assesment of public pension system in sea - 2015-aper asher and bali

8/17/2019 Assesment of PUblic Pension System in SEA - 2015-APER Asher and Bali

1/22


2/22

Public Pension Programs in Southeast Asia:An Assessment

Mukul ASHER 1† and Azad S. BALI 21LKY School of Public Policy, National University of Singapore, 2School of Management & Governance,Murdoch University

This paper assesses public pension programs in select Southeast Asian economies (Indonesia,Malaysia, the Philippines, Singapore, Thailand, and Vietnam – henceforth referred to as the SEA6)and the key issues facing them. The criteria used in assessing pension systems are the philosophy of pension design, the extent of coverage, investment policies and performance, and administrativeand compliance costs. The paper argues that three broad reform directions to strengthen publicpensions merit consideration. The rst direction is to enhance the professionalism of the existingprovident and pension fund organizations, including their governance practices. The second direc-

tion is to strengthen the role of noncontributory budget-nanced pensions (e.g. social pensions).The third is to adopt a systemic perspective to pension reform that includes reforms in comple-mentary areas (labor markets, public nancial management practices, and the civil service); devel-oping a nancing-mix of pensions; and lastly, improving effective coverage by exploringcomplementarities between health care and pension programs.

Key words: pension reform, public pensions, retirement income policy, social protection, socialsecurity, Southeast Asia

JEL codes: H55, H75, J26

1. Introduction

Current demographic trends suggest that most economies in Southeast Asia will age atrelatively low incomes, and at a pace that will allow a small window of opportunity foradjustments in the design of pension programs and reforming institutions that supportsocial protection systems. Pension systems will have to nance retirement expenditurefor an ageing population for a longer duration, and will therefore have to increase theshare of society’s resources to devote to the elderly. Additional funding will also requirechanges in the nancing-mix used to provide pensions in these countries. 1

While there is signicant heterogeneity among pension systems in Southeast Asia in

terms of their social protection philosophy, institutional features, macroeconomicsustainability, and coverage rates, a common theme that emerges is that the contributionof the public or government-organized pension programs in ensuring old-age incomesecurity is relatively small. This stems from many reasons.

First, most of these countries have large shares of the labor force that are employed inthe informal sector, and extending social security coverage and ensuring a high density of

†Correspondence: Mukul Asher, LKY School of Public Policy, National University of Singapore,469c Bukit Timah Road, Singapore 259772. Email: [email protected]

bs_bs_banner

doi: 10.1111/aepr.12100 Asian Economic Policy Review (2015) 10, 225–245

© 2015 Japan Center for Economic Research 225

mailto:[email protected]:[email protected]


3/22

contributions is challenging. 2 Second, there are many governance challenges that haveweakened professionalism and organizational effectiveness of provident and pensionfund organizations in developing economies. Third, social protection reects in somemeasure a contract between the state and its people. The terms of reference that guidethis contract vary across economies and stem from differing philosophies, ideologicalpreferences, and collective opinion on the role of the state in provision and nancing of social policy.

Diminished growth prospects in the aftermath of the economic crisis have accentu-ated the economic and political need for strengthening social safety nets and in particu-lar expanding publicly nanced and organized programs. Many developing economies inAsia (including China, India, Indonesia, Thailand, and Vietnam) are at the precipice of implementing or have implemented wide-ranging social protection reforms that outlinea greater role of the state in administering and nancing social policy, particularly healthcare and pension programs (Ramesh, 2009).

While public pension programs (contributory and noncontributory) have an impor-tant role in managing old-age income security (smoothening consumption, mitigatingpoverty, and insuring longevity, ination, and survivors’ risks), they also have an impor-tant macroeconomic function in supporting economic growth (Barr & Diamond, 2008).Over the past decade, the importance of noncontributory retirement income transfersnanced from the state has also been part of policy debate in social security reform.

In the above context, this paper assesses public pension programs in the SEA6 econo-mies, Indonesia, Malaysia, the Philippines, Singapore, Thailand, and Vietnam, and thekey issues facing them. The characteristics used in assessing pension systems are philoso-phy of pension design, the extent of coverage, investment policies and performance, andhow these economies manage pension-related administrative and compliance costs.

The next section reviews demography trends in the SEA6 countries which suggestthat there is an urgent need for pension reform. This is followed by a comparative assess-ment of pension programs on the basis of the aforementioned criteria. The paper con-cludes with suggestions for broad directions of reforms for public pensions.

2. Demographic Trends: Key Characteristics

Advances in longevity at all ages through the life cycle is historically unprecedented andhas implications for ensuring old-age income security. Assuming constant retirement

age, a higher longevity implies that an increasing proportion of the average life span isspent in retirement. One of the main drivers of longevity is the compression of mortality at advanced ages, resulting in an increasing share of the old-old (people aged above 85)in the population (Eggleston & Fuchs, 2012). Not only do health care and pension pro-grams need to be nanced for a longer period, but also for a growing number of indi-viduals. While aggregate pension spending in real terms (adjusting for prices) is largely afunction of time spent in retirement by a cohort of retirees, health-care spending on theother hand increases disproportionately with age. Successive cohorts of retirees are there-fore spending relatively more on health care in real terms as they live longer. This implies

Public Pension Programs in Southeast Asia Mukul Asher and Azad S. Bali

© 2015 Japan Center for Economic Research226

https://www.researchgate.net/publication/227466905_Reforming_Pensions_Principles_and_Policy_Choices?el=1_x_8&enrichId=rgreq-f154c033-6ca1-4fd7-a51d-e581dbcac9cf&enrichSource=Y292ZXJQYWdlOzI4MDkxNzUwMDtBUzoyNzQ4OTYwNjQzNDgxNjFAMTQ0MjU1MTczNzQ2NQ==https://www.researchgate.net/publication/227466905_Reforming_Pensions_Principles_and_Policy_Choices?el=1_x_8&enrichId=rgreq-f154c033-6ca1-4fd7-a51d-e581dbcac9cf&enrichSource=Y292ZXJQYWdlOzI4MDkxNzUwMDtBUzoyNzQ4OTYwNjQzNDgxNjFAMTQ0MjU1MTczNzQ2NQ==


4/22

that a greater share of society’s resources will have to be devoted to managing age-relatedexpenditure.

Before analyzing the demographic indicators and their implications, it is importantto acknowledge that there may be variations within and across cohorts that are notrevealed by aggregation. For instance, there are variations in the life expectancy at birthof males and females, with women living longer than men. A similar trend is evidentacross life expectancies in advanced ages (United Nations Department of Economic andSocial Affairs [UNDESA], 2012). Bell and Miller (2005) report the relative life expectan-cies at birth and at age 65 for American men and women born in 1900 through 2000.Not only have life expectancies at birth increased, but the probabilities of surviving untiladvanced ages have increased. Studies of human ageing nd that not only are individualsliving longer as a result of delays in mortality, but are reaching old age in better health,thereby delaying human senescence (Vaupel, 2010). However, the relative increases in lifeexpectancies (both at birth and at age 65) and the relative probability of reaching age 65have been greater for men than for women.

Similar variations are also evident across income distributions. Poterba (2014) ana-lyzes the mortality gradient (disparity in health outcomes across socio-income distribu-tions) and nds that those at the upper end of distribution have a longer time horizon toplan and save for retirement relative to what the aggregate measures of life expectancy may suggest. The corollary implies that the poor will have lower lifecycle supply of laborand smaller time horizon to plan for retirement. The implications of this in dened con-tribution 3 (DC) organized pension plans with limited risk-pooling and societal transferswill be more severe.

Another important impact of demographic change on social security and publicpension programs relates to the nancing-mix of pensions and how these vary acrossand within cohorts of retirees. The relative contribution of instruments used to accessfunding will vary across countries, and even across and within cohorts in a givencountry. As each cohort of retirees is expected to live for a longer period of time, it isexpected that there will be heterogeneity in the source of funds or the nancing-mix used to support their retirement. This heterogeneity must be recognized and built intopension design.

Demographic and labor market trends for the SEA6 countries are provided inTables 1 and 2 on the basis of which the following observations may be made.

Even in the relatively conned geographical area of Southeast Asia, there is signi-

cant variation in total population and the level and pace of ageing. The total populationof the aged in all economies, except Thailand, is projected to increase signicantly overthe next two decades. Data in Table 1 portends a very rapid pace of ageing. The share of the aged in the total population is projected to more than double in all economies(except the Philippines) over the next two decades. This is a relatively short period of time to ensure adequately preparing for an ageing population. The data also suggest dif-ferent scaling-up challenges across these economies. For instance, Singapore and Malay-sia’s pension arrangements will have to cater for between 1 and 3 million additional individuals entering retirement; while in the Philippines, Thailand, and Vietnam this

Mukul Asher and Azad S. Bali Public Pension Programs in Southeast Asia



5/22

ranges between 5 and 10 million; and for Indonesia numbers about 20 million. This sug-gests that there is relatively small window for reform in not only pension design to adaptto rapid ageing, but also in supporting institutions such as labor markets and in publicnancial management practices. The last two columns of Table 1 depict the share of

Table 1 Demographic characteristics of SEA6

PopulationPopulation aged over 65(share in total population)

Population aged over 80(share in total population)

2010 2035 2010 2030 2010 2035

Indonesia 240.7 303.4 12.1 (5.0) 33.2 (10.9) 1.9 (0.8) 4.4 (1.4)Malaysia 28.3 38.5 1.4 (4.8) 4.3 (11.2) 1.8 (0.6) 0.7 (1.8)Philippines 93.4 135.9 3.5 (3.7) 9.6 (7.0) 0.4 (0.4) 1.4 (1.0)Singapore 5.1 6.8 0.5 (9.0) 1.6 (23.0) 0.1 (1.9) 0.4 (6.3)Thailand 66.4 66.8 5.9 (8.9) 15.3 (22.9) 1.1 (1.6) 3.4 (5.1)Vietnam 89 103.3 5.8 (6.5) 15.9 (15.4) 1.6 (1.8) 3.3 (3.2)World 6916.2 8743.5 530.1 (7.7) 1118.5 (12.8) 108.3 (1.6) 240.1 (2.8)

Notes: The unit of measurement for all population numbers is millions and shares are measuredas percentages.Source: Authors’ calculations based on data from the United Nations Department of Economicand Social Affairs (UNDESA) (2012).

Table 2 Employment and labor force participation rates in SEA6

LFPR (15–64)in 2010

LFPR (65+ )in 2010

Changein LFPR (15–64) from2010 to 2020

Changein LFPR (65+ ) from2010 to 2020

% Change in the shareof economically activein population over theperiod 2010–2035

Indonesia 70.4 52.7 0.3 1.7 2.5Malaysia 64.7 23.8 − 0.7 0.4 0.6Philippines 66.1 37.4 − 0.2 − 2.4 4.1Singapore 71.5 18 1.6 2.0 − 10.3Thailand 77.8 30.6 0.1 1.0 − 8.0Vietnam 77.9 13 0.0 − 1.3 − 1.5World 69.9 19.5 1.0 0.4 − 1.2

Notes: All numbers are %; share of economically active in population is between ages 15 and 64.LFPR (15–64) and LFPR (65+ ) refer to the Labor Force Participation Rate for ages 15–64 and 65

over, respectively.Sources: The data are adapted from the Statistical Annex Part A in International LabourOrganization (ILO) (2010) and United Nations Department of Economic and Social Affairs(UNDESA) (2012).




6/22

those aged about 80 in the population in 2010 and 2035. These shares will also morethan double in all economies, with the exception of Indonesia and Vietnam. The con-sumption patterns of those about age 80 can be expected to differ signicantly fromthose for people aged 65. This suggests that adequate infrastructure to take care of theold-old (traditionally dened as those above age 80), such as health and palliative care,will need to be developed rapidly.

Table 2 presents labor market trends in the SEA6 countries. Greater lifecycle laborsupply enables individuals to sustain (for a given trade-off) higher annual consumptionduring retirement. Theories of economic growth assume a strong role of employment indriving increased savings and investment and contributing to the demographic dividend.While not illustrated in Table 2, in most economies labor force participation rates(LFPR)4 for men are higher than for women. If pension programs are designed on prin-ciples of commercial insurance and not on social insurance or solidarity principles,such trends will give rise to lower replacement rates 5 for women and inequity within thepension system. This is because women (as a group) have lower lifecycle labor supply and lower incomes, and, therefore, lower resources, but (as a group) live longer than menand will have to nance retirement spending for a longer period. The LFPR numbers alsomask trends between rural and urban areas. In the developing economies of Indonesia,the Philippines, Thailand, and Vietnam, it is reasonable to assume that improved accessto basic amenities such as water, electricity, and sanitation will improve LFPR, in turnhelping to plan for retirement savings.

In most of the economies in Table 2, the gains in LFPR for the age groups 15–64and above 65 will be marginal over the next decade. The Philippines and Vietnam areexpected to experience a reduction in LFPR for those aged over 65. The last column inTable 2 is particularly relevant for pension systems. The share of the economically active(i.e. those aged between 15 and 64 years) is expected to decline for Singapore, Thailand,and Vietnam, and grow marginally for Indonesia and Malaysia and signicantly for thePhilippines over the next decade. When this is viewed in the context of the data pre-sented in Table 1, it suggests that a smaller number of individuals can be part of the laborforce and potentially employed to support the elderly population. While this share is afunction of mortality and fertility rates, and cannot be adjusted in the medium term, thepolicy goal is to improve generation of sustainable livelihoods or gainful employment forthe vast majority of those in the economically active age group so that the elderly popu-lation can be supported for a longer period of time.

These demographic trends suggest that greater funding will have to be made availableto meet old-age expenditure needs. With increasing longevity, current age-specic con-tributions by individuals and households to national savings, consumption, and invest-ment may change. As individuals will have to sustain consumption for a longer period of time, without transfers, this will have to be balanced with participation in the labormarket for a longer period of time or through higher savings or reduced levels of currentconsumption or a combination of the above (Poterba, 2014). The data presented in thissection suggest that there will be marginal improvements over the next decade in laborforce participation in both the economically active age groups and those above age 65,




7/22

and, therefore, greater funding through transfers, particularly from the state, will play animportant role in providing old-age income security to economies in the SEA6.

3. Assessment of Current Pension Arrangements

The criteria used in assessing pension systems are the philosophy of pension design, theextent of coverage, investment policies and performance, and how these economiesmanage pension-related administrative and compliance costs. The unit of analysis is thepension system, and while the design of the programs is summarized below, the discussionwill not delve into institutional arrangements of various pension programs across theSEA6.6

3.1 Philosophy of pension designPublicly organized pension programs may be viewed as a social contract between a stateand its people, and the terms of reference that guide this contract depend on the prevail-ing views on the role of the state in organizing and paying for social protection in asociety. This in turn impacts on the choice of instruments (social or private insurance,mandatory or voluntary savings, and social pensions) and the relative weights of theseinstruments (where more than one instrument is used) in organizing a pension system.The philosophy of pension design thus refers to the choice of instruments and the rela-tive weights among instruments in organizing pension arrangements for different occu-pational groups within a society. This in turn has implications for the extent to whichhouseholds, businesses, or people collectively through the government bear retirementsecurity risks, the extent of risk-pooling, and ultimately impacts on the adequacy, acces-sibility, fairness, and sustainability of the pension system.

In SEA6, with the exception of the Philippines, no one philosophy of pension designis currently applied across occupational groups. In particular, differences in pensionarrangements between private and public sector workers (including military personnel)are noticeable. In the Philippines, social insurance, which enables risk-pooling, has beenincorporated in the programs for public sector and private sector workers. Even in thePhilippines, the credibility and the generosity of pension promises for governmentworkers is far greater than for private sector workers. This is also the case in Thailand,Malaysia, and Singapore, though the details of arrangements vary among them (see Park,2013).

The former British colonies of Malaysia and Singapore rely on mandatory savingsusing the DC-method, in which individuals bear the longevity, ination, and survivors’,as well as investment risk during the accumulation and pay-out phase. There is limitedpooling or sharing of risks: Malaysia permits a lump-sum withdrawal on reaching retire-ment age, and Singapore mandates the purchasing of an annuity that is priced on com-mercial insurance principles (Asher & Bali, 2013). This implies that income inequalitiesand vulnerabilities that exist during the accumulation phase or working years are contin-ued, if not accentuated, during the retirement period. There is a small element of socialpensions and social insurance for workman compensation schemes in Malaysia.




8/22

The pension programs for the Philippines, Thailand, and Vietnam are organized onthe principles of social insurance for both formal sector workers. Thailand, however, hasa voluntary provident fund for private sector workers and a mandatory provident forgovernment employees. The recent reforms in these economies have, however, expandedthe role of social pensions.

Indonesia’s Sistem Jaminan Sosial Nasional (SJSN) Law, enacted in 2004 and cur-rently being implemented, uses social insurance principles to organize on a universalbasis both the health care and pension systems, representing a marked departure fromexisting practices. The above discussion suggests that the pension philosophy in theSEA6 are quite varied, and drastic changes in them are feasible as evidenced by the initia-tion of social insurance methods for private sector workers in Thailand in 1998 and by Indonesia’s SJSN law in 2014.

3.2 CoverageThere are three broad dimensions to coverage. The rst refers to the number of people orretirees that are enrolled in a program that provides some form of insurance againstvarious risks during old age. The second refers to the range of risks covered. In pensionsthese usually include longevity, survivors’, and disability risks. The third dimension of pension coverage refers to the adequacy of pension benets in providing a replacementrate that not only covers ination risks and mitigates old-age income poverty, but alsosmoothes their consumption. This is illustrated in Figure 1. The same criteria are rel-evant for health-care programs as well (i.e. the number of people covered, the adequacy of benets, and lastly, the range of risks covered). 7 As health-care needs increase dispro-portionately with age, the lack of adequate health-care coverage across all three dimen-sions reduces the adequacy of pension benets. This is particularly important in theSEA6, where most health-care expenditure is nanced out-of-pocket.

In the SEA6 countries, coverage is mostly focused on increasing the number of indi-viduals that are “covered” under a statutory program and the range of risks covered.This is commonly referred to as legal coverage. The data in Table 3 suggest that there isuniversal legal health-care coverage in all economies except Indonesia and Vietnam;however, these programs do not provide adequate benet levels as most of health-carespending is nanced out-of-pocket (except in Thailand). This reduces the real value of pension benets as retirees will have to draw down their pension balances to pay forhealth care. For pension programs, the Philippines, Singapore, and Thailand have

between half and three-fourths of the current working-age population covered by socialsecurity laws, while the ratio is much smaller for Indonesia, Malaysia, and Vietnam.However, there is wide variation in the active contributors to the pension program. Inmost DC-type pension programs, the density of contributions is important to ensureadequate replacement rates. The share of active contributors is much lower than thosecovered by the program. The last column in Table 3 is the share of the elderly populationthat receives pensions, and again demonstrates large variation. It is very high in Thai-land, but less than 40% in other SEA6 economies. This suggests that there is considerablescope to improve effective coverage in the SEA6. Given the low effective coverage, this




9/22

suggests that signicant shares of retirement expenditure and health-care expenditurewill have to be nanced from individual and household savings. Improvements in orga-nizational effectiveness and coverage ratios of public pension schemes are an urgentimperative in economies such as Indonesia, the Philippines, and Vietnam.

Administrative and compliance costsThe core functions of any pension or provident fund may be stated as reliable collectionof contributions/taxes and other receipts; payment of benets for each of the schemes ina correct way; efcient nancial management and productive investment of providentand pension fund assets; maintenance of an effective communication network, includingdevelopment of accurate data and record-keeping mechanisms to support collection,payment, and nancial activities; and the production of nancial statements and reportswhich contribute to effective and reliable governance, duciary responsibility, transpar-ency, and accountability of old-age institutions (Ross, 2011).

Figure 1 The three dimensions of coverage.Source: Adapted from a gure in the World Health Organization’s (WHO) Health Financingfor universal coverage. Available from URL: http://www.who.int/health_nancing/strategy/dimensions/en/Note: The box with solid lines refers to the coverage of health-care risks, and the speckled box refers to coverage of old-age income risks.



http://www.who.int/health_financing/strategy/dimensions/en/http://www.who.int/health_financing/strategy/dimensions/en/http://www.who.int/health_financing/strategy/dimensions/en/http://www.who.int/health_financing/strategy/dimensions/en/


10/22

Table 4 provides estimates of the performance of provident and pension fund orga-nizations in select economies in SEA6. Lowering the administrative costs of providentand pension fund organizations implies a higher rate of return for members, and, conse-quently, higher replacement rates. Similarly, increasing the value of the denominator(namely, total income, contributions, and total assets) will also improve investment per-formance and replacement rates.

As these estimates are based on data from the annual reports of the respective organi-zations, and given the heterogeneity among the countries, the indicators may not be

Table 3 Legal and effective coverage of pensions and health-care programs in SEA6

Estimateof health

coverage asa % of totalpopulation

Health-carespending not

nanced by OOP spending(2011)

Estimate of legal coverage

for old age (%of working-agepopulation)

Activecontributorsto a pension

scheme (% of working-agepopulation)

Proportion of elderly populationabove statutory

pensionable agethat receiveold-age pensions

Indonesia 59.0 50.1 < 25 14.1 8.1Malaysia 100.0 64.6 63.8 19.8Philippines 82.0 44.1 50–75 54.7 28.5Singapore 100.0 39.6 50–75Thailand 98.0 86.5 50–75 21.3 81.7†Vietnam 42.0 43.9 < 25 12.4 34.5

Source: Adapted from International Labour Organization (ILO) (2010, 2014) Column D.†These proportions refer only to beneciaries of the old-age or disability social pensions.

Table 4 Estimated administrative and compliance costs of select pension and provident fundorganizations in SEA6

Variable

EmployeesProvidentFund(Malaysia)2012

SocialSecurity System(Philippines)2012

CentralProvidentFund(Singapore)2012

SocialSecurity Organization(Thailand)2012

Operating expenses as share of gross income

1.77 6.03 2.48 3.65

Operating expenses as share of contributions

2.92 8.20 0.77 4.77

Operating expenses as share of assets 0.25 2.13 0.09 0.63

Source: Adapted from Asher and Bali (2014).




11/22

strictly comparable. Moreover, indicators are crude as they focus only on operating costand not on manpower deployed to provide services. They, for example, do not take intoaccount the scale of membership of these various schemes or the physical size of theeconomy. It should be noted that some provident and pension funds include investmentexpenses as part of their operating expenditure (e.g. Malaysia). With these caveats, thefollowing observations can be made.

Operating expenditure as a share of assets is relatively high in Thailand and the Phil-ippines. This may reect the extent to which the pension system design permits accumu-lation of member balances and the time since the scheme has been in operation. Thus,a possible reason for a higher ratio in the Philippines is the partial funding nature of itssocial security scheme which is a dened benet (DB) scheme. Member contributionsand investment income during the year are used to pay for the benets of current retir-ees. It is only the current demographic structure of the Philippines, resulting in a smallernumber of retirees in relation to contributing members, which has led to the accumula-tion of funds. Thus, as the population of the Philippines ages and the proportion of retirees relative to working population increases, its total assets will decline or at leastexhibit less rapid growth.

In contrast, the schemes of Malaysia and Singapore are DC-type. As a result, the con-tributions of members plus investment returns, less preretirement withdrawals, accumu-late with the provident fund organization, and are returned to members only at the time of retirement, as a lump sum payment, an annuity, or other types of periodic payments.Thus, funds under management can potentially be higher than under a DB scheme, whichutilizes the current contributions of workers plus investment incomes to pay for pensions.

The ratio of operating expenditure to total income is substantially higher in the Phil-ippines (6.3%) than in other countries. The ratio for Thailand (3.7%) is also relatively high when compared to Malaysia and Singapore. An explanation of the comparatively high operating costs to contributions ratio includes the following. First, low compliancelevels are a main reason for low contribution levels. Second, the proportion of inactiveor multiple account-holding members is high. It is estimated that only one-third of Social Security System (SSS) (Philippines) members are active contributors, which raisesadministrative costs. Third, the contribution rate and the wage ceiling have been rela-tively stable, while operating costs have risen over the years. The Philippines is notunique in these respects.

3.3 Investment policies and performancePension contributions play a relatively smaller role as compared to the duration forwhich they are invested, and the net real return after subtracting fund management andother related costs credited to the account of the members. Investment policies and per-formance of accumulated pension assets is a challenge globally, especially in the currentglobal environment of low or negative real interest rates. Nevertheless, as Chile’s experi-ence shows, real annual returns in the range of 4–6% are possible. 8 The investment per-formance of the SEA6 has not matched that of Chile, and there is, therefore, scope forpolicy and governance initiatives to improve investment performance.




12/22

In any accumulation fund, the relationship between real gross domestic product(GDP) growth, real wage growth, and the real rate of return credited to members isimpacts the replacement rate. The annual average relationship for Singapore for the1987–2011 period was as follows: the real growth was 7.9%; the real wage growth ratewas 5.0%; and the real rate of return credited to Central Provident Fund members was1.42% (Asher & Bali, 2013). For Malaysia, estimates suggest that annual real returnsaveraged 3.3% between 1961 and 2009, and the real rate of return uctuated between2.6% from 1961 to 1980; and 4.42% from 1980 to 1996. For the most recent decade,1999–2009, the return was 2.8%, which was lower than the average return earned since1961 (Asher & Bali, 2012; p. 59). For the 10-year period ending 2013, the annual realreturn generated by the Government Provident Fund of Thailand (assets equivalent to4.9% of GDP) was − 0.72% (a nominal rate of return of 4.79% and an ination rate of 5.51%). Since its inception in 1997, the real rate of return was 1.37%. 9

This implies a low replacement rate at retirement, and as any ination adjustment isabsent, and longevity risk is covered in a limited way, and with the use of commercialinsurance principles implying premiums for annuities varying by age and gender, thereplacement ration will exhibit a downward trend.

Similar estimates for other SEA6 countries are not available; there have, however,been measures taken by Malaysia and Thailand to take a more aggressive stance for theinvestment of pension assets, both in terms of diversication of assets to different risk categories and geographically by mandating fund managers to invest abroad. Such diver-sication has enhanced the risk prole. As the scal risks and contingent liabilities fall onthe scal budget, it is essential to estimate them and incorporate them in public nancialmanagement.

Pension funds of the Philippines are also considering initiatives along similar lines,though because of the social insurance principle employed, the accumulation of balancesis relatively low. Singapore’s Central Provident Fund (CPF) organization has turned overthe accumulated balances to Singapore’s Sovereign Wealth Fund, GIC Private Limited,which in turn invests globally. Reliable data on returns obtained by the fund are notmade publicly available.

Characteristics and limitations of nancial and capital markets impact on the invest-ment patterns and so does the political economy of the country as pensions assets aremainly in government organizations. Malaysia’s Employees’ Provident Fund (EPF) isconsidered too large for the size of domestic capital markets (EPF balances have grown

from 17.1% of GDP in 1980 to around more than 57% or $178 billion in 2013), and itspolitical economy encouraging the EPF to support the government’s infrastructure ini-tiatives (not always successfully) impacts on investment policies and performance. InIndonesia, pension funds, which are long-term savings, are primarily invested in short-term instruments such as bank deposits. Asset allocation of public pension funds in theSEA6 economies are presented in Table 5.

The payout of benets phase poses important investment challenges. The commer-cial annuity markets have not sufciently developed globally. Uncertainty about longev-ity trends poses additional risks for insurance companies providing annuities. While




13/22

some innovations in annuity markets are being attempted (such as longevity bonds androlling annuities), they are not likely to be of much relevance to SEA6 (Roy et al ., 2013).

The above suggests that the likely annuity organizers (and bearer of at least somerisk) would need to be government. Singapore’s annuity scheme (called CPF Life) thusis organized and administered by the CPF, national provident fund, though its desireto have the members bear future longevity and ination risks through lower payoutsadversely impacts on retirement adequacy. Investment policies and performance duringthe pay-out period will be of increasing relevance to other SEA6, particularly Malaysia.

4. Reform directions

4.1 Role of social pensionsThere is increasing recognition of the role of social pensions or noncontributory retire-ment income transfers that are nanced from budgetary sources in ensuring old-ageincome security. These pension schemes may be provided on a universal basis to allelderly, or be subject to means-testing (such as assets, income – at the individual orhousehold level). The aim of social pension is to mitigate old-age income poverty par-ticularly for the old-old. Since these transfers are nanced from budgetary sources, they allow for an element of risk-pooling in nancing retirement support for the benecia-

ries. However, with all expenditure programs, the scal resources required and theirequity effects must be integral to the analysis of social pensions. 10

The scal costs or total outlays required to nance social pensions are a function of the benet level, the number of beneciaries, and importantly, the administrative costsof the program. The benet level can be absolute or relative (i.e. indexed to per capitaincome or median wages). While universal schemes may have potentially lower adminis-trative and compliance costs, the total scal cost of the scheme will be much higher.However, making the social pension taxable beyond a certain income threshold willreduce these total costs.

Table 5 Asset allocation for pension and provident funds in SEA6

Country Equity BondsGovernmentsecurities Others

Indonesia 0–25% 0–25% 0–25% 0–25% (Deposits)Singapore (CPF) 75–100%Malaysia 0–25% 25–50% 25–50%Philippines 0–25% 25–50% 0–25% (Real estate and securities)Thailand (Social Security) < 10% < 10% 25–50% 0–25% (Commercial banks/state

enterprises)Vietnam 0–25% 50–75%

Note: Indonesian data are for the largest pool of pension assets (i.e. private pensions) and areassumed to be representative of all pension assets (see Guerard, 2011).




14/22

Table 6 provides an overview of the main features of social pension schemes in theselected countries of this study. All economies except Singapore have some element of social pensions in their nancing-mix. However, there is large variation in the number of beneciaries and the benet level. While the benet level (as share of average wage)is relatively low in Thailand, all individuals above age 60 receive the social pension. Indo-nesia’s social pension is approximately a fourth of wages, but less than 0.1% of the popu-lation above the age of 65 receives it. Singapore, as discussed in the earlier section onpension philosophy, does not have a social pension scheme. The data suggest that there isroom to develop social pensions as an integral component in the nancing-mix to supportretirement. Social pensions would provide an element of retirement security to those thathave not been able to participate in formal contributions-based social security programs.

How much would social pension schemes cost? Estimates in a recent paper onnancing social pensions in developing Asia (which included all economies in SEA6except Singapore) suggest that a universal social pension, covering everyone over age thatis indexed to per capita income (i.e. benet level is 15% of per capita income), wouldcost between 0.6% (Philippines) and 1.33% (Thailand) of GDP in 2010 and rapidly growing to between 1.4% (Indonesia) and 2.9% (Thailand) by 2030. 11 In advancedeconomies such as New Zealand, scal costs in 2009–2010 were 4.3% of GDP and areexpected to increase to 8% of GDP by 2050; in Australia, the scal costs of means-testedpensions were 2.7% of GDP in 2009, and are projected to be 3.9% of GDP in 2050(Bateman & Piggott, 2011).

4.2 ProfessionalismIt is imperative that the ve core functions of provident and pension funds identiedearlier must be done with greater professionalism than has been the case so far for theSEA6. This, in conjunction with strong regulation, would enable higher levels of pensionbenets from lower contribution rates than is the case now. The focus of these organiza-tions should be on providing benets to their members, which are commensurate withthe contribution rates while minimizing administrative, compliance, and transactionscosts.

Some of the SEA6 economies, such as Indonesia, have high administrative and com-pliance costs. They have not been able to undertake record keeping and managementinformation system tasks adequately, even for a relatively small proportion of the laborforce comprising formal sector workers. Their plans to sharply expand the coverage to

include informal sector workers under the SJSN Law may therefore be severely under-mined by their inadequate record-keeping capabilities.Investment policies and performance also remains a challenge in the SEA6. Limita-

tions of domestic nancial and capital markets, legal restrictions on international diver-sication (e.g. Malaysia, Indonesia), and relatively low importance given to duciary responsibilities (which require maximizing returns of provident and pension fund bal-ances for the benet of the members) have contributed to this outcome.

As prefunding arrangements, through retirement savings or accumulation of reserves, become increasingly common (pension assets are expected to grow signicantly




15/22

T a

b l e 6

N o n c o n t r i

b u t o r y p e n s i o n s c

h e m e s : m a i n

f e a t u r e s

( l a t e s t a v a i

l a b l e y e a r

)

D a t e

M e a n s t e s t

N a m e o f p r o g r a m

A g e

M o n t h l y b e n e

t

( a s s h a r e o f

a v e r a g e w a g e )

# o f R e c i p i e n t s

( a s s h a r e o f

p o p u

l a t i o n

a b o v e a g e 6 5 )

I n d o n e s i a

2 0 0 6

A s s e t , i

n c o m e , p e n s i o n s

E l d e r

l y S o c i a l S e c u r i t y P r o g r a m

6 0

3 2 . 0

( 2 3 . 2 )

1 3 , 2

5 0 ( 0

. 1 )

M a l a y s i a

1 9 8 2

I n c o m e , p e n s i o n s

E l d e r

l y A s s i s t a n c e S c h e m e

6 0

9 4 . 4

( 1 2 . 7 )

1 5 , 2

5 2 ( 8

. 0 )

P h i l i p p i n e s

2 0 1 1


S o c i a l P e n s i o n s S c h e m e

7 7

1 1 . 5

( 6 . 0

)

1 4 8 , 7 6 8 ( 4

. 0 )

S i n g a p o r e

N o t a p p l i c a b

l e

T h a i l a n d

1 9 9 3


O l d - a g e A l l o w a n c e

6 0

2 0 . 0

( 6 . 0

)

6 1 2 , 3 7 0 ( 1 0 0 % )

V i e t n a m

2 0 0 4


S o c i a l A s s i s t a n c e B e n e t

6 0 – 7

9

6 . 3 ( 4

. 8 )

8 0 8 , 7 7 3 ( 2

. 4 )

8 0

9 . 4 ( 7

. 1 )

1 3 9 , 3 3 8 ( 1 3 . 8 )

N o t e s : M o r e r e c e n t

d a t a f o r T

h a i l a n d s u g g e s t t h a t a l

l i n d i v i d u a l s a g e 6 0 o r m o r e a r e e l i g i b l e f o r a s o c i a l p e n s i o n . T h o s e a g e d

b e t w e e n 6 0 a n

d 6 9 r e c e i v e

6 0 0 b a h t ( U S D 1 9 ) , b e t w e e n 7 0 a n

d 7 9 r e c e i v e 7 0 0 b a h t

, 8 0 a n

d 8 9 r e c e i v e d 8 0 0 b a h t

, a n d a b o v e 9 0 r e c e i v e 1 0 0 0 b a h t

.

S o u r c e : I n t e r n a t i o n a l L a b o u r O r g a n i z a t i o n

( I L O ) ( 2 0 1 4 )

.




16/22

in Asian countries), development of domestic nancial and capital markets has becomeessential. Provident and pension funds will need to increasingly acquire competencies todeal with sophisticated investment strategies using diverse asset classes (e.g. debt, equity,real estate, and currencies) and diverse players (such as hedge funds, private equity inves-tors, and sovereign wealth funds).

Such sophisticated strategies, however, should not be attempted without adequatepreparation; and without understanding potential downside risks. In many low- andmiddle-income countries, it may be prudent to not fully attempt to obtain upside poten-tial from investments or from nancial innovations such as credit default swap (CDS) tominimize downside risks, and thereby limit scal risks and contingent liabilities on thestate.

It is also important to ensure that the investment strategies and policies adopted by pension organization recognize the institutional and operational context. For instance,the effectiveness of social insurance organized pension systems is reduced signicantly ineconomies that have a high share of the population employed in the informal sector(where formal employer–employee relationships do not exist).

4.3 Systemic perspectiveThere are three aspects of this perspective that needs to be addressed.

Complementary reformsThe rst aspect involves complimentary reforms in other areas such as labor markets,scal policy, and nancial and capital markets that are essential for effective social secu-rity reform.

Effective social security reform is greatly facilitated by sustainable macroeconomicpolicies that lead to high and stable growth whose benets are distributed widely. This isbecause the single most important variable for the economic security of both the youngand the old is the long-term trend of economic growth. The labor market regulationsand functioning must provide an appropriate balance between creating new jobs andpreserving existing jobs. High employment is negatively correlated with poverty, andtherefore creating economically viable and sustainable jobs is essential. Traditional insti-tutional retirement between 58 and 60 years in most economies in the SEA6 will need tobe gradually extended as individuals will need to rely on greater lifetime labor supply tonance their retirement expenditure. This will have to be done in a calibrated manner, as

it has wide-reaching economic implications, let alone impacting age-specic employ-ment rates.Civil service pension reform should form a part of scal policy reforms. These

should be based on the full cost (including unfunded liabilities) of pension (and health)benets being provided to the civil servants; and to improve the delivery of governmentservices (including social assistance or social pensions for the elderly). Without full andexplicit costing of civil service benets, it would be difcult to allocate society’s resourcesdevoted to the elderly equitably and efciently. In many countries, without civil servicepension (and health care) reforms too large a share of national income devoted to all




17/22

elderly will accrue to civil servants. This creates intragenerational inequities, and may strain social cohesion.

Financial and capital market reform is essential as the demand for quality invest-ments by provident and pension funds should be matched by the corresponding supply of nancial assets, based on both debt and equity. Unlocking the value of state enter-prises through partial or full divestments will be an important avenue in many Asiancountries for increasing the supply of such assets.

Developing a nancing-mix to access funding The second aspect of the systemic perspective concerns the nancing-mix of pensions.One approach to thinking about the various sources of nancing required is the WorldBank multi-pillars approach. This approach discusses the roles of public and privatesources of money, and how they could be accessed during retirement. The different tiersaim to provide a balance between social risk-pooling and individuals bearing investment,longevity, and other risks; between contributory and noncontributory schemes; and ex-ibility in managing and accessing retirement contributions or savings. While there may be theoretical limitations to this approach (Barr & Diamond, 2008), it can be adapted toemphasize complementary organizational approaches that governments can employ inmanaging retirement systems to suit specic contexts and objectives.

The public economics literature has widely recognized that any mandatory contri-bution to a social insurance or a savings program, even if it offers benets linked tocontributions, is equivalent to a tax. 12 What matters therefore from a scal imperativeis not the relative shares of public or private monies or share of contributory and non-contributory sources, but the total resources devoted to managing age-related expendi-ture. This suggests the need to differentiate between funding (i.e. the resources required)and nancing (the instruments used to access, allocate, and distribute) in pensionsystems.

The discussion in the preceding sections have argued that greater share of society’sresources, usually proxied by the share of GDP devoted to the pension system, will needto be devoted to managing retirement spending; that is, funding has to increase, and thishas to be achieved through a nancing-mix. Funding can also be improved if savingsin administrative and compliance costs are realized, or social insurance and providentfunds are able to improve their investment performance, and ultimately if there are sus-tained increases in national income.

Individuals commonly rely on multiple sources of nancing to meet their retirementexpenditure needs. The relative weight of each instrument in a nancing-mix varies fromcountry to country. In the SEA6, the Philippines, Thailand, and Vietnam have pension(and health) systems based on social insurance principles (though coverage of popula-tion in each country is far from being universal). However, it is important to note thatwhile Vietnam’s nancing-mix emphasizes the role of social insurance, due to weak agency functioning of the social insurance fund, it has accrued decits over the past few years that are covered by tax revenue. Malaysia has primarily relied on individual savings,which is also used for housing and health care.



https://www.researchgate.net/publication/227466905_Reforming_Pensions_Principles_and_Policy_Choices?el=1_x_8&enrichId=rgreq-f154c033-6ca1-4fd7-a51d-e581dbcac9cf&enrichSource=Y292ZXJQYWdlOzI4MDkxNzUwMDtBUzoyNzQ4OTYwNjQzNDgxNjFAMTQ0MjU1MTczNzQ2NQ==https://www.researchgate.net/publication/227466905_Reforming_Pensions_Principles_and_Policy_Choices?el=1_x_8&enrichId=rgreq-f154c033-6ca1-4fd7-a51d-e581dbcac9cf&enrichSource=Y292ZXJQYWdlOzI4MDkxNzUwMDtBUzoyNzQ4OTYwNjQzNDgxNjFAMTQ0MjU1MTczNzQ2NQ==


18/22

Different nancing instruments have different organizational and institutional pre-requisites, which must be taken into account when developing a nancing-mix.

For example, individual retirement accounts have been relatively easy to develop inSingapore and Malaysia, but Thailand and Vietnam have found it difcult to sustainthem on a mandatory or voluntary basis as the labor market is more formalized in Singa-pore and Malaysia, and administrative systems are more developed.

Social insurance requires high density of contributions that are channelized intonancial and capital markets through appropriate investment strategies and asset classesto enable the fund to be viable over the long-term. If contributions are entirely usedto invest in a particular asset class (such as government bonds) or used on a pay-as- you-go basis, it defeats the inherent advantages of social insurance over the tax systemas an instrument of nancing. The tax system can well be used to pool-risk and cross-subsidize individuals.

Similarly, provident funds require robust nancial and capital markets, includingannuity markets. Yet countries that rely on them have relatively less developed annuity markets such as Malaysia and Thailand. Since 2013, Singapore has introduced manda-tory annuitization once members reach their retirement age and are eligible to draw down their accumulated balances. The annuity, CPF Life, is organized by the govern-ment, whose premiums are set according to commercial and not social insurance prin-ciples. Thus, they vary by gender (with women being charged more) and age (premiumsincrease with age).

Across most instruments in the nancing-mix, one of the main constraints relate tolimited investment options to manage longevity and ination risks during the pay-outphase. Uncertainties about longevity trends are also a constraining factor, as these lead toconservative pricing of annuities, making them unattractive in comparison to otherinvestments (and in some cases unaffordable, creating adverse selection problems). Thepay-out phase requires greater attention of the policymakers in SEA6. Variations of phased withdrawals, with some social risk-pooling in the form of above-market interestrates, nanced from the budget merits consideration. Research on group annuities andsuch methods also merits serious consideration as alternatives to individual purchase of annuities.

Lower fertility rates, urbanization, changing values, and expectations of both the young and the old are signicantly increasing the need for more formal pension systemsin SEA6, consistent with the experiences of current industrialized countries who faced

these trends earlier.It is also essential to recognize the importance of personal savings, home ownership,investing in human capital, and opportunities for participating in livelihood activities inold age as integral elements of any pension system. If their importance is reected in tax,regulatory, and government expenditure allocation decisions, these can play a usefulsupplementary role in addressing pension challenges. In some countries with well-developed micro-nance institutions, micro-pensions could also play a useful role.

The weight of different instruments used in a nancing-mix reects to some extentdiffering historical legacies (eg. provident funds were established in Malaysia and




19/22

Singapore under the colonial rule) and differing political philosophies (eg. Singapore’saversion to societal risk-pooling to nance retirement expenditure). There will also bevariation in a nancing-mix across cohorts of retirees and even within a cohort group.Poterba (2014) notes the large variation is the sources of income during retirement incurrent retirees in the USA and how the variation has increased over the years. Thepolicy goal should not be to develop a particular nancing-mix a priori but to be ablerely on different instruments based on an individual’s nances and how a country’spension system is structured.

To the extent SEA6 governments will nd it difcult to improve investment returnson provident and pension funds additional resources will be required to fund retire-ment expenditure. This will have to come from individual and household savings,higher taxes, and reallocation of government expenditure. Over the past decade,there has been increasing recognition on the role of government-nanced transfersto individuals during retirement that can be provided universally or targeted to indi-viduals below certain income levels. These transfers can prevent individuals fromfalling into poverty during old age and are expected to play a greater role in thenancing-mix. While governments with limited scal space may be skeptical of devel-oping this instrument, effective targeting through household income testing as done inAustralia or by asset and income means tests as done in Vietnam will reduce the totalscal cost.

Effective coverage The third aspect of systemic perspective refers not only to strengthening sources of retirement income and improving replacement rates for retirees, but ensuring that thereal value of their pension income is not reduced by disproportionate consumption of any age-related consumption expenditure. This is true in most economies in the SEA6(except Thailand) where health-care spending by retirees is largely nanced out-of-pocket, effectively reducing the purchasing power of their pension to nance otherneeds.

Column 2 in Table 3 provides the share of health-care expenditure that is notnanced by out-of-pocket spending, and is lowest in Singapore (implying that 60% of health-care spending is out-of-pocket) and highest in Thailand (implying that only 15% of health-care spending is paid out-of-pocket). This is particularly striking given

the severe consequences of a lack of risk-pooling to meet health-care expenditure.The stochastic nature of the onset of health-care shocks and the uncertainty associatedwith the cost of treatment leaves individuals ill-prepared to nance them out-of-pocket, particularly in retirement. Therefore, health-care risks are an importantdimension of coverage and must be included in the range of risks covered duringretirement in a pension system. This aspect of systemic perspective has not beengiven adequate recognition. Despite achieving legal universal health coverage,out-of-pocket spending nances a third of spending in Malaysia and three-fths inSingapore.




20/22

5. Concluding remarks

The analysis in the paper has emphasized that country context-specic pension reformsare needed in the SEA6, informed by global and regional experiences. Given the hetero-geneity in pension systems, demographic and labour market proles, institutional andorganizational capacities even in a relatively conned geographical area in SEA6, inap-propriate generalizations of a particular pension systems or model should be avoided.

Any reform, whether in a tax system or a pension system, needs to improve trade-offsimplicit in the existing arrangements.

A major limitation faced by pension researchers in the SEA6 is the lack of reliable,timely, and disaggregated data on a consistent basis. Unless this gap is addressed, itwould be difcult to engage in rigorous but nuanced empirical evidence-based research.Regarding socioeconomic data, including on pensions, as a public good to be shared,rather than as a strategic or a tactical instrument, would help in advancing pensionresearch in the SEA6.

The paper suggests three main directions for pension reform in SEA6. The rst direc-tion is to enhance professionalism of the existing provident and pension fund organiza-tions, including their governance practices. The second is to strengthen the role of noncontributory budget-nanced pensions (e.g. social pensions). The third is to adopta systemic perspective to pension reform that includes reforms in complementary areas (labor markets, public nancial management practices, civil service pensionprograms); develop a nancing-mix and lastly, improve effective coverage by exploringcomplementarities between health care and pension programs.

Notes

1 We differentiate between funding and nancing . Funding, in the context of this paper, refers tothe proportion or the share of the total economic resources available to meet age-relatedspending. This will necessarily imply trade-offs with competing public and private expenditurepriorities. Financing refers to the various instruments or mechanisms through which resourcesare accessed or allocated. These include, for example, social insurance contributions, manda-tory and voluntary savings, general budgetary revenue, and family and community support.

2 The density of contributions is the number of contributions made by an individual to a retire-ment scheme divided by the maximum contributions possible during the economically activetime period.

3 A DC pension scheme or plan is one where its members are required to contribute in a denedmanner, but benets are left undened. Investment and macroeconomic risks are borne by individuals and not by the plan sponsor.

4 Labor force participation rate is the proportion of the population aged 15 and older who areeconomically active.

5 The replacement rate refers to the share of income during retirement from all sources (includ-ing personal savings, pension income, and social security) relative to an individual’s salary prior to retirement.

6 The country case studies, including those on the SEA6 countries, in Park (2013) discuss theseinstitutional arrangements in detail.




21/22

7 It is recognized that a replacement rate of about 60–70% from all sources of retirement incomeis considered to be an adequate pension benet during retirement. It is however more chal-lenging to dene an adequate benet level of health care. It is usually proxied by the share of health-care expenditure that is not nanced by an individual through out-of-pocket spending.

8 In Chile, between September 2002 and May 2014, the average annual rate of return credited to

members ranged between 4.14% for the most conservative portfolio and 6.95% for the mostrisky portfolio (Sojo, 2014).

9 Calculated from estimates available at URL: https://www.gpf.or.th/eng2012/invest_growth.asp(accessed October 30, 2014).

10 As those in the upper end of the income distribution, as a group, live longer than those in thelower end of the distribution, the net equity impacts of a universal social pension scheme may be regressive. This is however an empirical question that deserves greater scrutiny.

11 See Asher & Bali (2014; p. 77).12 Moreover, the emphasis should be on the ultimate economic incidence or burden of social

security contributions, rather than on statutory provisions of division of contributions by employers, employees, and the government.

References

Asher M.G. & Bali A.S. (2012). Malaysia. In: Park D. (ed.), Pension Systems in East and Southeast Asia: Promoting Fairness and Sustainability . Manila: Asian Development Bank, 53–65.

Asher M.G. & Bali A.S. (2013). Fairness and sustainability of pension arrangements in Singapore:An assessment. Malaysian Journal of Economic Studies, 50 (2), 175–191.

Asher M.G. & Bali A.S. (2013). Financing social protection in developing Asia: Issues and options. Journal of Southeast Asian Economies, 31 (1), 68–86.

Barr N. & Diamond P. (2008). Reforming Pensions: Principles and Policy Choices. Oxford and New York: Oxford University Press.

Bateman H. & Piggott J. (2011). Structuring the Payout Phase in a Dened Contribution (DC)Scheme in High Income Countries: Experiences of Australia and New Zealand . Jakarta: EconomicResearch Institute of ASEAN and East Asia (ERIA).

Bell F.C. & Miller M.L. (2005). Life Tables of US Social Security 1900–2100. US Social Security Administration. Accessed 1 January 2015. Available from URL: http://scholar.google.com.sg/scholar?q= life+ tables+ of + us+ social+ security + 2005&btnG = &hl= en&as_sdt= 0%2C5

Eggleston K.N. & Fuchs V.R. (2012). The new demographic transition: Most gains in life expec-tancy now realized late in life. Journal of Economic Perspectives, 26 (3), 137–156. doi: 10.1257/ jep.26.3.137

International Labour Organization (ILO) (2010). World Social Security Report 2010–2011: Provid-ing Coverage in Times of Crisis and Beyond . Geneva: ILO.

International Labour Organization (ILO) (2014). World Social Protection Report 2014–2015: Build-ing Economic Recovery; Inclusive Development and Social Justice. Geneva: ILO.

Park D. (2013). Pension Systems and Old-Age Income Support in East and Southeast Asia: Overview and Reform Directions. Manila: Asian Development Bank.

Poterba J.M. (2014). Retirement security in an aging society. Working Paper no. 19930, NationalBureau of Economic Research. Accessed 1 January 2015. Available from URL: http://www.nber.org/papers/w19930

Ramesh M. (2009). Economic crisis and its social impacts: Lessons from the 1997 Asian EconomicCrisis. Global Social Policy , 9 (1), 79–99.



https://www.gpf.or.th/eng2012/invest_growth.asphttp://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://www.nber.org/papers/w19930http://www.nber.org/papers/w19930http://www.nber.org/papers/w19930http://www.nber.org/papers/w19930http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5http://scholar.google.com.sg/scholar?q=life+tables+of+us+social+security+2005&btnG=&hl=en&as_sdt=0%252C5https://www.gpf.or.th/eng2012/invest_growth.asp


22/22

Ross S. (2011). Collection of social contributions: Current practice and critical issues. In:Takayama N. (ed.), Priority Challenges in Pension Administration. Tokyo: Maruzen, 3–26.

Roy A., Punhani S. & Hsiesh A. (2013). Demographic insights into policy: Asia’s big 3. Globaldemographics and pension research, Credit Suisse. Accessed December 2014. Available fromURL: https://plus.credit-suisse.com/researchplus/ravDocView?docid = mAdeuH

Sojo A. (2014). The Chilean System of Contributory Pensions as Locus of Rivalry and of a New Social Compact . Santiago: UNECLAC.

United Nations Department of Economic and Social Affairs (UNDESA) (2012). World populationprospects: The 2012 revision. Accessed 30 September 2014. Available from URL: http://esa.un.org/wpp/

Vaupel J.W. (2010). Biodemography of human ageing. Nature , 464 (7288), 536–542. doi: 10.1038/nature08984


https://plus.credit-suisse.com/researchplus/ravDocView?docid=mAdeuHhttps://plus.credit-suisse.com/researchplus/ravDocView?docid=mAdeuHhttps://plus.credit-suisse.com/researchplus/ravDocView?docid=mAdeuHhttp://esa.un.org/wpp/http://esa.un.org/wpp/http://esa.un.org/wpp/http://esa.un.org/wpp/https://plus.credit-suisse.com/researchplus/ravDocView?docid=mAdeuH

assesment of public pension system in sea - 2015-aper asher and bali

Documents