In May 2009, the government introduced the voluntary New Pension Scheme, with flexible contribution rules. An Indian citizen between the ages of 18 and 55 can be a member of this scheme, in principle. The minimum contribution required per year is Rs 6,000, but there is no ceiling. Under Section 66 (to replace the current Section 80C of the Income Tax Act) of the proposed Draft Direct Taxes Code, expected to be effective from financial year 2011, the aggregate deduction for investments made in savings instruments, including the voluntary NPS, is set to increase to a maximum of Rs 3 lakh from the current limit of Rs 1 lakh.
Enrolment in the voluntary NPS has been much lower than anticipated. By August 2009, it is estimated that only about 2,500 individuals joined the NPS. Given the demographic trends in India, where the share of working age to total population is increasing, and the number of those above 60 years is set to rise from 100 million in 2010 to 330 million in 2050, the negligible enrolment is a major cause for concern. Through their savings, individuals will need to bear a greater proportion of retirement financing than has traditionally been the case in India.
Making the voluntary NPS mandatory and more attractive should be a high priority for the government. There are several measures to achieve this goal, which merit serious consideration. These require that the PFRDA bill, 2009, be passed as expeditiously as possible before the end of 2009. Any delay will be rightly construed as lack of seriousness about making the NPS work.
Among the major perceived constraints of the NPS has been the uneven tax treatment of retirement savings as opposed to that for other products and organisations, such as the Employees’ Provident Fund (EPF) and the Public Provident Fund (PPF). The Draft Direct Taxes Code has, however, signalled a shift in taxation of retirement savings to provide a level playing field for different schemes (mandatory and voluntary) and providers. The proposed system is the EET, under which only withdrawals are taxed at marginal personal income-tax rates, provided they are not redeposited in approved savings intermediaries. Adequate transition arrangements have been provided.
It is proposed that from 2011, any withdrawals from retirement savings will be taxed at the applicable marginal income-tax rate. This will reduce the incentives for pre-retirement withdrawals and for lump-sum withdrawals at retirement, while increasing income flows for the elderly.
However, some concerns about the taxation of the NPS remain, and need to be addressed. Greater clarity is required in the Draft Code on whether the second tier of the mandatory NPS or the preretirement withdrawals from the voluntary NPS will be subject to personal income tax. The logic of the EET suggests that they should be taxed, but this will reduce the attractiveness of the NPS. Greater equality can be attained if the pre-retirement withdrawals of the EPFO are taxed similarly.
One way to address this issue would be to specify both the purpose and the proportion of balances that can be withdrawn before retirement. The cumulative proportion of permissible withdrawals could be kept at around 20% of the total amount.
Another concern involves the treatment of commutation benefits under the old civil service pension scheme, the NPS (both voluntary and mandatory), and other private and public sector pension schemes. Would such withdrawals be subject to normal marginal personal income-tax rates? If yes, then the tax burdens for those opting for these will be high. If it is not, then there will be gaps in the EET treatment of retirement savings.
To address this issue, it may be useful to consider the following. At the time of statutory withdrawal, a one-time lump-sum withdrawal (periodically adjusted for inflation and income levels) to meet social and other obligations could be permitted without attracting personal income tax. This could be combined with a phased or programmed withdrawal option to be offered by approved entities. The PFRDA should encourage the development of such options.
An individual would then be able to turn the accumulated retirement savings into periodic payments (say every quarter) in such a way that the principal and interest are exhausted over a period of 15-20 years. If a person dies, the balance can accrue to the nominee. There will thus be no insurance and insurance risk-pooling under the phased withdrawal, but some social risk-pooling through interest rate subsidies could be incorporated.
The purchase of annuities, which is currently mandatory under the NPS, can then be made voluntary. Even in countries with well-developed financial and capital markets, the use of annuities is not widespread. Even if annuities are mandated, pricing could be an issue as the insurance companies providing them make overly conservative assumptions concerning longevity and investment returns.
The introduction of the Draft Direct Taxes Code provides an opportunity to reform some aspects of the EPF scheme. The plan currently permits unlimited contributions, which are exempt from personal income tax. This provision needs an urgent review. An option worth considering is to make it mandatory for all workers in establishments with 20 or more employees (to be gradually reduced to 10 as the EPFO acquires greater administrative and technological capabilities) to be covered up to a specified monthly wage ceiling. For instance, Rs 20,000 per month can be automatically adjusted every three years in accordance with the average nominal wage increase for the organised sector workers. This will eliminate the current anomaly, wherein those who begin their employment with a monthly salary of Rs 6,500 are excluded, although many continue to contribute to receive tax benefits and a politically negotiated rate of interest.
No contribution above the mandatory wage ceiling should be accepted by the EPFO, which will allow it to focus on being a service provider. The exempt funds should, after approval from the PFRDA, become part of the mandatory NPS scheme. This will provide better scale to the NPS, while addressing the anomaly of the EPFO being both a service provider and regulator of exempt funds.
Finally, the PFRDA’s budget for publicising the NPS and its architecture needs to be increased substantially from the current Rs 10 crore. At the same time, the PFRDA needs to show greater dynamism and leadership to organise different stakeholders and communicate better the need for longterm retirement savings, particularly by those who are not in formal employer-employee relationships. The PFRDA should also sponsor greater empirical evidence-based pension policies and product designs, while permitting the expansion of the distributional network for the NPS.
Mukul G. Asher is Professor, Lee Kuan Yew School of Public Policy, NUS.