39149 Pension Reform and the Development of Pension Systems: An Evaluation of World Bank Assistance Background Paper Regional Summary: Asia John Piggott Director-General, Independent Evaluation Group: Vinod Thomas Director, Independent Evaluation Group, World Bank: Ajay Chhibber Senior Manager: Ali Khadr Task Manager: Emily Andrews 2007 This paper is available upon request from IEG. The World Bank Washington, D.C. ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Independent Evaluation Group is an independent unit within the World Bank Group; it reports directly to the Bank's Board of Executive Directors. IEG assesses what works, and what does not; how a borrower plans to run and maintain a project; and the lasting contribution of the Bank to a country's overall development. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank's work, and to provide accountability in the achievement of its objectives. It also improves Bank work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. IEG Working Papers are an informal series to disseminate the findings of work in progress to encourage the exchange of ideas about development effectiveness through evaluation. The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot 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 on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Contact: Knowledge Programs and Evaluation Capacity Development Group (IEGKE) e-mail: eline@worldbank.org Telephone: 202-458-4497 Facsimile: 202-522-3125 http:/www.worldbank.org/ieg Acronyms and Abbreviations AAA Analytical and Advisory Services ADB Asian Development Bank EET Exempt, Exempt Taxable ESW Economic and Sector Work FSU Former Soviet Union IEG Independent Evaluation Group (formerly Operations Evaluation Department) IMF International Monetary Fund OED Operations Evaluation Department (changed it name to IEG in December 2005) NDC Notionally Defined Contribution NPS National Pension Scheme PAYG Pay-as-you-go PROST Pension Reform Options Stimulation Toolkit SOE State-Owned Enterprise Content Preface........................................................................................................................... i 1. Introduction and Purpose of Overview............................................................. 1 2. Economic and Demographic Summary ............................................................ 2 3. Formally Evaluated Countries ­ China and Republic of South Korea ......... 3 China......................................................................................................................3 South Korea ...........................................................................................................5 4. World Bank Work Elsewhere in Asia............................................................... 7 East Asia ................................................................................................................7 South Asia ..............................................................................................................7 5. Perspectives on Bank Penetration and Activity............................................... 9 Reference ................................................................................................................... 11 Tables Table 1: Demographic and Economic Summary of Countries in East and South Asia with World Bank Pension Reform Work ..........................................................................................2 Table 2: Summary of World Bank Pension Reform Activity in South Asia............................8 i Preface This paper belongs to series of 19 country and regional case studies commissioned as background research for the World Bank's Independent Evaluation Group (IEG) report "Pension Reform and the Development of Pension Systems." The findings are based on consultant missions to the country or region, interviews with government, Bank, donor, and private sector representatives involved in the pension reform, and analysis of relevant Bank and external documents. This paper was authored by John Piggott in 2004. John Piggott is Professor of Economics and Director, Centre for Pensions and Superannuation (CPS), University of New South Wales, Sydney, Australia. He would like to thank, without implicating, Robert Palacios, and especially Yvonne Sin and Xiaoqing Yu, for their assistance in preparing this summary. Responsibility for the document is the author's alone. ii 1 1. Introduction and Purpose of Overview 1.1 Since the early 1990s, and especially following the publication of the World Bank volume Averting the Old Age Crisis (1994), the World Bank and its personnel have been actively involved in offering advice and policy recommendations on pension reform throughout the developing world. Important regions of influence have been Eastern Europe, including the former Soviet Union (FSU) nations, Latin America, and to a more limited extent, Africa. Activity in the Asian region has been much more limited, even though two countries in the Asian region, China (in East Asia) and India (in South Asia) account for approximately 40 percent of the world's population over the age of 65. This ratio is projected to rise in coming decades, as the trend to increasing longevity spreads from richer nations. 1.2 The Bank's pension reform activities in China and South Korea have already been reviewed by the Independent Evaluation Group (IEG). In the sense that it is these countries in Asia that have received the most focused pension reform assistance from the World Bank in the last decade, this is appropriate. However, as a sample of the potential of World Bank efforts elsewhere in the Asian region, this selection leaves major gaps. The countries of South Asia have, on average, less developed financial sectors; they have somewhat higher fertility; and several were at one time or another subject to British rule, and have inherited comprehensive Civil Service pensions which have survived to the present day. 2 2. Economic and Demographic Summary 2.1 The major countries in East Asia where the Bank has been active are China and South Korea. A summary of the detailed IEG assessments will be offered in the following section. In addition, there has been limited work in Indonesia, Malaysia, the Philippines and Thailand, and discussion is underway in Mongolia and Vietnam. In South Asia there has been more widespread Bank activity, but it is on the whole of more recent vintage. In most cases, pension reform work in South Asia has taken place in this millennium. Countries which have sought help, and to which there has been Bank response, include Afghanistan, Bhutan, India, Nepal, Pakistan, and Sri Lanka. 2.2 Table 1 tabulates essential information about the demographic and economic status of these countries. In each region, there are about 1.6 billion people living in countries which have been involved in some form of pension reform dialogue with the World Bank. Excluding South Korea as an outlier, countries in East Asia enjoy per capita incomes double on average of those in South Asia. Aged dependency ratios are on average 50 percent higher in East Asia. By 2030 the weighted average aged dependency ratios will be 6.8 percent in the South Asia countries listed, and 11.3 percent in East Asia. 2.3 In terms of sheer size, India accounts for almost two-thirds of the South Asia states with which the Bank has a pension reform dialogue, and well over half the population of all countries not covered by the earlier in-depth IEG assessments. For this reason, it will be emphasised in the summary of non-assessed countries later in this report. Table 1: Demographic and Economic Summary of Countries in East and South Asia with World Bank Pension Reform Work Country Population Population Aged Projected GDP GDP per cap (Millions) 65+ Dependency Dependency (USD (US dollars, Ratio ratio (2030) Billion, 2003) 2003) Afghanistan 28.5 2% 4.5% 6% N.A. N.A. Bhutan 2.2 4% 7.1% 8% 0.6 295.1 China 1,298.8 8% 10.7% 24% 1,409.9 1,085.5 India 1,065.1 5% 7.6% 14% 599.0 562.4 Korea (South) 48.6 8% 11.5% 32% 605.3 12,455.8 Malaysia 23.5 5% 7.2% 16% 103.2 4,385.6 Nepal 27.1 4% 6.3% 8% 5.8 215.5 Pakistan 159.2 4% 7.3% 8% 67.9 426.5 Philippines 86.2 4% 6.5% 12% 80.6 934.3 Sri Lanka 19.9 7% 10.3% 23% 18.5 929.6 Thailand 64.9 7% 10.6% 20% 143.2 2,207.1 Sources: CIA Factbook; World Bank; UN Population Statistics. 3 3. Formally Evaluated Countries ­ China and Republic of South Korea China 3.1 World Bank advice to China on pension reform began in the early 1990s, and gathered momentum in the mid-90s. It has therefore been on-going through the two major State Council Circulars (26 and 42) which have mapped Chinese pension reform in the last decade. The overall impression is that of a delicate relationship, in which neither party is inclined to insist too heavily on its own agenda. Its size and importance gives China options in sources of assistance that are not available to many of the Bank's clients, and there is no adjustment lending. For these reasons there is an expectation that Bank influence on Chinese policy thinking might be limited. 3.2 China has a surprisingly fragmented decision-making framework, where provinces and counties, as well as a number of central government authorities, stake claims to influence. China is therefore in a somewhat different position from many of the Bank's other clients. 3.3 The World Bank became involved in pension reform in China as a consequence of the adoption of market-oriented resource allocation, a process begun in the late-1970s. China's social security has evolved from the State-Owned Enterprise (SOE) welfare support. An early Bank initiative, broader than but embracing pensions, focused on separating SOEs from their welfare responsibilities, with the overall goal of SOE reform. This project was concerned with the implementation issues involved in the separation of various social support mechanisms from SOE employment contracts ­ housing, health, and social security. The move towards the separation of social security from the SOE had begun as far back as 1986, when the State Council ordered separate labour contracts for all SOE and Commune Owned Enterprise employees, with pension entitlements separately accounted for. By the early-90s, labour market mobility had made the old defined benefit SOE arrangements impractical. 3.4 More specifically directed at pension reform was a 1995 World Bank Mission which visited China to study pension policy. The resulting report was highly influential. It analysed the systems then operating in China, which were extremely fragmented, and called for a unified national system. It proposed a three-tiered structure comprising a social pooling dimension, a mandatory individual account and a voluntary contribution account. Rates were established, and transition arrangements specified, and their influence can be seen in current national policy parameters. 3.5 Chinese authorities appreciated the importance of information flows in rendering a centralized social security system operational, and this has led to the Jin Bao (Golden Protection) project. This is currently under way, with support from a number of agencies, including the Bank. 3.6 The World Bank has played an important role in China's pension reform in general. From the intellectual approach of the unpublished 1995 report ,"Reforming Enterprise-based Social Service in China" to the current Liaoning pilot project, the strategy proposed by the World Bank and the actual policies issued by the Chinese government form a coherent and well argued policy and implementation strategy. 4 Circulars 26 and 42, and the most recent Enterprise Annuity Document No. 20 all indicate that China is moving towards the three-pillar system proposed by the Bank. 3.7 Finally, a range of AAA initiatives have provided a systematic approach to guide Chinese officials through a rigorous methodology of pension analysis. World Bank staff have worked tirelessly to make their work more influential and acceptable. The PROST workshops in particular have reached out to the provincial level, and have complemented the intellectual framework advocated in the Bank's other work. It seems clear that the PROST methodology, whether applied in a funded or notional environment, pushed the Chinese government to think beyond current practice and get ready for tomorrow's reform. 3.8 Related to this is the question of whether the Bank was too doctrinaire in its promotion of a funded second pillar. Although the 1995 report did mention that the proposed pension reform would not be feasible unless coverage is extended to include the non-state sector, and the real rate of return on pension funds is, over the long term, at least equal to the rate of increase in real wages, it did not suggest PAYG or Notional Individual Accounts as alternative approaches. Regulation of the financial sector was emphasized, but the question of the maturity of the finance sector was simply ignored. The combination of high wage growth and low investment return has been an outstanding issue for about a decade, and still no substantial research has been done by the Bank into pension design options other than funded individual accounts. 3.9 There is little current impetus for Bank-funded pension reform work in China. Yet major issues remain for China's pension system. The actual transition to a funded second pillar has only just started and the regulation for enterprise annuity (the third pillar) has only just been issued. Fund management and capital market development, and the regulatory structure being built to facilitate them, are pivotal issues whose resolution remains uncertain. These are the crucial elements for a successful funded pillar as promoted by the Bank in its earlier interaction with the Chinese authorities. China's current high real wage growth, and comparatively low official interest rate, makes the development of a financially sensible funded pillar a serious challenge. 3.10 Equally important for pension reform is the pattern of regional labor migration to urban areas. Urbanization of the labor force in the richer provinces, which have the most flexibility in pension reform, has been on-going for some time. Various provinces are exploiting this phenomenon to expand social security coverage to young newly urbanized workers, thereby finding a means to finance current social security liabilities. Yet almost no work has been done on the implications of this strategy, including the future liabilities such a course of action entails. 3.11 There is a real danger that the Bank's good work to date will not be converted into a worthwhile final outcome because of the lack of funds to support relevant projects. An agenda for future work might include an investigation of the linkage between the funded pillar and financial sector development, the question of expanding coverage, particularly to the rural sector, urbanization and new farmer citizens, and the potential role of notional defined contribution (NDC) structures in China. 5 South Korea 3.12 The World Bank's involvement in pension reform in South Korea stemmed from the Asian crisis of 1997, and the financial crisis that engulfed South Korea in that period. In 1990, South Korea's per capita income had passed the World Bank guidelines for graduation from Bank lending and in 1991 South Korea and the World Bank agreed on a four-year graduation plan. Over this period World Bank funding was limited to small projects in areas of mutual interest. In 1994 the World Bank field office in Seoul closed. 3.13 However, in 1997, a financial crisis spread across Asia, and by late 1997 the Korean economy faced severe difficulties. The exchange rate had depreciated by 50 percent against the US dollar, real interest rates rose to around 30 percent, and the government was having difficulties financing an external debt of US$120 billion, of which around 50 percent was short-term debt and US$20 billion was due by the end of that year. In November 1997, the South Korea Government formally asked the IMF/World Bank for assistance. 3.14 At that time the World Bank was not operating in Korea. But using existing staff expertise, an Economic Reconstruction Loan was rapidly developed in late 1997. It was broadly agreed that Korea's economic problems did not stem from poor macroeconomic fundamentals, but from structural deficiencies in the financial and corporate sectors. 3.15 The World Bank agreed to recommence lending to Korea with up to US$10 billion to support structural reforms. This led to Bank assistance in the area of pension reform as part of three structural adjustments loans (one in 1997 and two in 1998), as well as technical assistance and policy advice (in the form of an analytical pension reform paper prepared by the World Bank). 3.16 The World Bank assistance was both demand and supply driven. It was demand driven in the sense that the South Korean government had made the initial request for assistance, and supply driven to the extent that the specific pension reforms required were developed by Bank (in conjunction with the South Korean government). It is emphasized, however, that pension reform was only a small part of the structural reforms required under the structural adjustment loans. 3.17 An overview of the required pension reforms indicates some influence of Averting the Old Age Crisis. Both support consideration of the multi-pillar approach based on a balance of public and private pensions. The basic strategy towards pension reform was as a component of broader structural reforms. It had been agreed that South Korea's macroeconomic and financial problems resulted from structural deficiencies in the financial and corporate sector and in relation to labor markets and the social safety net. As a result, the focus of the pension reforms was long-term structural change. However, the structural adjustment lending also recognized the immediate impact of the Asian financial crisis on the current aged. 3.18 Technical assistance included the preparation of an analytical report, which was officially released after the third structural adjustment loan had closed, but the analytical work of earlier drafts had been used to design the pension components of the reforms required as part of the structural adjustment lending. 6 3.19 To address poverty among the aged, a non-contributory pension was introduced for the current elderly poor in 1999. Structural difficulties were partly overcome. Five-yearly actuarial projections are now conducted for all public pension schemes, most recently for the National Pension Scheme (NPS) in 2003 and for the special public pensions in 2000. A phased reduction of mandatory investment of NPS reserves in public sector assets was achieved over the period 1999 to 2002 and now stands at 14.5 percent (2003). Various parametric reforms to the NPS have been undertaken, along with the establishment of uniform EET taxation of all retirement savings. Moreover, consideration and debate of pension reform proposals has been ongoing since that time within the bureaucracy, and among academics and interest groups, and government committees are currently investigating a range of issues including management of NPS reserves, restructuring the NPS to ensure universal coverage, portability between all public pension schemes and the introduction of multi-pillar pension provision. Systemic pension reforms are now supported by many policymakers and academics, although there is less support from employer organizations and little support from labor unions. In terms of individuals, public pension coverage has almost doubled to 80 percent since the mid 1990s, and as a result of parametric reforms the financial stability of all public pension schemes has improved. 3.20 From an institutional development perspective, the largest impact has been to greatly improve the capacity of South Koreans to formulate their own pension policies and to manage their public pension schemes. Most important here was the establishment of a Pension Reform Task Force with broad representation across government ministries, public pension corporations, research institutes, labour unions and employer organizations. This exposed a cross-section of Korean pension experts to international best practice and provided training in modern analytical techniques. The success of this strategy is evidenced by continued analytical review of pension policies by academics, research institutes, public pension corporations, and interest groups such as labor unions and employer groups. As a result of the Bank assistance, Korean pension experts are more aware of the issues associated with the governance, management and financial stability of their public pension schemes, and sustainability of the reforms seems more likely as a result. 7 4. World Bank Work Elsewhere in Asia East Asia 4.1 Pension reform work by the World Bank in the East Asia region has been dominated by the countries whose IEG assessments were summarized in the previous section. Limited work has been carried out in Thailand and the Philippines, but in neither case has the scale of assistance reached a point where it could be reasonably assessed. 4.2 In Thailand, a dialogue has continued for several years. In 1998, a Social Security scheme was introduced, on an unfunded, replacement rate basis. A seminar was held there in 2000, in cooperation with the Asian Development Bank (ADB). The World Bank was subsequently approached about the design of a funded second pillar. However, the Bank indicated it would be prepared to be involved only if the approach to reform was more holistic, and discussions ceased. 4.3 The Philippines have a Social Security scheme, a Civil Service scheme (the GSIS), and a defense services scheme. There has been dialogue between the Philippine Government and the Bank at various stages throughout the 1990s about pension reform, but governance structures were so fragile that little constructive reform was possible. There is no current activity. South Asia 4.4 In contrast, the World Bank has been involved in pension reform activity with a range of countries in South Asia. These have comprised ESW and TA work, rather than major projects. Overall, in South Asia, the Asian crisis, followed by a realization of high debt associated with civil service pensions drove countries to seek assistance. Because most of these countries require loans, there is leverage to encourage pension reform. On the whole, before such assistance was sought, there was no activity in pension reform in South Asia. 4.5 The most important is India, not only because of its large population, but also because the reform efforts there appear embedded in a forward-looking structure. In common with most other countries in this region seeking Bank assistance, the motivation behind India's call was the realization that large and continuing liabilities would result from the civil service pension. 4.6 The first World Bank overview on pension reform in India was started in late- 98, and presented to the government in 2001. The eventual outcome was OASIS, a proposal to reform the Civil Service pension to be a contributory defined contribution scheme. The OASIS structure was put forward as a framework for pension saving in the non-government sector, but more recently, a pension regulatory commission has been established. Income tax breaks are envisaged, but such a small proportion of private-sector workers pay income tax that this will not be an effective incentive for most. It will perhaps encourage the self-employed to save for retirement. Current work includes a three-state investigation into establishing a social pension, and a joint World Bank /ADB conference on pension reform is planned. This last development suggests the possibility of cooperative support between the Bank and the ADB in South Asia, which would be a resource saving innovation. 8 4.7 The South Asia countries that are at least peripherally involved in pension reform activity with World Bank assistance are listed in Table 2, along with an indication of the focus of activity. The importance of the fiscal driver of a burgeoning civil service pension debt is clear. Table 2: Summary of World Bank Pension Reform Activity in South Asia Country Nature/focus of pension reform assistance India Civil Service pension reform, social pension design, regulatory structure, second pillar establishment Pakistan Civil Service pension reform; 3rd pillar advice on legal structures with the Securities Exchange Commission Maldives Civil Service pension reform and social pension advice Indonesia Dialogue only Sri Lanka Civil Service under reform, but still Defined Benefit. New entrants now contribute. First grant for pension regulatory reform Nepal Civil Service pension reform, plus a demogrant design Bhutan IDF grant for pension work Afghanistan Post war Civil Service Pension Reform 9 5. Perspectives on Bank Penetration and Activity 5.1 Overall, the experience of China and South Korea, the two countries in Asia selected for in-depth assessment, are not typical of the interactions between Asian nations and the World Bank in the area of pension reform. In most countries where the Bank has been active, especially in South Asia, fiscal imperatives stemming from Civil Service pension liabilities, rather than a sense of social protection, or even financial sector sophistication, drove countries to seek assistance. These perceptions were brought to a head by the Asian financial crisis of the late-1990s, and pension reform activity did not, in most cases, begin until after 2000. 5.2 Opportunities may well have been lost to help these nations with both pension reform and financial market development in the mid to late-1990s. Such activity would have led to earlier action being taken in the face of demographic transition, and may also have shored some financial sector institutions against the upcoming financial crisis. 5.3 Nevertheless, now that countries in the region are seeking assistance on pension reform, the Bank has responded by strengthening its presence in the region institutionally. In the last two years, a Social Protection Sector Manager has been appointed for South Asia, along with two lead specialists based in the region. (In East Asia, Social protection still falls under the Health Sector Manager's portfolio.) This partially reflects the perception that there is less opportunity in some East Asian countries to build impetus for reform, because there is less lending potential. 5.4 Asia would seem to be an ideal opportunity to develop workable protocols for cooperation with other international institutions in providing assistance to countries in the region. Specifically, the ADB would appear to provide the potential for fruitful partnership. Such intentions have been voiced at senior levels, but actual co-operation has been limited. In addition to the examples given above, in 2000 India hosted a combined ADB/ World Bank capacity building conference in New Delhi, focused on pension reform, and there are plans for a 2005 joint conference on pension reform. But there are no examples of joint project work, even though there would appear to be complementary skill bases in the two organizations which could be productively exploited. Institutional and turf barriers remain a constraint, despite lip service to the contrary. This definitely requires attention. 10 11 Reference Bateman, Hazel. 2006. Pension Reform and the Development of Pension Systems: An Evaluation of the World Bank Assistance: South Korea Country Study. IEG Working Paper. World Bank, Washington, D.C. Piggott, John and Lu Bei. 2006. Pension Reform and the Development of Pension Systems: An Evaluation of the World Bank Assistance: China Country Study. IEG Working Paper. World Bank, Washington, D.C. Conversations and discussions with Robert Palacios and Yvonne Sin.