. 3.specification The following is the for public The following services : for public services : is the specification can be found here: https://www.worldbank.org/en/research/brief/LTGM. 41 The LTGM with various extensions = [ ] = [ ] A Silver Lining: Productive and Inclusive Aging for Malaysia 137 (2A) (2A) Each firm takes technology (TFP), and public services as given, that is, these are exter is the private capital stock, is effectivelabor, which can be further decomposed into per worker and Each firm , the takes = number technology ( of )− ( workers. (TFP), )− and public services as given, that is, th ( and are private capital and labor incom − is the private capital stock, is = ( ) effective ) labor, which can be further deco per worker and , the number of workers. − and are private capital an Methodological Annex 3. The following Each firm is the takes specification technology (TFP), for public and services public services : as given, that is, the is the Each firm private capital stock, is effective labor, which can be further takes technology (TFP), and public services as given, deco 3. The following is the private is the specification capital stock, for public is effective services labor, : which can be fu = per [ worker ] and , the number of workers. − and are private capital and per worker and , the number of workers. − and are private 3. The = [ following ] The following is the specification for public services St : is the specification for public services : 3. The following is the specification for public services : (2A) = [ ] (2A) = [ ] 4. is the efficient physical public (2A) capital stock —the public capital that is Gt is the efficient physical public production. captures whether public capital capital stock—the public capital that is is actually subject to used congestion in production. (or not). is the u ζ captures whether public capital capital is (more subject 4. technically to congestion is the the efficient elasticity(or not). physical ϕ of is output the usefulness public to efficient capital of public public stock capital capital): — the public capi (more (2A) technically the elasticity of output to efficient production. public capital): captures (2A) whether public capital is subject to congestion (or no = capital (more technically the elasticity of output to efficient public capital): 4. is the efficient physical public capital stock —the public capi (2B) production. 4. = captures is the efficient physical public capital stock —the pu whether public capital is subject to congestion (2B) (or no capital (more production. technically captures the elasticity whether of output public tocapital efficient is public subject to congest capital): (2B) capital (more technically theθ elasticity of output Due to corruption, mismanagement 5. Due to corruption, or pork-barreling, mismanagement only a fraction or pork-barreling, t ≤ 1 of measured public only a to fraction efficient public ≤ of ca = capital isThe useful for production. = The measured is what iscapital stock is what is recorde capital is useful for production. 5. measured Due to capital corruption, stock mismanagement recorded or pork-barreling, in international only a fracti statistical databases, constructed usingcapital statistical databases, the perpetual (2B) constructed inventory using method. the perpetual θ is the inventory average method. efficiency/quality is the average e the public capital stock. is useful Equations (2B) for production. (1), (2A) and t The (2B) measured can be written capital in stock a more conven is wh 6.translated Equation of Equation the public (3) per can capital be stock. translated Equations into per (1), statistical (2A) worker and databases, (2B)terms can be constructed written in using aboth more the conventional perpetual production inventory method. is can be 6. into (3) worker can 6. beEquation terms translated function by (3) as: dividing into can5. per be Due bothworker translated toterms sides corruption, by into dividing by : per dividing mismanagement worker both sides terms sides by by or by : : pork-barreling, dividing both sides only bya fractio : a function as: 6. Equation 6. Equation (3)capital the can (3)be public can translated 5. be capital translatedDuefor stock. tointo corruption, per intoworker Equations per worker mismanagement(1),terms (2A) terms by dividing and by capital (2B) or both dividing can pork-barreling, be bothsides written sidesby is in by : onl − −− is useful production. The measured stock wha [ ( )− ] ≡ ( ≡= ) −− [ = [ ( ) ( ) − ] ≡ ( ] ) (= = [ function ) −− ( ( ) as: − capital ) ( −− ] ) is (constructed useful −− ) −− ( (4) for ) production. The measured(4) (4) (3) capital stock statistical ( − databases, using the perpetual inventory method. is t ) can ≡ = [ − ] ( )−− 6. Equation (3) can 6. (3) be Equation translated (3) into ≡ per be translated worker = [ ( terms into )terms per by dividing worker ]) ( termsboth sides sides by dividing by : :both sides 6. Equation can be translated into the per public worker statistical capital = ( ) ( ) stock. databases, by dividing Equations constructed both −− (1), ((2A) by using and the (2B) perpetual canby beinventory : written in met a )L : (1), (2A) and (2B) can be wri Equation (3) can be translated 6. Equation into (3) per can worker be the terms translated public by capital dividing into per stock. both worker Equations sides terms by by function as: : dividing both sides per by : is is t t per worker where where 6. and is ≡ output Equation is is =output ≡per private [ worker (3) (= per can capital ) − [ be worker ( and translated ) per− ]is and( worker ) ] private into −− ( is and per )private −− − capital worker ] capitalis measured per terms worker per −− by worker dividing public and and capitalboth sides measured per is by measured public public (4)(4) capital is per capital ( ) where ≡ = 1 [ output The LTGM per with worker various ( function extensions and ) canas: is be private found here: capital per https://www.worldbank.org/en/research/brief/LTGM worker and measured (4) public lower worker case). worker (note = (note the lower , the where lower case). worker where is case). where total (note = is output = population, the is , where output lower per , where worker case). per is is worker the total and = is working total population, and is private age-population population, is private capital is −− the capital isper working the workerper working worker age-population and age-population and is is measured is measured publ ,)where is( total )population, the ( )= − ( ) (−− ) (4) working age ≡ = [ ( )− ≡ ] the ( = [ )−− LTGM ] ( −− (4) terms labor force ratio6. ratio and Equation participation is and the (3) is laborrate the can labor be force (labor ratio worker translated force participation force-to-working-age worker and (note participation is the into (note 6. per rate lower the (labor rate worker Equation lower 1 The case). population (labor case). terms force-to-working-age with (3) = various force-to-working-age can by ratio). = be dividing extensions , ( translated where , can both population where be ) force-to-working-agefound is sides population ( ) into total here: is per by ratio). population, total worker ratio). ( :population, https://www.worldbank.org/en/research/b ) is by the dividing is working the ratio). bo worki ag labor force participation rate (labor population where is output perratio worker and ratio and and is the is labor private force capital participation per worker rate and (labor is measured force-to-working-age public capital populationper ratio). where is output per worker and is private is thecapital labor force per worker participation and − rate isworker (labor force-to-working-age measured andpublic capital perpopulation rat where is− output per worker and isprivate capital per is measured public capita −− quation 7. can worker 7.The where then (note worker be above The ≡ yabove the t used (note is = [ output equation lower worker thecalculate to equation case). lower per ( can (note ) worker case). then can the = growth then and be lower ] = ( used be rates) , case). is used where to calculate private ,of where to output = capital calculate is ≡ total isper growth per = total , growth where[ worker population, worker population, rates ( rates ) and from of is output total of is to ] is output the is ( the population, measured per + working )−− working worker per worker publicage-population age-population from is (4) capital the from to working per to + +age-popul where ratio and is output the is the 7. where per labor The workerforceabove and is output participation equation is private per rate 1 The can worker (labor capital LTGM then force-to-working-age with per and be various worker used extensionsis private to and calculate can capital is population be found measured growth per here: worker ratio). rates public and https://www.worldbank.org/en/research/b of output capital ratioper isper measured worker pu fr ratio: and : worker is the worker (note (note labor ratio the lower force and lower 7. case). participation is The 7. the L t = ϱt above above Thelabor ωforce Nt, equation rate where(labor,equation participation Ncan then t is total force-to-working-age can be population, rate then used (labor be used to ω calculate force-to-working-age is to t population the calculate working growth ratio). age-population growth rates population rates of output ofratio). output per worker per wor case). = where is total population, is the working age-population t : worker (note the lower case). 1 The = LTGM with , various where extensions is total can be population, found here: https://www.worldbank.org/e is the working and ratio ϱtand is the labor is the force labor : participation force : participation rate (labor rate (labor force-to-working-age force-to-working-age population population ratio). ratio). from to population where + + (−) 7. is The output above per equation worker ratio (−) can and and (−) −− then is be is the used private labor where to capitalforce calculate is (−) output participation per −− growth −− worker per rates worker and rate of (labor output and is force-to-working-age per measured is private −− worker public capital capital per +per and worker ratio). 7. ] The : above [ + + = [] equation [ + 7. + + + + + + + = [ ] The [ + can ] ] above [then ] [] be equation + + = [ used ] [ [ + + + [ ] [to ] can + calculate + + ][ ] then + + + [] ](note [ + be ] (−) growth used[ ] [ + + ] to ][ rates (5) calculate [ + + ] [ ] of [ + output + ] growth ] + [ per ] rates worker −− (5) of output [ (5) from + −− ] per to worker + from(5) to worker(note the lower case). = worker , where is(−) + the total +lower population, case). + = is the , where working + age-population is + total population, : 7. The above equation The above : equation can can then bethen + = [ used be + + = [to used + + to calculate calculate + + ] ] growth [ growth ] [ [ rates + ]] +rates [ of [ of + ] output output ] [ [ + per per + ] ] [ worker worker + ] [ from from ] [ t to to] t + + 1 : (5) (5 ratio and : is the labor force 7. participation The above (−) equation ratio rate and (labor can is then the force-to-working-age −− labor beused force + to participation calculate population growth rate ratio). (labor rates of force-to-working-age output per worke p + + + + + + + + can be rewritten 8. Equation in (5) terms =can [ of be growth rewritten : ] rates in [ fromterms ] [ of to ] growth [ ] [ rates ] from [ to ] : (5) 8. + Equation (5) can be rewritten in terms of : growth rates from to + : +from to + : (−) (−) + −− + −− 8. Equation [(5) ] can be rewritten in [ terms + of growth rates + + + + + + + + ] + = [ ] 8.+ + =Equation [ then [ + (−) ]be (5) ] can [(5) ] be [ [ be + rewritten + ] [ −− ] in [ + terms ]] [ of + ]output growth [ rates (5) ]to from to + :+ (5) (5) 7. (−) The above equation + = [ + + (−) can 8. ] Equation [ + used ] [ 7. + to ] The can calculate [ + above ] [ rewritten (−) + growth ]equation −− [in rates +terms can ] of then of growth be used per rates worker (5) −− calculate −− from from to to growth + : rates of ou + + ) + ]( + + = [( + = )( [( + + ) )(−)]( ) ( + ]( + + )( +) )( ( + + ) ( ) + )( + ( ) + ( ) + ( + ) ) ) ( + +( + ) ) −− : ,8. +,Equation (5) can be rewritten in terms of,growth rates from to : + + + + + + + ,+ + +,++ = , [( = +[ + , , : + ) (−) + ] ]( , + , + + [ ] [ )( + , +], + [ + + , +] + ) [ ( , + + + ] , + ,+ [ ) ] ( , + + , + ,+ −− ) (5) ( − ,+ + (−) (−) , + , 8. Equation Equation (5) (5) can 8. can be Equation be rewritten + rewritten + (5)]( in can = in terms [( terms be = + [( rewritten of of + growth ) growth ) ]( rates rates + ]( from from + )(tot to + )( +t + 1 + : : ) ( ) + ( + ) ( ) + ( + ) ,+ ) ( in terms of growth rates : from to + : −− + , + , + , ,+ 8. Equation , + , + + , + , + −− ++ ,+ (5) can be rewritten + in terms of growth rates from to + ( + = [( + + )(−) )( + −− + ,+ + + + (−) + , + + ,+ + ) + ( + + + , + + ) ( + + (−) + ,+ ) + , + ) + (6) (6) = [ (−) ] 8. Equation [ ] [ (5) ] can [ + be ] [ rewritten = [ ] in [ terms ] ] of growth [ ] [ rates ] −− [ (5) from [to + ] + ] + : −− [ ] + ,+ + = [( + [( + + )+ (6) = ]( [( (−) ++ , + ) )( (−) + ]( + ) + ( + )( + ) ( ) ( ++ ( + ) −− ) ( ( ++ ) ) (6) ( + ,+ = ) ]( )( + ) ( + ) ) ( + ) + + , , + , , + (6) (6) + + , + , + , + , , + + , + , + , + , , + + ,+ rate of where a where variable the (6) growth the from growth rate to rateof where a + of variable + a is the denoted variable growth = from [( by from rate + to + to) , and (−) + is denoted ]( isis the denoted + growth by by )( , +rate + , and of , the and ) is (the is+ growth the growth rate) rate ( of + the of the ) −− 8. number Equation (5) can be rewritten in terms 8. ofof , + Equation growtha variable (5) rates can from to + , + :is denoted by + + , rates and from is ,the growth from be rewritten to in terms of growth to + , + , + s: number of workers: of workers: where where the ,+ growth the growth rate of rate at+ variable of ais variable from from to + + to is + denoted is denoted by, , by , , and , and is+the is growtthe + g (6) where where (6) the the growthgrowth (6) number raterate ofof a a of variable variable workers: x from from t to to + 1 is denoted denoted byby , + , , andand Γ is is the the growth growth , rate rate + of of the + the number of workers: number (−) number of workers: of workers: (−) −− ( + + ,+ number )( , where ++ + = ++the [( of , += workers: + + growth ( )( = + + + ( ) rate + , , (6) + +of )( , ) a ]( + +)( variable + + + , , + = + )( from , )( ( + + ++)( + , to + ,, + + ++ = ) )( )+ , [( is ( + ) denoted + + + )by )( ,+ + ) ]( ( , + and + ) , , is + + )( the ) growth + , ( + rate + ) of ( ,the+ + ) ,+ ) ( + where the growth rate where the growth of a variable rate from + of a variable to + is ,+ from denoted , to by + + + , is , + , denoted and ,+ is the growth by ,+ , and is the(7) rate of the growth rate o + ++ = + ( = + ( + )( + )( , ++ )( + )( , + + ) ,+ ) number (7) of+ workers: + = ( + ,+ )( + )( +, ) + + , + ,+ number of workers: number ofwhere workers: (7) the growth rate of a variable (7) , , + (7) from to + is denoted by ,+ , and is the grow (6) + = ( + number )( of +workers: (7) (6) )( + ) (7) (7) from equation + ++ + drops (6) + + in drops out drops = the ( from congestion out + + out from equation + from , + + equation )( default , equation + + (6) = drops ( in (6) ,( + (6) + the out = in)( , in,congestion ) the from + the +. congestion )( congestion,+ equation + ) ,+ ,default + (6) default )( default in(+ the =( ) = congestion . +) ). . default ( = ). + + + , where the + growth + drops rate out from of + a variable equation ++ drops + (6) (7)drops out from in where the = from out ( to congestion + equation from the + growth equation is default )( (6) denoted + in rate ( (6) the = by ofin congestion ) a the )( . variable , congestion + + , and default from is default the ( to = growth ) ( + .= ) is. denoted rate of the by ,+ , + ) + number of workers: (7) + number of (7) ,+ workers: ,+ , tput per 9. capita, 9.To To obtain To+ obtain obtain output + drops out from from output output equation per 9. per per capita, capita, equation (4), capita, (6) from ≡from in from equation = equation equation (4), . default (4), Rewriting (4), ≡= this ≡ = equation = .Rewriting .(4), Rewriting in . ≡ this Rewriting this this =equation equation equation . in in in the congestion To obtain output per ( ) . capita, from equation Rewriting this + +terms 9. To drops of out obtain growth from + output equationrates: 9. drops per To 9. (6) outcapita, obtain To in from the obtain output equation congestion from output equation per (6) default capita, perin the (4), capita, (7) ( congestion = from ) ≡ . from equation= default equation .(4), ( Rewriting = ) this (4), . ≡ equation ≡ = = in . Rewriting . Rewritin thi ates: terms terms of growth of + growth + rates: = ( rates: + +terms ,+of )( + )( + + ) = ( + )( + )( + ) growth , rates:+ ,+ + ,+ ,+ + , terms 9. To of growth obtain output rates: per +terms capita, drops out from from equation equation (4), (6) in ≡ the congestion = . Rewriting default (this = ) . terms of + growth of growth rates: rates: equation in = (9. + To ,+ obtain)( ++ + output , + , 9. + = , + ( )( per To = + + ( obtain capita, + , , + + )( ) , output + + + (7) )( from + , per + = )( equation , capita, ( + +)( + + , + (4), ) from )( , + + ) ≡ equation )(= (8) + (4), (7). Rewriting ) ≡ = this (8) equation . (8) Rewriting in (8) this equati terms of growth + ,+ = ( + ,+ )( rates: , + + , )( + , , + + ) ,+ ,+ (8) + + =+ ( = + ( , + + )( + )( ,+ + )( + )( , + + ) ) terms + of + drops growth rates: out from equation 9. To (6) obtain in the, + output congestion , + +per + drops capita, default , out from + = ( from ). equation equation , + (6) in (4), , + the congestion ≡ = default (. = Rewriting). t tput growth, 10. 10. To obtain (8) To isobtain output multiplied +terms output of growth,=with growth ( growth, + population (8) rates: ,+ is(8) )( multiplied is+ multiplied growth:,+ )( with + with population , + ) population growth: growth: (8) To 10. obtain To obtain output ,+ output 10. growth, growth, To obtain (8) is(8) multiplied is output multiplied growth, with with (8) population population is multiplied growth: growth: with population growth: + 10. terms To obtain of growth output rates: growth, (8) ,+ = ( + + , + )( 10. + To = obtain ( , + + )( ,output+ + )( ,+ + ) growth, , +is )((8) multiplied + is multiplied + ) , with with population population growth: (8) growth: (8) = ( +9. To ,+ )( 10. To obtain + + + obtain , + + output = , , + + ) ( ,output = + + = ( per ( + capita, growth, , + + , )( + , + + , + +)( (8) )( is + + , from + multiplied= 9. ) , , ( + + equation To ) + ) obtain with (4), population )( output + ≡ per growth: ) capita, = (9) . Rewriting from equation this (9) (9) equation (4), (9) in≡ = ++ == ( + )( + , + , + ,+ )( + ,+ ) terms of growth rates: , + + , +, terms ( = + of ( growth + , , + + )( ,rates: + )( + , + , +) ,+ ) (9) 10. To obtain output +10. ,+ growth, To = obtain( + (8) is multiplied output )( + growth, + with ) (8) is population multiplied + growth: with population growth: (9) d quantity 11. 11. The of 11. The measured public The measured capital measured quantity (as quantity 11. quantity in The of public international , of + measured ofpublic public capital ,+ capital capital quantity statistical (as (as in (as of international in in databases) international public international capital statistical accumulates statistical (as statistical in international databases) databases) databases) accumulates accumulates statistical accumulates databases) a andard according capital according according + + accumulation to a to ,standard + = to a = (a standard ( standard + + 11. capital identity, + 10. ,The )( capital 11. + To capital )(accumulation + obtain measured with The = + accumulation ( accumulation the measured + , output +) next )( quantity + period’s )( growth, identity, quantity + identity, +identity, , + of ) , (8) with public stock + of with ) = is with ( the public multiplied coming the capital+ next the next capital , next + period’s (as from )( period’s with in period’s + (as the population international instock stock , + )( stock international coming coming + growth: (9) coming (8) statistical , from +from )statistical the from the databases) the databas (9) 11. The measured ,+ according , quantity + , + to of public a , + ,+ capital standard capital (as in , accumulation + international is identity, statistical databases) accumulates stock with the next period’s comin s undepreciated previous previous previous period’s stock, period’s period’s ( undepreciated − undepreciated undepreciated according ) according(where stock, to stock, a stock, standard to ( is a − the ( ( standard − −public ) capital) ) (where capital capital (where accumulation (where depreciation accumulation is the the is public the public identity, publicrate) capital identity,capital with capital depreciation the depreciation with depreciationnextthe nextperiod’s rate) rate) period’s rate) stock stock comi according to a standard previous capital period’s + . accumulation ,+ undepreciated = ( + identity, ,+ stock, )( with the + ( ,− +next ) ) period’s (where stock coming is the from public thecapital deprec 11. 10. The To138 and obtain .measured new output public11. quantity The growth, investment, previousmeasured of previous public (8) period’s is multiplied capital period’s 10. undepreciated ((as To −undepreciated with obtain in population international output stock, stock, growth, A ( growth: Silver statistical − ( Lining: (8) ) is multiplied (where databases) (where is with accumulates the is population public the public capital growth: capital depre d . quantity of public capital (as in international statistical databases) accumul Productive and Inclusive Aging for Malaysia nvestment, and new and new previous public public investment, period’s investment, undepreciated and new . public stock, investment, ) . (where is the public − ) capital depreciation rate) according to a standard according = ( − capital and 11. to) new and a The accumulation standard + public new measured public capital investment, identity, investment, quantity accumulation with .of public the . ( + next identity, capital period’s with (as + stock the ininternational next coming (10) period’s from stock statistical databases the coming from + and new public investment, . ) − previous + = ( , ++ = − = ( ( − ) + ) + , )( + + ,+ ) + + (10) , = ,+ )( ,+(10) ) (10) (9) period’s + + undepreciated previous period’s + stock, = ( undepreciated ( − − ) ) + stock, (where ( − is ) the public (where capital is depreciation the public capital rate) depreciation (10) com r = ( − ) according + + to = a (= standard − ( ) − capital ) + + accumulation identity, with (10) the next period’s stock (10) (1 9. To obtain9. output To per capita, obtain output per from capita, equation from (4), equation ≡ (4),= ≡ .Rewriting = this eq . Re terms terms of growth of growth terms rates: rates: 9. Torates: of growth obtain output per capita, from equation (4), ≡ = . Rewr terms of growth terms rates: of growth rates: terms of growth rates: + + , + = = ( ( + + + ,+ )( , + )( =+( + + ,+ , )( + , +)( )( + ++ + ,)++ , ))( + ,+ ) (8)(8) ,+ , + ,+ = ( + ++ )( ,+ + , = ( + , + )( , + + )( +, + + , ) )( + ,+ ) (8) + ,+= ( + )( + ,+ )( + ) ,+ ,+ ,+ 10. 10. ToTo obtainobtain 10. output output To growth, obtain growth, output (8) (8) is multiplied growth, is multiplied (8)with with is multiplied population population with growth: growth: population Methodological growth: Annex 10. To obtain10. output To growth, obtain output(8) is growth, multiplied (8) with is population multiplied growth: with population growth: 10. obtain output To ) growth, (8) is multiplied with population growth: + + ,+ + , = = ( ( + + + ,+ , , +)( + )( = + + ( + ,+ ) , ++ , )( + , + ) (9)(9) + ,+ = ( + + ,+ ,+ )( = + ( + , + ) + )( + ,+ ) (9) + = ( + )( + , ) ,+ ,+ ,+ 11. 11. The The The measured measured measured quantity quantity 11. quantity The of measured public of capitalof public public quantity capital (as capital in of (aspublic(as international in in international capital international (as in statistical statistical international statistical databases) databases) statistical databases) accumulates accumulates databases) ac accumulates 11. The measured11. The quantity measured of public quantity capital of (as in capital public international (as in statistical databases) international accu statistical da according according according a to toto a a standard according standard standard to 11. capital capital capital a accumulation The standard accumulation measured accumulationcapital quantity identity, accumulation identity,identity, of with public with with the the identity, capital next thenext next with (as period’s period’s period’s instock the stock next international stock period’s coming coming coming statistical from from stock from thethe comin data according to a standard according capital a standard tostandard accumulation capital identity, accumulation with the identity, next period’s with the stock next coming period’s previous the previous previous period’s period’s according undepreciated previous period’s period’s stock, undepreciated undepreciated to a stock, undepreciated ( ( − − capital )stock, ) (where (where (where accumulation ( − ) is is is (where thethe the identity, public public public capital capital capital iswith the the depreciation public depreciation depreciation next period’s capital stos rate) deprec rate) previous period’s previous undepreciated period’s stock, undepreciated ( − ) stock, (where ( − ) is the public (where capital is the depreciat public cap and and rate) new new and new public public and new previous investment, public investment, public investment, . period’s undepreciated . . stock, ( − ) (where is the public capita and new public and investment, new public . investment, . and new public investment, . + += = ( ( − − ) ) = + + ( − ) + (10) (10) (10) ++ ( − ) = =+ ( − ) + (10) (10) + = ( − ) + + 12. 12. The The gross gross growth 12. The growth rate gross rate of measured growth rate of measured public of public capital measured capital (not public (not perper worker) capital worker) is: is: (not per worker) is: The 12.rate gross growth 12. The rate gross measured ofgrowth ratepublic of capital measured (not per public worker) capital is: (not per worker) The gross growth of measured 12. The public gross capital growth (not rate per of worker) measured is: public capital (not per worker) is: is: ⁄ ⁄ ⁄ − ) + ) / + / = ( = ( − / + )+ = ( − ⁄ ⁄ (11) (11) (11) + / + + = ( ⁄ − ⁄ + / ) + = ( − ⁄ ⁄ ) + (11) + / = ( − ⁄ ) + ⁄ (11) ⁄ 13.13. TheThe growth growth 13. rate rateThe of measured growth of measuredrate public of public capital measured capital perpublic per worker, worker, capital whichwhichperenters enters worker, equation which equation (6), enters (6), equation is: is: (6), is: 13. The growth rate of The growth measured rate 13. The public ofgrowth measured capital rate per public of worker, measuredwhich capital per worker, public enters capital equationwhich per (6), enters equation worker, is: which (6), is: enters equat measured 13. The growth rate of ⁄ ⁄ public capital per worker, which enters equation ⁄ (− )+ )+ (− (− )+ + + + + ⁄ ⁄(− )+ ⁄ ⁄ (− ⁄ + + ≡ ≡ + + + / / = = ≡ / = )+ )+ ⁄ (12) (12) (12) ,+ ,+ + ,+ (+ ⁄ (− ⁄ (12) (12) ≡ / = + +)(+ ,+ (++ ,+ )(+ ,+ , ≡ ,+ + (+ )(+ ,+ )(+ /,+ ,++) = ),+ )(+,+ )(+ ,+ ) ⁄ + ,+ +≡ + (+ / + ,+ = )(+ (+ ,+ )(+ ,+ )(+ ,+ ) ,+ )(+ ,+ ) (+,+ )(+,+ )(+,+ ) 14. 14.14. 14.The The 14.The The 14. The stock stock The stock stock The stock stock of of of stock of of efficiency-adjusted efficiency-adjusted efficiency-adjusted efficiency-adjusted of efficiency-adjusted of efficiency-adjusted efficiency-adjusted 14. 14. The The 14. stock stock public The public public public of public of public stock capitalcapital public capital efficiency-adjusted efficiency-adjusted capital of efficiency-adjusted capital capital (which (which capital (which (which (which (which is is(whichis is is actually actually actually actually is public public actually actually is actually used used used public capital capital used used used in in inincapital in production)production) production) used production) production) (which in (which in is production) (which production) is actually actually evolves evolves evolves is evolves actually evolves evolves used used based based evolves based based ininused based basedbased production production in pro 14. The stock of 14. efficiency-adjusted The stock of efficiency-adjusted public capital (which public is capital actually (which used is in actually production) used in evolves production) based evolv 14. on on on the 14. onon The the The the on thethe the stock previous previous stock the previous previous previous previous of ofprevious efficiency-adjusted period’s period’s efficiency-adjusted period’s period’s period’s period’s on period’s on efficiency efficiency the efficiency the efficiency-adjusted on efficiency efficiency efficiency previous previous the public-adjusted public -adjusted -adjusted previous -adjusted -adjusted capital capital period’s period’s -adjusted undepreciated (which undepreciated undepreciated period’s (which undepreciatedundepreciatedefficiency undepreciated efficiency undepreciatedis is efficiency actually actually -adjusted -adjusted stock stockstock stockused stock used stock -adjusted and and and stock in in and production) and undepreciated andproduction) undepreciatedand undepreciated efficiency-adjusted efficiency-adjusted efficiency-adjusted efficiency-adjusted evolves efficiency-adjusted efficiency-adjusted efficiency-adjusted evolves stockstock based new new new based and stock new and new new and new ef efficienc efficiency on the previous on period’sthe previous efficiency period’s -adjusted efficiency undepreciated -adjusted undepreciated stock and stock efficiency-adjusted and efficiency-adjus new on on the investmentthe investment investment investment investment previous previous . . period’s period’s . . . investment efficiency efficiency investment -adjusted -adjusted . . . undepreciated undepreciated stock stock and and efficiency-adjusted efficiency-adjusted new new . investment investment investment . investment investment investment . . . ) + = ( − ) = ( − (13A)(13A) (13A) (13 + (13A) = = = ( ( (= − − − ( = − ) ( ) ) − + + + ) + ) + = = ( ( − − ) + (13A) (13A) (13A) (13A) (13 + ) + + + + + + + + = + ( − ) ++ = ( + − + ) (13A) + + + = =( (− − ) ) + + (13A) (13A) 15. 15.15. 15.θ isis is 15. the isthe the is the average average average is average the efficiency efficiency average efficiency efficiency efficiency of of of 15. existing existing of existing existing of is public existing the public public average public capital capital capital public capital efficiency (rather (rather (rather capital (rather than than of than (rather existing than thethe the than the efficiency efficiency public efficiency efficiency the of efficiencycapital ofof new new of new (rather newinvestment). investment). investment). of new investment). than investment). the efficien 15. the is the average average efficiency efficiency 15. 15. is of is the existing the of existing average average public public efficiency efficiency capital capital of (rather of existing (rather existing than public than public the the efficiency capital efficiency capital (rather of (rather new of new than than investment). investment). the the efficiency of newof efficiency of ne n 15. t is the average 15. efficiency is the average of existing efficiency public of existing capital (ratherpublic capital than the (rather efficiency than of the new efficiency investment). inve 15. 15. Substituting Substituting Substituting Substituting Substituting is the is Substituting the average average = = = = efficiency = efficiency into into into into Equation Equation of Equation into of existing Equation existing Substituting Equation 13A 13A (13A) public 13A public and and and 13A and = rearranging capital capital rearranging rearranging and rearranging (rather rearranging (rather into as as as Equation 13B, asthan 13B, (13B), than 13B, asone one the onethe13B, 13A one efficiency can efficiency can can and can see one see see can of of rearranging see the thethe seenew the new the investment). investment). evolves evolves evolves as evolves 13B, as as as evolvesa one aas a a can as a se Substituting = into Substituting into Equation Equation 13A = 13A and and rearranging into rearranging Equation 13A as as 13B, 13A 13B, one and one can rearranging can see see the the as 13B, evolves one evolves as as can a see a thethe Substituting = into Equation 13A and rearranging as 13B, one can see + + + Substituting = = Substituting = into Equation and rearranging as 13B, one + can + + +see the evo into Equation 13A and rearranging as 13B, one can see the evolves as a + Substituting Substituting weighted weighted weighted weighted weighted weighted weighted averageaverage average average average = average of average of of the the of of the of the quality the thequality of into quality into weighted quality quality the weighted quality Equation of Equation of quality of existingweighted existing of of existing of average existing existing average of existing 13A 13A public existing public average public of public of and public publicand the the capitalcapital public rearranging capital rearranging capital quality quality of capital capital the capital ,of θ quality and ,of , and ,existing and and ,, and the existing as and as the thethe , of 13B, quality andthe the13B, existing quality quality quality public quality the publicone quality one of of quality of of capital can public new can of new capital new new of newsee investment newsee of capital investment investment investment new , the , the andand investment investment investment the the , + + and + quality evolves . quality evolves .. . the quality . . of . of as as new new aa of investme investmen new in weighted average weighted of the quality average of of the quality existing public of capital existing t public , and the capital quality of, and new the quality of investment new investment . weighted weighted average average of of (− the (− the (− ) quality ) quality ) ) of of existing existing publicpublic capital capital (− , ,and ) the and the quality quality of ofnew new investment investment . . . (− = = = = = (− (− (− ) ) ) + + + + +(− = ) = (− (− ) ) + (−) + (13B) (13B) (13B) (13B) (13B)(13B) (13B) (13 = + = + + (13B) (13 + + + + (− = + + (− ) ) + + + (− (− (− (− (− ) ) ) ) (−) + + )+ (− + (− + + (− ) ) ) (−+ + ) + + + + + += = = (− (− (− (− (− ) ) ) ) + + + + + + + (− (− (− (− (− ) ) ) ) ) (− + (− + + + ) ) + + (− ) (− (− + ) ) + + (13B) (13B)(13B) (− ) ) + + + (− (− ) + ) + + 16. 16.16. 16.As As As 16. 16.As such, such, such, Assuch, such, As such,the the such, the the quality/efficiency the quality/efficiency quality/efficiency the quality/efficiency quality/efficiency quality/efficiency quality/efficiency 16. 16. As As 16. such,such, of of of As the the the of the the of such, the stock the stock of stock stock quality/efficiency stock the quality/efficiencystock the of of stock of of quality/efficiency public of public public of public public publicofthe capital capital public capital capital capital capital of ofthe only capital the only onlyonly stock onlyof stock onlychanges changesthe only changes changes changes ofof changesstock public public when changes when when of when when when public capital the capital the when the the the quality the quality capital quality only quality the quality only quality ofofonly new quality of changes changes of new of new of changes new of new when when new th the w 16. As such, the 16. quality/efficiency As such, the quality/efficiency of the stock of public of stock capital of only public changes capital when only the changes quality whenof the new qualit 16. 16. As As such, investment investment investment new investment such, investment investment investment projectsthe projectsthe projects projects projectsquality/efficiency quality/efficiency projectsis projectsis different isdifferent is different is different is isdifferent investment different investment frominvestment different from from ofof from from the that projects the that projects from that from stock that that of stockof of projects the that the that is ofofthe is of of the existing different the different public of public existing of existing the is existing theexisting existing capital different capital public existing from public public from public public that only from only capital that capital public public capital of changes capital ofthe capital that changesthe stock: stock: capital stock: existing existingof stock: stock: when the when stock: ≠ existing ≠public ≠ the the≠ public . .Using quality quality Using ≠ . public Using capital capital of equationof new capital new equation ..equation Using Using stock: stock: equation stock: ≠ . Using .stock: Using equation equation ≠ ≠ ≠ investment projects investment is different projects from is that different of the fromexisting that of public the existing capital stock: public capital ≠ . Using equation . Using investment investment (13B), (13B), (13B),the (13B), the (13B), equation (13B), the the growth growth the projects growth projects growth the (13B), growth in in growth the quality in in in quality is growth qualityis different quality quality different in (13B), which which quality in which (13B), which quality which the the from enters from (13B), enters which enters growth growthenters which enters that that the equation equation entersequation in of growth equation enters in equation of quality the the equation quality (6) (6) equation (6)existing inexisting can(6) can (6) which quality can which becan (6) canbe be written (6) public be can entersbepublic which written enters can written written be written be capital as capital enters equationas writtenwritten equationas asfollows:as follows: follows: stock: stock: equation follows: as follows:as (6)(6) follows: follows: can can be (6) be ≠ ≠ writtencan written . be .Using Using written asas follows: as equation equation follows: follows: (13B), the growth (13B), in quality the growthwhich in enters quality equation which (6) enters can equation be written (6) as can follows: be written as follows: (13B), (13B), the the growth growth in in quality quality which which enters enters equation equation (6) (6) can can be be written written as as follows: follows: + + ⁄ ⁄⁄ ⁄ ⁄ + ⁄ ⁄ + + + + + ≡ ≡≡ + ≡ + = + = = ≡ [( + = [( [( − [( − = − − [( ) ) +) + − + ) + ) + + ] ⁄ ⁄ /( ] /( ] /( ≡ ] /( / ] / /( = / [( ) ) − / ⁄ ) ) ) + ⁄ ]/( (14) (14) ) / (14) (14) ) / ) )(14) + ++ +, ,+ + ,+ , + + , , + ≡ , + + = + [( [( − + + ≡ ) , + + , + + = ≡ ⁄ ≡ ⁄ [( ⁄ , ⁄ + − = = ]⁄ /( [( + [( + ) + + − + + +− / + ) ) + +) ]/( ] ] /( /( / ⁄ + ) + / / + (14) (14) (14) (14 (14 + ++ ,+, ,+ + ≡ ≡ ≡ + + = = = [( [( − − − , + ) ) + )+ + ⁄ ] ⁄ ⁄ /( ] /( ] /( + + + / / / ) ) ) ⁄ ⁄ ⁄ + (14) (14) (14) ⁄ ⁄ ⁄ 17.17.17. 17. The 17. TheThe The The 17. quantity The quantity quantity quantity The of of quantity quantity quantity of of of private private private of private of private private 17. 17. capital privateThe The capital 17. capital capital capital capital quantity The follows follows follows capital follows quantity follows quantity the follows follows of thethe the of the same same same the private private same of same the same private accumulation accumulation capitalsame accumulation accumulation capital accumulation capital accumulation accumulationfollows follows follows process process process the process theprocess process process same same as as the as as same as public public public public as accumulation accumulation public as public accumulation capital. capital. capital. public capital. capital. But But But with capital. But capital. process process But with with But with process with But with as withasca public aspublic p c 17. The quantity17. of The private quantitycapital of private follows capital the same follows accumulation the same accumulation process as public process capital. as public But with capital. 17. 17. as as TheThe the as as as the the quantity quantity the private private private as the private ofof capital capital private private private capital depreciation capital capital capital depreciation depreciation follows as follows depreciation the rate, the private rate, the rate, and and same same and rate, capital asand asaccumulation accumulation as private private depreciation private as investment. investment. investment. private process process rate, The investment. as as The public andpublic The growth growth The capital. ascapital. growthprivate rate rate rate growth ofBut of investment. But ofwith with private private private rate of private The capital depreciation rate, and as private investment. as the the private private capital capital as the as as theprivate depreciation the depreciation private private capital capital rate, capital rate, and and depreciation depreciation depreciation as private rate, private andrate, investment. rate, and investment. and as The private as as The private private Thegrowth growth growth investment. investment. investment. rate rate of of rate private of private The Thegrowth private The growth growt rate o capital capital as as capital as the the per per the capital private private perprivate worker worker worker per iscapital is capital as worker capital asis asis follows: follows: follows: as depreciation depreciation capital depreciation follows: rate, rate, per rate, and worker and and is as as as as private private follows: private investment. investment. investment. The The The growth growth growth rate rate rate of ofof private private private capital capital capital per worker per per worker is capital worker isas is follows: as as capital follows: capital per follows: per per worker worker worker is as isis follows:as as follows: follows: capital capital capital per per per worker worker worker isisas isasas follows: follows: follows: ⁄ ⁄ ⁄ ⁄ ⁄ (− (− (− )+ )+ (− )+ ⁄ )+ ⁄ ⁄ ⁄ (− ⁄ )+ (− (− (− )+ ⁄ )+ )+ ⁄ ⁄ ⁄ ⁄ (− )+(− (− )+ ⁄ )+ ⁄ + ,+ = (15) ⁄ + + + + + + = = = = = = (− (− (− )+ )+ + )+ + ⁄ ⁄ ⁄ ⁄⁄ = = ⁄ ⁄ ⁄ (15)(15) (15) (15) (15) (15) (15 (15 , + , + , + , + , ,+ + , (+ + (+ (+ + (+ )(+ )(+ )(+ )(+ = )(+ ⁄ )(+ )(+ ,, + + ) ) )(+ ) ) (+ )(+ )(+,+ ) (15) ++ == + ,+ = ,+ ,+ (+ ,+ (+ (+ ,+ ,+ )(+ ,+ ,+ ,+ , ,+ )(+ )(+ + ,+ ⁄ ,+ )(+ ,+ ,+ ,+ )(+ )(+ ,+ (+,+ ,+ ,+ ,+ ))(+ (+ (+ ,+ ,+ ,+ )) ,+ ,+ )(+ )(+ ,+ ,+ )(+ ,+ ,+ ,+ ,+ )(+ )(+ ,+ ) ,+ )) ,+ (15) (15) (15) ,+ (+ (+ (+ ,+ ,+ )(+ )(+ )(+ ,+ ,+ )(+ )(+)(+ ,+ ) ) ,+ ) ,+ ,+ ,+ 18. 18. 18. ToTo To better better 18. better Tounderstand understand better understand and and and simplifysimplify18. simplify To the better the the analysis analysis understand analysis of of the theof the drivers and drivers driverssimplify of of growth,of growth, the growth, analysis it it is is ithelpful helpfulis of helpful the to to drivers take to take take aa log- of log- a growth, a log- it 18. 18. To 18. ToTo better To better better better understand understand 18.understand understand understand To18. 18. better and and To and To and simplify and simplify better better understandsimplify simplify simplify the understand understand the the the analysis and analysis the analysis analysis simplify analysis and and of of of of simplify thethe simplify the the the of drivers drivers the drivers analysis drivers thedrivers the ofof of growth, analysis of analysis of growth, growth, growth, of growth, the of drivers ofit itis the it isthe it is is helpful is itdrivers drivers helpful helpful of helpful is growth,helpful toto of to of take to take take growth, growth,take it to isatake a a log- helpful log- log- it it a is log- is helpfu help to ta 18. linear linear linear 18. linearTo better approximation Toapproximation better linear approximation approximation understand understand approximation of of equationofand equation and equation of simplifylinear simplify equation (6). (6). approximation the (6). the Specifically, Specifically, (6).analysis analysis Specifically, Specifically, of of equations equations the of the equations equation drivers drivers equations (12), (12), of (12), of (14) (6). growth, (14) growth, (12), (14) Specifically, and and (14) and it (15) it (15) is and helpful (15) arehelpful are (15)equations are substituted substituted toareto take substituted take (12), substituted a aintolog-log- into (14)into intoand (1 linear linear log-linear approximation approximation approximation linearof linear approximation of linearof equation equation equation of approximation approximation equation (6). (6). (6). of (6). Specifically, Specifically, equation Specifically, Specifically, of ofequation equation equations (6). equations equations equations Specifically, (6). (6). (12), (12), Specifically, (12), Specifically, (12), (14) equations (14) (14) (14) and and and and (15) equations equations (15) (15) (15) (12), are are are are (14) substituted substituted (12), (12), substituted substituted and (14)(14) (15) into and into and are into into (15) (15) substit are are linear linear equation equation equation approximation approximation equation (6). (6). equation equation (6). (6). Then, (6).Then, Then, Then, (6). Then, taking taking Then, taking taking taking of of logs logs takingequation logs equation equation equation logs and logs and and equation and using logs and using (6).using (6). (6). and (6). using using the Then, the Then, using the (6). Specifically, the Specifically, approximation approximation the taking taking Then, approximationapproximation the approximation approximation logs taking logs equations equations and and logs ( ( usingusing ( + and +( + (12), (12), ) ( ) approximation the + the using ≈ ( ) + ) ) ≈ ≈ (14) + (14) approximation approximation ≈(for ) ≈ (for the (for and ≈ (for approximation (for and smallsmall small (15) small (for(15) small g) g) small g) one are one g) are one ( g) ( one can one substituted can +g)substituted can arrive one )can can ( arrive arrive ≈ + arrive can arrive ) at at arrive (for at ≈ into the at into the the at small the theg)(for at the g) sma one equation equation (6).(6). Then, Then, equation taking taking (6). logs logs Then, andand using taking using the the logs approximation and approximation using the ( + ) ≈ (for (for ( small small + ) g) g) ≈ one one + )(for cancan ≈ small arrive arrive (for g) at at small one the can arr one equation following: following: following: equationfollowing: following: (6). following: (6). Then, Then, taking taking logslogs following: following: and and following: using using the the approximation approximation ( ( + + ) ) ≈ ≈ (for(for small small g) g) one one can can arrive arrive at at the the following: following: following: the following: following: ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄⁄ [ ≈ ≈ ≈ ≈ ≈ ≈ + + + (+ ( + ( ( ( + + ( + + + + ≈ ≈ + + + + + + ≈ + + + ) ) − ) − ( ( − ( )) ( −+ ( − − −)( −( ()( − + )( + − )( − ( )( − )( + ) )( + ) + + ) + + ))+[ + [ ) + [ ) )+ − [ − [ ( ([ − ⁄ − ) − ⁄ )( )( ( − − − − − − )( ] ] ) ] ) + − ] + ] ] [ [ ) + ,+ ,+ ,+, + , ,++ , ,+ + , + , , + , , + ++ , + , + , + , ≈ +, , , + ++ , , , + , + , + + , + , , ++ , , + + + + ( , , , + , + + , + + , + , , , , + +++ + , + , , + + + , , + , , , , + ++ + , + , ) , + ++ , − + ( , + ,+ ,+− )( , + ⁄ ⁄ ⁄ ) ⁄ + , , +⁄ +[ , + − ] ≈ ≈ ,+, + ≈ , , + + + + + ( ( ( , + , +, + + + + , + , , + + + + + , , + + , + ) ) − )− − ( ( ( , − − + − )( )( )( , , + , + +) ) + )+ + [ [ [ , + ⁄ ⁄ − − − ] ] ⁄ ] ⁄ ⁄ , + , + ⁄ ⁄ ⁄ ⁄ , ⁄ + , + , + ⁄ , + ⁄ ⁄ ⁄ (16) +( +( +( +( − +(− − − +( − − − ) − − ) − ) ) ( (− ) ( ( ) ( ⁄ −− ⁄ ( − − − ) ) ) +( +( − ) ) )− − +( − − ⁄− ) ( ) ( − ⁄ ) ⁄ ( − − ) − ) (16)(16) (16) (16) (16) (16) ) ⁄ +( ⁄ ⁄+( ⁄ ⁄ ⁄ − ⁄ − ) ( − ) ⁄ (16 −− −− ) (( ( − ) (16) ⁄ ⁄ ⁄ ⁄ ⁄ +( +( − − ) ) − − ) ) ⁄ (16)(16) ⁄ ⁄⁄ 19. 19. 19. In In terms 19.In terms terms Inof of terms of implementationimplementation implementation of In implementation of 19. of the of the In theterms LTGM, LTGM, of LTGM, the of the the LTGM, implementation the future future future the growth growth growth of rates rates the rates of of LTGM, the of the the laborlabor the labor future force force force growth participation participation participation rates of the la 19. 19. 19. In In terms In terms terms of of implementation of 19. 19. In implementation 19. implementationterms In terms of terms of of the of implementation of the of the LTGM,implementation LTGM, LTGM, implementation the the the future of future thefuture future growth of LTGM, ofthe growth the growth growth LTGM, therates LTGM, rates future rates rates of thethe of the of future future growth the the of labor the labor labor labor growth growth force rates force force of force participation ratesrates the participation participation participation of of labor the the force labor labor forc parti for 19. rate rate 19. rate rate ( In (In terms rate ( terms ( , ),( ), of the the of ), implementation the working working ), working implementation the age working age age to to rate of to population of age the population the ( to LTGM, population LTGM, population ), ratio ratiothe the ratio (working future ( future ratio ( growth ), ( ), age population growth ), population to population rates population ), rates of ( ( of population the ( the )ratio labor ( ) laborand and ) ( force and pure pure force) and pure participation TFP ), TFP participation pure population TFP( ( TFP ( ), ), ( ( ), ), ) rate (, ), the ),+the working working rate rate age (), ( age to topopulation), ),the theworking population working ratio ratio to age (, age ( to to ),population populationpopulation ),+population ratio (, ratio ( ( (+ ) and ) ), ), and pure population population pure TFP TFP (, +( ( ( +), ) )), and and pu pu ,+ , + ,+ rate ( + , + , + + , ,+ + ), , the rate ( working ,+ age the to, , + + working population age ratio population ( + , + + , , , , + ++ , ), ratio ( population + , ( ,+ ), + , population , ,++ , , + , + ) , + and + pure ( TFP , ) ( +and , + ,+, , , , + , + pure + ), , + TFP rate arerate are are ( exogenously ( exogenously exogenously ), ), thethe working working determined. determined. determined. age age to to are The population population exogenously growth ratio rate ratio determined. of ( ( ), ), population population The growth ( ( rate of) ) and and measured purepure TFP TFP public ( ( capital ), ), per work are exogenously determined. The The growth growth rate rate of of of measured measured measured public public public capital capitalcapital per per per worker worker worker ( ( ( ), ) is is ) given givenis given by by by ,+ The growth rate of , measured + public capital , per+ worker ( ) , is +given by are A exogenously are exogenously ,+ , + determined. are are determined. are exogenously exogenously exogenously The The growth growth determined. determined. determined. rate rate of measured The measured growth, , The + The+ public growth growth rate public of capital rate capital rate of measured per of , measured per measured + worker ,+ public worker ( public public capital ( capital capital per ) is worker ) given is per , per given+ , + worker ( by worker by ( ( ,+) is , + , + + ,+ are Silver exogenously Lining: Productive and Inclusive determined. Aging The for Malaysia growth rate of measured public capital per worker ( ,+ , ,+ + ) is 139 given by are are equation equation exogenously exogenously equation (12), equation (12), (12), usingusing (12), determined. determined. usingthe the the growth usinggrowth growth the rateThe growthThe rate equation rate of growth ofgrowth the theof rate the public rate public rate (12), of public the of of using capital capital publicmeasured measured capital the stockstock capital growth stock public public (equation (equation stock rate (equationcapital capital of (11) per the (11) (equation per (11) aspublic as worker an worker an (11) as ancapital intermediate intermediate as ( ( intermediate an stock intermediate ,+ ) step. ) (equation step. is is given step. given Private Privatestep.Private by by (11) Private as an in equation equation (12), (12), using using the equation growth the equation growth equation (12), using rate rate (12),(12),of the of the using the using public growth public thethe capital growth growthcapital rate of the stock rate rate stock (equation of public of the the (equation public public (11) capital(11) capital (11) capital stock as an as an stock (equation intermediate stock (equation intermediate (equation ,, (11) asstep. + + step. (11) (11) step. Private an intermediate as asPrivate an an intermed intermedi ste equationequation equation capital capital capital per per capital (12), per (12), worker (12), worker using worker per usingusing growth growth worker the the the growth growth growth growth ( ( growth ( raterate ) ( rate )capital is is of ) of given of given the is the the per given ) public by is public by public worker by equation equation given capital equation capital by capital growth (15). equation stock (15). stock stock (15). The () The (equation (15). (equation The (equation growth growth )by growth The is given (11) (11) rate rate growth rate of as of by as theasan the of ratean an equation intermediate the intermediate intermediate efficiency efficiency of efficiency the (15). of efficiency of The public step. of public step. growth publicof Private Private capital capitalPrivate public capital rate capital of the e capital capital per per worker worker growth capital growth ( capital capital per ( worker , + , + , + per per , ) + , ,+ + is worker ) given worker growth ,+is given bygrowth ( equation by growth equation( ( ) is (15). given ,+ ,+ (15). ) The is is by The given given growth ,+ equation growth by rate rate equation equation (15). of the Theof (15). (15).the efficiency growth efficiency TheThe growth growth rate of public ofof public rate the rate capital of of efficiency capital the the efficienc efficiency of pub capital capital per per isworker worker growth growth ( ) usingis given by equation ,+ (15). (15). The growth rate of the efficiency of public capital (capital ( ( ,+ ,+ )) , ( is+per is )given given worker given ) is by by growth by equation given by(equation equation ( equation (14)(14) ,+ (14) ,+ ,+ using )() using is is (14) given given , the + the ) is using by theby growth growth equation given equation growth the by rate rate growth equation rate of (15). of the of the rate TheThethe public (14) public of growth growth public the using capital rate rate capital capital public the of stock of stock the growth capital the stock efficiency from efficiency from stock rate fromequation equation of from ofthe equationof public publicpublic (11) equation (11) capital capital (11) as as capital an (11) as an an as stock an f equation (6). Then, linear taking approximation logs and using of( equation (6). ( Specifically,( equations (12), (14) and equation equation equation equation (6). equation (6). (6). (6).Then, Then, (6). Then, equation Then, Then, taking taking taking taking taking(6). logs logs logs logs Then, and and logs and and using taking using and using using using the the logs thethe approximation approximation the approximation and approximation using approximation the the ( (( approximation + + approximation + + ( )) ≈) ) + ≈ ≈ ≈ ) (for (for (for ≈ (for small (for small + small small ) g) small g) +≈ one g)g) one ) one one g) (for can ≈ can one can can arrive(for small arrive can arrive arrive small g) at arrive at at one at the the g)the at the can one the can arrive ar a equation (6). Then, taking logs and using the approximation ( + ) ≈ (for sm following: following: following: following: following: following: following: following: ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ , ,+ , + , ≈ + ≈ ≈ +, ≈ + ≈ , ,+ + , ,++ + + + ,+ + ( ( ( ,+( + ≈( , , ,+ + , , + + + +≈ , ++ ,+ + + + , , , + + ( , , + + + + , ++ + ( + ,+ , ,+ + ) , + , + +)− ,+ )− , ) + − − (+ ( , ) + − ( ( − − + − − ( )( , + )( )( − )( + )( ,+ ,+ ) , , ,+ , − ) + + ) ++ ( + ) , ) +)+ + − ) ( )( [ [ + [ [ − [ )( , + ) + , + − −) [ − +− ] − ] [ ] ] ] − ] − Methodological Annex ,+ ≈ ,+ + (,+ + ,+ + , ⁄ ⁄ + ⁄ ) ⁄ − ⁄ ( − )( ⁄ , + ) ⁄ + [ ⁄ ⁄ ⁄⁄ ⁄ ⁄ ⁄ +( +( +( +( −− − − +(− − ) − − ) ) − ( ) ( (( +( ) ( − −+( − −− − − ) − ) ) )− ) ( )) ( − − ) ) ⁄ (16) (16) (16) (16)(16) (16) (1 ⁄ ⁄ ⁄⁄ ⁄ − − ) ( +( ⁄ ⁄ − ) ⁄ 19. 19. 19.19. In In 19. In In terms terms Interms terms termsof of of 19. of implementation of In19. implementation implementation implementation implementation terms implementation In terms of implementationof of of of of of implementation the the the the of the LTGM, the LTGM, LTGM, LTGM, LTGM, LTGM, the the the thethe of future the future of future future future the LTGM, future growth growth growth LTGM, growth growth the growth the rates future rates rates rates rates of future rates of of growth of the ofthe the thethe of labor growth labor the labor labor labor rates labor force force force rates force force offorce theparticipation participation of participation participation participation the labor labor participation forceforce participa par rate ( ), the rate working ( age ), to the 19. working populationIn terms age ratio of to implementation ( population ), ratio population of(the ( LTGM, ), population ) the future and pure (TFP growth ( ) and rates ), pure of the TFP rate rate rate (rate(( rate ( ), ,+), the), rate the the ),, working working the the ( working working working ,+ ), age age age the ,+to age to age to working population population to population to population age population ratio to ratio ratio ratio population ( ( ratio ( ( ), ,+ ), ratio population population population ), , population ( ), population ( ,+ ( ), ( population ,+( ) , + ) and and ) and ) and purepure and pure pure( TFP pure TFP TFP ) (TFP ( and (+ , ( pure ), ,,+), ), TFP +), ( , ,+ + , + , + rate ( ,+ ), the , ,+, working + + , + age to population , ,+ + , + , + ratio (, + , ,+ + , ,+ ), population (,+ + , are are areare exogenously are exogenously exogenously exogenously are exogenously exogenously are determined. determined. are determined. exogenously determined. determined.exogenously determined. The The The determined. The growth The growth The determined. growth growth rate growth growth raterate rate of rateThe of rate ofof measured of The measured growth measured measured of measured measured growth rate public public rate public of public public of measured capital capital public measured capital capital per capital capital per per per public worker worker per per worker workerpublic worker capital worker (( ( ( capital per ( ) per worker + is , ) )is ) is given isworker is given given ) given is( given given byby ( by by ,+by ) , is ) +give are exogenously determined. The growth rate ofmeasured ,+ + , , + , + public capital per wo equation equation equation equation equation by equation (12), (12), (12), (12), using using equation using (12), using (12),using the equation thethethe using growth growth (12), the growth growth the growth (12), using rate rate growth rate rate of using the ofthe rate ofof the the the growth of the rate public the publicpublic of growth public the rate capital public capital rate capital capital public of the stock capital stockof stock public stock capital the stock public (equation (equation (equation capital (equation stock capital (11) (equation from stock (11) (11) (11) as stock as (11) equation as (equation an as an an an (equation intermediate intermediate as intermediatean intermediate (11) (11) intermediate as an (11) as step.anstep.step. intermediate as step. an intermediate Private step. Private intermediate Private Private Private step. ste Pr capital per worker equation growth (12), ( using ) is the given growth by rate of (15). equation the public The capitalrate growth stock of (equation the efficiency (11) ofaspub an capital capital capital capital capital per step. per perper worker worker Privateper worker worker capital worker growth growth capital growth growth per growth per ( (worker ( ( worker ( ) ,+) isgrowth is ) growth ) is given is given given ) given is ( by by given byequation equation by equation by ) equation is equation is given (15). given ,+ ,+ (15). (15). (15). by The by The (15). TheThe equationgrowth growth The equation growth growth (15). rate growth rate (15). rate rate of TheThe of ratethe of of the the growth theof growth efficiency the efficiency efficiency efficiency rate efficiency rate ofof of public of theofthe public public public of efficiency capital public efficiency capital capital capital capital of public ca , + + , , + capital ,+ per worker growth (,+ ) is given by equation (15). The growth rate of th ( ( (( ( , , ,+ + of ) ,+ +)is )is public , ) is given + is given )givengiven is (by by given capital by by equation ,+( by equation equation is+given ), equation equation ) (14) (14) is is (14) (14) given by using given using (14) using equation using by by the equation the using the the growth growth equation (14) growth thegrowth using growth(14) rate (14) raterate rateofusing using the of rate the ofofthe the growth the of the the public public the growth public public growth rate capital of capital capital publiccapital raterate the capitalstock of of stock stock public stock the the from stock public from public from capital from from equation capital equationequation equation capital stock equation (11) stock stock from (11) (11)(11)as as (11) from from equation an asas an as an an equation an (11) ( ) is given by equation (14) using the growth rate of the public capital stock intermediate intermediate intermediate intermediateintermediate equation step. step. (11) step. intermediate step. step. as an intermediate intermediate step. step. step. ,+ intermediate step. 20. 20. 20. 20. Finally, Finally, 20. Finally, Finally, the the the the Finally, Finally, the 20. model the model20. model model Finally, Finally, model isis model is is is closed the closed closed closedis the model closed closed by by by by by updating model byupdating is updating is public closed updating updating closed updating public public by public by updating updating public public capital-to-output capital-to-outputpublic capital-to-output capital-to-output capital-to-output capital-to-output public using capital-to-output using capital-to-output using using using equation equation equation equation using equation equation (17) using (17) and using (17) (17) (17) the equation and (17) andand andequation the the and private the the (17) (17 the and private private capital-to-output private ratio capital-to-output 20. capital-to-output using Finally, equation ratio the ratio (18) using model using (with equation is closed equation the (18) growth (18) (withby updating (with rates the the in per-worker growth public growth rates capital-to-output rates in in terms):per-worker per-worker usin term terms): private private private private capital-to-output capital-to-output capital-to-output capital-to-output ratio ratio ratio using using ratio using capital-to-output ratio using equation equation equation using equation (18) equation (18) (18)(with (18) (with (with (18) the (with thethe (with the growth growth growth growth the growth rates rates rates rates in in in per-worker ratesper-worker in per-workerper-worker in terms): per-worker terms): terms): terms): terms): private capital-to-output ratio using equation (18) (with the growth rates in (+ (+ (+ (+ (+ )) ),+ ) ) (+ (+) ,+ ) ,+ ,+ ,+ ,+ ,+ + + = + + = = = + = + = + = (+,+ ) + + + ++ + + + + ,+ + ,+ + ,+ ,+ ,+ + + + ,+ ,+ + = (17) + +,+ (17) (17) (17) (17) (17) (17) (17) (17) + = (18) + D. Aged Care Policies, Legislation and Programs Malaysia has started to formulate its strategic aging framework and aged care policies since the mid-1990s and involved two leading ministries—the Ministry of Women, Family and Community Development (MWFCD) and the Ministry of Health (MOH). In 1995, the National Policy for the Elderly (NPE) was the first strategic policy on aging in Malaysia. To implement this policy, the National Advisory and Consultative Council for Older Persons under the chairmanship of MWFCD was set up in May 1996 and a Plan of Action for the Older Persons was formulated in December 1998. Based on a review of the first strategic policy and the Plan of Action, the National Policy for Older Persons (NPOP, Dasar Warga Emas Negara) and Plan of Action for Older Persons (Pelan Tindakan Warga Emas Negara) were introduced in 2011 with a broader goal for the old persons related to their development, health and well-being, an enabling and supportive environment, to allow them to have a high sense of self-worth and dignity, which includes short-term, medium-term, and long-term actions up until 2020. Aged care normally includes two aspects of social care and health care. Following the 1995 NPE, the Government developed the National Plan of Action Plan for Health Care of Older Persons in 1997 and issued the National Health Policy for Older Persons 2008 (Dasar Kesihatan Warga Emas Negara) to promote healthy, active and productive aging through integrated and comprehensive health and health care services. In addition, the National Plan of Action for Nutrition of Malaysia III (2016-2025) provides nutritional guidance for older persons. Those policies and action plans have articulated relevant Government programs to support the development of the aged care sector. Aged care legislation in Malaysia has a relatively long history, rooted in its traditional welfare provision model dated back to the late 1970s, and since the early 1990s has kept evolving with a focus on regulating the private sector. The Destitute Persons Act 1977 (Act 183) and related rules in 1980s codified the approaches and procedures of managing and operating public welfare homes for destitute old persons. In 1993, as the numbers of elders increased and demand for private aged care services increased, the Care Centers Act (Act 506) superseded the previous rules and placed the regulation of private aged care 140 A Silver Lining: Productive and Inclusive Aging for Malaysia Methodological Annex centers (and other care centers), both day care and residential, under the responsibility of the Department of Social Welfare (Jabatan Kebajikan Masyarakat or JKM), MWFCD. This legislation was shortly followed by the Private Healthcare Facilities and Services Act 1998 (Act 586), which placed the regulation of private nursing homes under the responsibility of MOH. There has been a perception among stakeholders that the regulatory stringency required of care centers under the Care Centers Act was low and that the stringency required for private nursing homes under the Private Healthcare Facilities and Services Act was high for the typical needs of residential aged care centers. To address these concerns, a new piece of legislation—the Private Aged Healthcare Facilities and Services Act (Act 802)—was passed by Parliament in 2018. This most recent act supersedes the Care Centers Act allowing for a 5-year transition period during which care centers licensed under the earlier act will be automatically licensed under the new Act until 2024 and places the responsibility of regulating private aged care facilities, both day care and residential care, under MOH. Public aged care programs cover institutional care and home and community-based care and they are managed and supervised by JKM. Specifically, a short description of the aged care programs below summarizes the objectives and types of aged care services each program proposes to cover. However, the actual delivery of aged care services and their quality will be further discussed in the later sections. Except the activity centers program, all the programs should meet certain eligibility criteria to access the corresponding services (See Table D.1).42 • Rumah Seri Kenangan (RSK), Rumah Warga Tua (RWT) in Sabah, and Rumah Sejahtera (RS): Care homes provide care and protection to the poor independent older persons that allows them to live in peace and enjoy a good quality of life. The catalog of services includes care and protection, guidance and counseling, recreation, medical treatment, occupational therapy, physiotherapy. • Rumah Ehsan (RE): Nursing homes for chronically ill provide care, treatment and protection to the poor dependent older persons that allows them to continue to live in a comfortable and safe environment. The catalog of services includes care and protection, guidance and counseling, physiotherapy, recreation and job recovery, medical treatment. • Pusat Aktiviti Warga Emas (PAWE): Activity centers for older persons offer a place for senior citizens to participate in and perform daily activities in the communities. Activities include religions, recreation, therapy and rehabilitation, health seminar, training or courses, aiming to promote active, productive and healthy aging. It involves strategic cooperation between MWFCD, other Government agencies and non-governmental organizations (NGOs). • Home Help Services: Provide social outreach services and support to the low-income older persons and the disabled, especially to those living alone in the absence of family members. This program followed the initiative of NBOS7 - 1Malaysia Family Care, commenced in July 2012, carried out in a strategic partnership between JKM, MOH, and the Welfare Voluntary Organization (PSK). The catalog of services is broad from home visit to providing food and drinks, personal hygiene, house cleaning, to mobile services, physiotherapy, and training family caregivers. Case workers are trained volunteers. • Unit Penyayang Warga Emas (UPWE): “We Care Services” provide transport facilities and services to the older persons who live alone and have physical problems and mental disabilities that limit their movement to get treatment and other services from hospitals or clinics. It was introduced in 2008 by JKM and commissioned to the Central Welfare Council of Malaysia to deliver the services. 42 In addition to the public aged care programs listed here, JKM also provides Financial Assistance for artificial aids and assistive devices to assist person with disabilities (PWDs) who are unable to purchase devices such as artificial legs and arms, calipers, crutches, wheelchair, hearing aids, special glasses, special shoes and other aid devices that are recommended by the doctor or specialist. This program aims to improve PWDs capabilities and allow them to be self-reliant. A Silver Lining: Productive and Inclusive Aging for Malaysia 141 Methodological Annex • Respite Care: Provide an alternative service to the caregivers or guardians who could place the older persons temporarily in an institution for a specific period, due to various reasons that require a temporary arrangement. Institutions provide facilities and basic services such as food, health and shelter for the older persons. MOH has introduced various programs to promote health awareness and provide health care services for older persons. This includes activities such as health promotion and education, health screening and assessment, home visits and home medical services at the community level through PAWE and health clinics—Older Persons Clubs (Kelab Warga Emas). At public hospitals, long term care, rehabilitative care, and psychogeriatric care are also provided. Some public hospitals have also opened specialist geriatric units for older persons. MOH has also provided training programs such as six months post-basic course in geriatric nursing for primary care staff in handling health issues for older persons. In addition, home visit, home care nursing, and some basic training for family caregivers were also provided. As a federation, aged care provision, financing and governance in Malaysia involve the Governments at the federal, state and local levels, and they have different roles in the administrative hierarchy. Both the Federal Government and State Governments have responsibilities to provide for older persons, as social welfare and social services are part of List III (Concurrent List) of Ninth Schedule of the Constitution of Malaysia. The Federal Government has aged care policies and funding for these policies, but State Governments have policies and funding as well. Districts and local authorities are largely responsible for policy coordination and implementation. They are encouraged to provide supplementary financing if local fiscal capacities allow. It is interesting that the Federal and State Governments are also involved in direct provision of aged care services, while Districts and Local Authorities are not. TABLE D.1. Eligibility criteria for public aged care programs Programs Eligibility Criteria RSK, RWT, and RS 1. Malaysian; 2. Age 60 years and over; 3. No income; 4. No infectious diseases; 5. No relatives / heirs; 6. No permanent residence; 7. Able to take care of themselves RE 1.Malaysian; 2. Not able to care for themselves; 3. No relatives / heirs; 4. No infectious diseases; 5. No income; 6. Authorized by Government Medical Officer as destitute patients PAWE No eligibility criteria; No charge Home help services Low-income older persons and the disabled; some eligibility criteria to volunteers UPWE The older persons who live alone and have physical or mental limitations Respite care Senior Citizens: (i) Bed-Ridden Disabled and Chronically ill; (ii) Age 60 years and above (husband / wife, brother and sister, mother / father, mother / uncle, grandfather / grandmother, grandfather / grandaunt); (iii) Able to manage themselves; and (iv) Malaysian; Heirs/ Guardian: (i) Heir to senior citizens; (ii) Custodian (foster family, friends); and (iii) Malaysian citizen; RM350 per person per month; Placement period: Minimum—one day (1 hour -24 hours); Maximum—30 days or subject to the approval of JKM Source: Authors based JKM. 142 A Silver Lining: Productive and Inclusive Aging for Malaysia Methodological Annex Local Governments in Malaysia largely fall under the purview of the State Governments but the Federal Ministries, specifically the Ministry of Housing and Local Government and the Ministry of Federal Territories, have an oversight in expressing, performing, and monitoring all laws related to Local Government (see Table D.2). There are three types of Local Government with a range of responsibilities reflecting their size and capacity: City Councils (in the Federal Territory of Kuala Lumpur and in Sabah and Sarawak), Municipal Councils, and District Councils. The scope of authority is related to urban planning, the development of basic facilities and infrastructure, monitoring development, public health, waste management, business licensing, maintenance of peace and landscape, and improving the local economy. Local Authorities receive revenue from taxes, non-tax revenue, and Federal or State Government allocations. TABLE D.2. Roles and responsibilities of Federal, State, and Local Government in aged care Level of government Governance Financing Provision Federal Consultation, policy setting Predominant source of Direct provision and and regulation financing commissioning State Consultation, policy setting Supplementary financing Direct provision and and regulation commissioning District and Local Policy coordination and Supplementary financing Commissioning of services Authorities implementation from NGOs Source: Authors. Under National Health Policy for Older Persons 2008, coordination of federal and state policy implementation is the responsibility of both State Action Councils and State Development Committees (chaired by Deputy Secretary of the State Government with the Director of the state- level Social Welfare Department as secretariat) at each state in accordance with the National Policy for Older Persons. Members of the State Development Committees include Government agencies, the private sector, and NGOs. Yearly implementation plans for each state, reflecting the priorities of each state, are to be submitted by the State Committees to the national-level Technical Committee. This state-level structure is approximately mirrored at the District or Local Authority level. Local Authorities further play an important regulatory role in the licensing of private aged care facilities. E. Additional References Devadas, Sharmila and Steven M. Pennings. 2019. Assessing the Effect of Public Capital on Growth: An Extension of the World Bank Long-Term Growth Model. Journal of Infrastructure, Policy and Development 3: 22–55. A Silver Lining: Productive and Inclusive Aging for Malaysia 143 CONNECT WITH US wbg.org/Malaysia @WorldBankMalaysia @WB_AsiaPacific http://bit.ly/WB_blogsMY