Pension Reforms in India: Myth, Reality and Policy Choices

Escalating costs of the pension system is forcing the Indian Government to reevaluate the formal programmes that provide social security to employees. The government has so far received three official reports (namely, OASIS, IRDA and Bhattacharya), which have examined the issue and suggested several measures to provide a safety net to the aging population. This paper examines the recommendations made in these reports and analyses the potential effects of them. It is organized around five policy questions: 1. Should the reformed system create individual (funded defined-contribution) accounts, or should it remain a single collective fund with a defined-benefit formula? The changeover involves a larger public policy choice issue: who should ultimately bear the risk? Should employees/retirees shoulder those risks alone arising from variations in asset yields and unexpected changes in longevity, or should these risks be shared more broadly across participants, if not society? Choice would depend upon to which group the individual belongs. Financially successful people may believe in individual ownership and choice, while low wage earners may want assured returns because they do not have other resources to fall back upon. Unfortunately most Indians, unlike those in many other countries, are in the latter category which cannot bear any risk, more so in the old age. 2. If individual accounts are adopted, should the reformed system move toward private and decentralized collection of contributions, management of investments, and payment of annuities, or should these functions be administered by a public agency? In privately managed funds, associated problems would be intermediation costs, agency problem (principal-agent fiduciary relationship), and greatly increased costs to administer the plan. Several studies across the world have shown that periodic fee may look deceptively low but, over longer time horizons, the cumulative effect can be dramatic, sometimes reducing the benefits by 30 to 50 per cent. 3. Should fund managers of retirement savings be allowed to invest in a diversified portfolio that includes stocks and private bonds? In recent years equity investments, particularly index investing, have become a favoured strategy. Index funds are subject to tracking error, and being loaded with few big stocks, there are much higher risks in index investing than people perceive. Over the period, real annual return on index funds may be more, but people retire only once. Equity markets are highly volatile and go through long periods of feasts and famine. Guarantees would have to be provided in the form of minimum return or providing minimum basic pension on retirement. World bank studies show that government ends up acquiring conjectural liabilities wherever a pension system based on private providers is mandated. How would that be different from the present system where a government agency (EPFO) provides retirement benefits? 4. Should the government move toward advance funding of its pension obligations for its employees, or should these obligations continue to be financed on pay-as-you-go basis? Studies have shown that a simultaneous implementation of funded, diversified, individual accounts is not a "free lunch" once you properly account for existing unfounded obligations and risk. The Bhattacharya Committees estimates show that the government would have to pay out more on account of pensions to its employees for the next 38 years before the new scheme starts showing reduced government expenditure. These amounts do not include the tax foregone by the government on the employees contribution. Several assumptions have been made about the scheme, which the committee hopes would remain valid and that the future governments would behave responsibly. The proposed scheme does not consider intermediation costs and agency risks; in fact, the committee presumes that agents would behave more responsibly than principals. 5. What should be the level of government fiscal support in the form of tax subsidy, foregone tax collections, grants, administrative costs incurred by its agencies, and level of assumed contingent liabilities in case the government guarantees minimum pension? The crucial question is: how much and to whom is this subsidy accruing? Are beneficiaries of the proposed system the ones who need subsidy? Tax treatment of pension is a critical policy choice. A generous tax treatment may promote savings but may be costly in terms of revenue foregone. Apparently, an exercise in balancing is necessary. The priority should, therefore, be putting in place a policy vision and road map with specific goals in relation to pre-determined milestones. These should include a tax financed and means-tested system for lower income groups. If government cannot afford it, then it has no moral or political justification to even consider providing further tax benefits to privileged income groups. If there are no government funds for the first pillar in the World Bank recommended multipillar system, the third pillar should remain out of policy discussions. Emphasis should be on strengthening the second pillar. Suggested reforms neither enhance efficiency nor make the social security system more equitable. It would only privatize the gains while costs and risk for the government would increase considerably. It would only help well-off segment of society in availing more tax concessions. Present problem in the government pension system is due to successive governments behaving like Santa Clauses ignoring the cost to exchequer. Fund managers would not be able to solve these problems. Specific fiscal and other measures for implementing a feasible and viable pension system in Indian conditions have also been suggested in the paper.