Time to Mainstream Micro-Pensions in India Mukul G. Asher* Savita Shankar** The paper has argued that to expand coverage of micro-pensions, social entrepreneurship (along with social responsibility) will be needed by the financial sector, including the MFIs, insurance companies and mutual funds. Such entrepreneurship should aim to drastically reduce overall transaction costs, including fund management costs. The micro-pensions sector requires a regulator. While the PFRDA as an overall pension regulator is an appropriate agency for regulating micro-pensions, But it must closely co-ordinate with EEBI, IRDA, RBI, and NABARD. *Professor of Public Policy, National University of Singapore. Email:
[email protected] . **Freelance Researcher. Email:
[email protected]
I Introduction There is increasing recognition that to meet the needs of 21st century India, the reform of India’s complex social security system comprising seven components (Figure 1) should proceed along three broad directions [Asher 2006]. First, provident and pension fund organisations in all seven components need to substantially enhance the professionalism with which they perform the core functions. Ross (2004) has identified five core functions. The first involves reliable collection of contributions, taxes and other receipts, including any loan payments; The second concerns payment of benefits for each of the schemes in a timely and correct way. The third involves securing financial management and productive investment of provident and pension fund assets. The fourth core function is maintaining an effective communication network, including development of accurate data and record keeping mechanisms to support collection, payment and financial activities. The fifth is production of financial statements and reports that are tied to providing effective and reliable governance, fiduciary responsibility, transparency, and accountability.
eSS Working Paper/Social Security/Pension System May 2007
1
Figure 1: India’s Social Security System From January 1, 2004, all newly recruited civil servants at the Centre (except for armed forces) are on a DC scheme. 16 states have also issued notification for a shift to a DC scheme, but their starting dates vary.
MFs Unions SHGs, Coops
Abbreviations Used DB Defined Benefit DC Defined Contribution EDLI Employees’ Deposit Linked Insurance Scheme EPF Employees’ Provident Fund EPS Employees’ Pension Scheme GPF Government Provident Fund GS Gratuity Scheme Coops Cooperatives CSPS Civil Service Pension Scheme MFs Mutual Funds NGO Non-Government Organizations SHG Self Help Groups
Source: Asher (2007a)
eSS Working Paper/Social Security/Pension System May 2007
2
The second factor concerns systemic perspective. It requires that all components of the social security system are well integrated in meeting the retirement needs of the population. This is consistent with obtaining retirement income security from multiple sources. The role of each component and the schemes within that component should be periodically reviewed to assess their suitability, scalability and sustainability. The third is the need to enhance financial literacy, particularly concerning long time horizons, need for empirical data based price discovery and tyranny of small numbers relating to insurance and pensions among both policymakers and the general public. Any pension arrangement is a financial contract spanning five to six decades. It should therefore command a high level of trust and confidence from the participants. Expanding the population with access to financial services i.e. financial inclusion is one of the priorities of the Reserve Bank of India. The provident and pension fund organizations and the pension regulatory structures should promote financial literacy and financial inclusion. The prevailing tendency of regarding pensions as a welfare measure without due regard for empirical evidence and rigorous economic, financial and fiscal analysis represents a formidable constraint in pension reform. This will require a mindset change among policymakers and governmental organization from current focus on administration to public policy and management will also be essential [Asher 2007b]. Demographic Trends and Ageing of India’s Population India is experiencing a demographic transition leading to lower total fertility rate, increased life expectancy and greater proportion of the aged in the population. The share of the elderly or persons aged 65 years and above in India’s population is expected to rise from 4.6 per cent in 2000 to 9 per cent in 2030. In absolute terms, the number of those above the age of 60 will rise from 87.5 million in 2005 to 100.8 million in 2010. By 2030 this number is projected to be around 200 million, which will increase further to 330 million by 2050. This suggests that lack of pension coverage will affect more persons than ever before. The share of elderly as a proportion of working age population namely the old age dependency ratio, will also increase. Increased life expectancy means that each elderly person will require support for a longer period. In 2005, an Indian man reaching 60 years of age can on an average be eSS Working Paper/Social Security/Pension System May 2007
3
expected to live for another 17 years and an average woman 18 years. This is expected to increase. Uncertainty about the pace of increase has major implications for annuity markets and for health insurance. Both the level and the pace of ageing are likely to provide significant challenges as even by 2030, India’s per capita income will be relatively low. If the current normal retirement age of 60 is not increased, then the challenge of financing the elderly will be even greater. As women are increasingly living longer than men; greater feminization of the population is taking place [Liebig and Rajan, 2003]. Providing secure and sustainable retirement income therefore ranks high in the social priorities. Presently, only about a fifth of India’s 460 million strong labour force and about a quarter of its 90 million elderly benefit from at least one of the components of India’s social security system. It is in the above context that this paper examines the role of micro-pensions in India. This article argues that as one of the seven components of India’s social security system, micro-pensions can play a limited but useful role if they are brought into the mainstream of the financial system. The paper is organized as follows. The nature and design of micro-pensions in India is reviewed in section II. A brief discussion of channels for expanding coverage of micro-pensions is provided in Section III. This is followed by a brief discussion of reform directions in Section IV. The final section provides concluding observations.
II Nature and Design of Micro-pensions A pension scheme may be of defined contribution (DC) or defined benefit (DB) nature. In a DC scheme, while contributions are explicitly defined, benefits are not. This is because the risks associated with investing them, and then converting them to a regular retirement income stream are borne by the individual members. In contrast, in defined benefit (DB) scheme, the benefits to be provided are explicitly stated, while contributions are left undefined. In a DB scheme, it is the plan sponsor, which bears the investment, mortality, and other risks. In any defined contribution (DC) pension scheme, there is an accumulation phase as well as a pay-out phase eSS Working Paper/Social Security/Pension System May 2007
4
Figure 2: Accumulation and Payout Phases of DC Schemes Figure 1 Accumulation and payout phases of DC schemes Cumulative Balances ($)
(Insert Figure 2 here)
During the accumulation phase, a member of a DC plan or an annuity purchaser1 Accumulation Phase
WorkingWorking-phase
Pay-out phase
Withdrawal Age
Retirement Period
Cumulative Balances = Net contributions (contributions minus withdrawals), withdrawals), plus interest credited on accumulated balances. Payout phase: phase: the funds accumulated can be spent rapidly or slowly. Death may may occur before the funds are exhausted or reverse is also a possibility. So need to protect against against longevity risk. As it is the purchasing power of the funds that is relevant, protection against the inflation inflation risk is also desirable. Source: Source: Author
contributes towards accumulating balances. The value of such accumulation During the accumulation phase, a member of a DC plan or an annuity purchaser1 contributes towards accumulating balances. The value of such accumulation depends on the amount of contributions (for a DC plan, covered wage level times the contribution rate) less pre-retirement withdrawals plus returns (net of investment management expenses) obtained from the investment of funds less applicable taxes. It is usual for administrative expenses to be borne by the members collectively. These however need to be transparent and benchmarked. A micro-pension scheme is typically designed as a defined contribution scheme. The scheme essentially operates on the principle of voluntary savings accumulated over a eSS Working Paper/Social Security/Pension System May 2007
5
long period. These savings are intermediated through financial and capital markets by a professional fund manager. At an agreed upon withdrawal age (usually 58 or 60 years) the accumulated balances can be withdrawn in a lumpsum, a phased withdrawal, annuity or some combination of these methods. Micro-pension Schemes in India Micro-pensions operate as a part of the broader framework of micro-finance institutions (MFIs). In any country, membership in micro-pensions is likely to be a fraction of the membership of MFIs. The key proximate objective is therefore to increase this fraction in a financially sustainable manner. In India, the insurance companies and two mutual funds have been offering pension plans for individuals for a considerable period. However only one mutual fund, UTI AMC (Asset Management Company), which is in the public sector began offering plans in 2006, which can be loosely termed as micro-pensions. As of April 1, 2007, there is therefore only one fund manager in the micro-pensions space, though it has teamed with a variety of third-party organizations. The key characteristic of the UTI AMC micro-pension plans are small sum (ranging from Rs. 50 to Rs. 200 per month); flexibility in payments (monthly or yearly contributions are not mandatory), and presence of a third party, such as cooperative, self help group (SHG) or an NGO.2 The non-mandatory nature is an important departure from the traditional pension plans. The role of the third party is to be a locus for generating a large number of members with common characteristics; undertake certain administrative functions; and act as channel of communication. They can therefore assist in reducing transactions costs involved in identifying members, in contributions, and in record-keeping. The third party must therefore command trust and confidence of the members as well as be competent in administration and financial matters. Moreover both the trust and the competence must be sustained over a long period. Increasing supply of such third parties is therefore critical to expanding the reach of micro-pensions.
eSS Working Paper/Social Security/Pension System May 2007
6
The role played by the partners is in collection of the monthly contributions from individual members. These savings are pooled and transferred to UTI for funds management. Each member receives a unique account number. The first micro-pension scheme with UTI AMC as the fund manager was launched nearly a year ago. Since then, UTI AMC has partnered with the Self Employed Women’s Association (SEWA), a micro finance institution based in Ahmedabad; COMPFED, a federation formed by milk producers in Bihar; Paradip Port Trust; and an urban cooperative bank run by women. More such partnerships are expected. A strategic tie-up between Bank of India (BOI) and UTI Mutual Fund for providing members of self-help groups an investment opportunity through a micropension initiative could help address the issue of relatively short life-span of most SGHs. But its impact on transaction costs and financial viability need to be empirically determined. The schemes involve contributions ranging from Rs. 50 to Rs. 200 per month per member, but they need not be made every month. The contributions must typically be made until age 55 and the pension payments begin after age 58. A maximum of 40 per cent of the corpus can be invested in equity and the balance invested in debt. Actual investment in equity is however around 20 per cent. The target appears to be an annual return of 10 to 12 per cent after all expenses. It is too early to evaluate the extent to which the target is likely to be achieved. For managing micro-pension funds, UTI AMC has made only minor adjustments in their usual charges levied for managing regular pension funds. The management fees range from 1.75 per cent to 2.5 per cent of assets depending on the assets under management. Unlike in the case of regular pension plans, there is no entry load. But exit load of 1 per cent is levied if amounts are withdrawn before the agreed upon retirement age. The exit load appears to be harsh given the fact that economic and financial uncertainties and liquidity and credit access constraints loom large for the members of micro-pension schemes. Perhaps zero exit load after five to eight year membership could be considered. The combined membership of the micro-pension schemes is difficult to estimate precisely. But knowledgeable observers put it at around 0.2 million. In contrast, in micro eSS Working Paper/Social Security/Pension System May 2007
7
finance schemes membership is around 35 million. This may be regarded as a proxy indicator of the potential coverage of micro pension schemes in India. The costs charged in the accumulation phase by the UTI AMC have been noted earlier. There are however, no empirical studies available as yet to estimate such costs as a percentage of contributions and of assets and their behaviour over time.. The transaction costs in the case of group micro-credit providers in India however range between 3.2 per cent and 11.3 per cent of the loans provided, depending on the geographical area [Shankar, 2006]. Payout Phase In the payout phase (usually coinciding with retirement3), longevity, investment and inflation risks need to be addressed. In addition, survivors’ benefits and disability insurance are also essential. This is particularly the case in India as life-time labor force participation of women is relatively low, and even when they do participate, women as a group earn less than men, and therefore require more years of support. Uneven property rights and social status of women, particularly of widows, are additional reasons for the need to provide survivors’ insurance benefits. More systematic consideration of these issues, informed by international experiences, and analytical insights, is needed. The longevity risk concerns the fact that while each person is certain to die, the age, the cause, and the place of death are not known. Some may die within a short period after retirement; while others may live for a much longer period. The latter category of persons may find their financial resources exhausted, while those dying early in retirement may not face this challenge The earlier the age at which final withdrawal is permitted, the longer the period for which the accumulated balances will be required to be used to finance old-age. In 2001, in India, average men and women at age 60 had life expectancy of 16 years and 17 years respectively. This implies that accumulated balance in provident fund at age 55 will on an average need to last for 21 years for men, and 22 years for women. For some groups, such as civil servants, the life expectancy is much higher4. Moreover, the dispersion around the mean life expectancy should also be taken into consideration.
eSS Working Paper/Social Security/Pension System May 2007
8
India’s demographic trend therefore suggests that there will be considerable lengthening of the payout phase. Regional and group variations will however be considerable. The structure of the payout phase in the UTI AMC’s micro-pension plans appears to have received less attention. A phased or systematic withdrawal plan would involve periodic withdrawal after retirement until the amounts are exhausted. Till then, the amount will continue to be invested as in the accumulation phase. This pattern may be too risky given the volatility in the capital markets. Moreover, there are many different ways to structure phased withdrawals. Greater attention to this design issue is warranted, including the costs to the members of different types of options. In savings-based micro-pension schemes, investment, macroeconomic and other risks are borne by the individual. Risk-sharing arrangements have therefore been often advocated. Some have argued for co-contributions by the government at the accumulation stage for participants in micro-pensions. Other options include risk sharing by the society (through government) at the payout phase. These could be in the form of special bonds or bank deposits with higher interest rates for senior citizens (with a cap on total investment) that vary according to market interest rates on long-term government bonds. This can be combined with well-targeted and reasonably funded old age assistance schemes financed from general budgetary revenue. More research is needed in the Indian context before designing risk-sharing options. In particular, political economy considerations, where populist policies often trump financial and economic sustainability, should play an important role in design of such options. It would appear that co-contributions by the government at the accumulation stage may be particularly vulnerable to dysfunctional populist policies. The issue of who should bear the costs of the services of the third parties needs to be considered. If full costs are charged to members, micro-pension schemes may become financially unviable. Rigorous research efforts are needed to separate accounting and financial costs on the one hand from economic costs, which involve costing all the resources, used in delivering micro-pension services. Once these costs are identified, then
eSS Working Paper/Social Security/Pension System May 2007
9
cost-sharing arrangements among different stakeholders in both accumulation and payout phases can be further investigated.
III Channels for Expanding Coverage This section focuses on potential channels for increasing coverage of micro-pensions. The role of micro finance institutions is of particular importance in this regard [Hermes and Lensink, 2007]. The core activity of most micro-finance institutions is providing micro-credit. Micro-credit loans essentially are short term in nature and range between one year and three years. While repeat loans are often observed, the time horizon is usually not comparable with those of micro-pension schemes which have a much longer time horizon. Hence there is likely to be timing mismatch which means that the two services may not overlap entirely and to that extent there may not be cost savings. Group loans account for 93 per cent of the micro finance in India (as per Sa-dhan, Industry association of Community Development Finance Institutions in India’s website www.sa-dhan.net). A strong case can be made that if the potential of micro-finance is to be realized, loans should be on an individual basis and savers and borrowers need to be separated. Record keeping in the case of group loans is usually on a group level. Micropensions schemes will however require individual record keeping. Individual members may fear that when the provider of micro-credit is also the administrator of the micro-pension schemes, at the time of default, the savings under the schemes may be adversely affected. This may reduce the attractiveness of the scheme. There is therefore a need to separate record keeping and other routine administration, fund management and custodian function of assets. The linkage of the scheme with a commercial bank such as the collaboration of UTI AMC with Bank of India SHG members may give further credibility to the scheme, but may raise costs. Only established micro finance institutions having the reputation required to market a scheme involving payouts over a long period of time can effectively market microeSS Working Paper/Social Security/Pension System May 2007
10
pension schemes. While customers may be willing to avail micro-credit from less established institutions, they may not be willing to place long term savings with them. However, in spite of these challenges, the case for using micro finance institutions who have passed due diligence process for micro pension services is strong. These institutions have incurred the establishment costs in various locations, not all of them are easily accessible. Many of them also have experienced personnel who can with some basic training market and administer these schemes. Other Channels for Expanding Coverage The other channel through which micro-pensions could be spread is by use of the post office network. Post offices have established ready presence in remote areas and hence this presence could be used to service micro-pension schemes. Being part of the Government machinery, post offices enjoy high degree of credibility. While post office personnel are experienced in handling small savings they need to be trained to service pension schemes. Moreover the post office may need to offer more flexibility in terms of deposit timings and even perhaps doorstep collection services to ensure regularity of savings. An experienced asset management company should handle funds management. Reverse mortgage arrangements to turn housing equity into retirement consumption schemes are another channel that can be considered. The 2007-08 budget has proposed that the National Housing Bank consider such an instrument. It appears that if the variety of channels sketched above to expand the coverage of micro-pensions is utilized, there is a scope for enrolling around five million members over the next few years. This is not an insignificant number in the context of total pension coverage in India.
eSS Working Paper/Social Security/Pension System May 2007
11
IV Reform Directions The analysis in the previous section has demonstrated that the micro-pensions sector is in the early stages of development in India. Relatively minor adjustments to standard individual pension plans been made in offering micro-pension products. The empirical basis of financial and economic sustainability of the schemes as well as risk management practices benefiting the members is still weak. As Prahalad (2005) has persuasively argued, in offering standard middle and high income products to those at the “bottom of the pyramid”, social entrepreneurship (along with social responsibility) which radically reduces transaction costs and results in genuine resource cost savings without compromising on quality is essential. This is the direction which micro-pension reforms should take. In addition to design changes concerning exit load suggested earlier, the fund management fees will need to be substantially lowered. Aggarwal (2007) noted that the Coal Mines Provident Fund Organisation (CMPFO) recently conducted an auction to pick fund managers for assets of Rs.20,000 crores (USD 4.6 billion). The lowest bids were for 0.01 percent of assets under management far lower than the 1.75 to 2.25 percent charged by the UTI AMF for micro-pensions. Admittedly, the two costs are not directly comparable. Nevertheless, they suggest possibilities for reducing costs which need to be encouraged. However, these will not come about automatically. The financial industry in general and micro-finance institutions in particular will need to be innovative. Minor modifications to existing schemes, while useful as a beginning, would clearly be insufficient to meet the needs. Micro-pensions represent a long-term financial contract with potential for significant agency problems and systemic risk to the financial system. Due to the sensitivity of the population, the contingent liability is on the Government.
eSS Working Paper/Social Security/Pension System May 2007
12
There is therefore a strong case for regulation on micro-pensions. The tabling of the bill of the Microfinancial Sector (Development and Regulation) Bill 2007 provides for regulation of the microfinance sector by the NABARD. This is a positive step, which will result in mainstreaming of this sector. Similar regulation of micro-pensions is required. It is suggested that Pension Fund Regulatory and Development Authority (PFRDA) when it is formally functioning should consider forming a separate division for micro-pensions.5 The division should work closely with SEBI, IRDA and NABARD so as to prevent regulatory arbitrage and bring coherence to the micro-pensions sector. The regulator should also play a development role by promoting professionalism and financial literacy. Concluding Remarks The current micro-pension arrangements in India are based on relatively minor modifications of the existing pension schemes. Not only the number of fund managers is limited, but the design of micro-pensions is also relatively rigid. The paper has argued that to expand coverage of micro-pensions, social entrepreneurship (along with social responsibility) will be needed by the financial sector, including the MFIs, insurance companies and mutual funds. Such entrepreneurship should aim to drastically reduce overall transaction costs, including fund management costs. The paper has also argued that micro-pensions sector requires a regulator. The PFRDA as an overall pension regulator is an appropriate agency for regulating micropensions. But it must closely co-ordinate with EEBI, IRDA, RBI, and NABARD. The PFRDA should have a separate wing for micro-pensions in recognition of the special requirements. This wing should help develop appropriate products and delivery mechanisms, help instill confidence, promote professionalism and encourage financial literacy and inclusion. Research on micro-pensions is in its infancy. Implications of alternative designs and delivery systems for micro-pensions need to be rigorously researched for financial sustainability and for their impact on the level of benefits to members. Different risksharing arrangements among all the stakeholders also need to be explored. The National Insurance Academy (NIA), NABARD (National Bank for Agriculture and Rural eSS Working Paper/Social Security/Pension System May 2007
13
Development), university academics and industry researchers are urged to ensure that further progress of micro-pensions is based on sound empirical and analytical foundations. There is a potential for micro-pensions to play a limited but useful role as an integral component of India’s social security system. [Thanks are due to Amarendu Nandy for useful comments. The usual caveat applies.] References Aggarwal, A. (2007):’Don’t Compare Mutual Funds with Pension Plans’, Business Standard, March 31. Asher M. (2007a), “Pension Reform and Old Age Grants in India” Paper presented at the conference on ‘Innovative Approaches to Social Security in India, Brazil and South Africa’, University of Johannesburg, Asher M. (2007b): ‘India must breed more policy wonks’ Far Eastern Economic Review Vol.170 No.2 pp.30-33 Asher, M. (2006): ‘ Pension Issues and Challenges Facing India’, Economic and Political Weekly. November 11,2006. pp. 4638-4641 Asher, M. and Nandy, A. (2006): ‘Issues and Options in the Pay-out Phase in Defined Contribution Schemes’, Pravartak , Volume 1, Issue, pp. 1-7. Hermes, N. and Lensink, R. (2007): ‘Impact of Micro finance: A Critical Survey’ Economic and Political Weekly February 10, pp.462-465 Liebig, P.S. and S. Irudaya Rajan (ed.): (2003), ‘An Aging India: Perspectives, Prospects and Policies,’ USA: Haworth Press. Prahalad, C.K. (2005): The Fortune at the Bottom of the Pyramid: Eradicating Poverty through Profits. Wharton School of Publishing, New Jersey. Ross, S.G. (2004): ‘Collection of social contributions: Current practice and Critical Issues. Paper presented at the International Conference on Changes in the Structure and Organization of Social Security Administration”, Cracow, Poland, 3-4 June. Shankar, S. (2006): “Transaction Costs in Group Micro Credit in India: Case Studies of Three Micro Finance Institutions” Working Paper, Institute for Financial Management and Research Shapiro, D. and Streiff, T.F. (2004): Annuities, 4th edition, Dearborn: Dearborn Financial Publishing, Inc.
Notes 1
An annuity provides for a systematic liquidation of a capital sum. It is distinguished by a variety of features such as how the funds are invested, when annuitized payments are scheduled to begin, and how they are paid for [Shapiro and Streiff, 2004, p. 2]. The mortality risk is the main risk faced by the life insurer, while investment risk is the main risk in provision of annuities. The increased life expectancy therefore reduces life insurance costs, but increases the annuities cost, i.e. lower annuity benefits for a given capital sum. The risk profile of those voluntarily seeking annuities increases if there is significant adverse selection problem, i.e. those who are especially likely to live longer constitute disproportionately large proportion of the demand for annuities. 2
If the third party uses traditional agent model to obtain members, costs will be higher, impacting the viability of micro-pensions. 3
The age at which final withdrawal is permitted may or may not coincide with the retirement age. In the Employees Provident Fund (EPF) scheme, administered by the Employees Provident Fund Organization (EPFO), the withdrawal age is 55 years, but the retirement age in India is generally higher, ranging from 58 to 60 years. In the pension plans of insurance companies, such as by HDFC Standard Life Insurance, the age at which vesting occurs (i.e. the pay-out phase begins) may vary between the age of 50 and 70 years. eSS Working Paper/Social Security/Pension System May 2007
14
4
According to LIC (1996-98) Occupational Pensioners Mortality, the life expectancy at age 60 was 22.5 years for all occupational pensioners.
5
The PFRDA Bill has been introduced in Parliament since 2004. But the left parties whose claim to speak on behalf of ordinary workers is dubious, have been using their temporary political clout to block the will of the majority by blocking the passage of the PFRDA Bill.
eSS Working Paper/Social Security/Pension System May 2007
15
