Pension reforms in India in the last decade have seen three major initiatives - a paradigmatic shift in the civil servants' pension scheme, the National Pension Scheme for all citizens and the New Pension System Lite for the economically disadvantaged sections with small savings. The NaPS has seen a lukewarm response so far, with a majority of subscribers being central and state government employees, for whom the scheme is mandatory. An analysis of the auto choice option under the NaPS and the demographic transition reveals potential future imbalances in the investment structure among the asset classes. Moreover, the NaPS does not even guarantee a minimum pension, thus defeating its "welfare" orientation.

COMMENTARY

National Pension Scheme: For Whose Beneﬁt?

Ayanendu Sanyal, K Gayithri, S Erappa

Pension reforms in India in the last decade have seen three major initiatives – a paradigmatic shift in the civil servants’ pension scheme, the National Pension Scheme for all citizens and the New Pension System Lite for the economically disadvantaged sections with small savings. The NaPS has seen a lukewarm response so far, with a majority of subscribers being central and state government employees, for whom the scheme is mandatory. An analysis of the auto choice option under the NaPS and the demographic transition reveals potential future imbalances in the investment structure among the asset classes. Moreover, the NaPS does not even guarantee a minimum pension, thus defeating its “welfare” orientation.

Ayanendu Sanyal (ayanendu@isec.ac.in), K Gayithri (gayithri@isec.ac.in) and S Erappa (erappa@isec.ac.in) are at the Institute for Social and Economic Change, Bangalore.

Economic & Political Weekly

february 19, 2011

he design and implementation of the deﬁned contribution-based pension system, popularly known as the New Pension System (NPS) has had three important initiatives till date. The ﬁrst was a paradigmatic shift in the Civil Service Pension System (CSPS) from a deﬁned beneﬁt (DB) system to a deﬁned contribution1 (DC) system launched in 2004. This reform aimed at containing the ﬁscal liability arising on account of the DB scheme for civil servants. The second m ajor reform, launched on 1 May 2009, extended the NPS to all citizens, so as to widen the coverage of the pension system. Hereafter we refer to this as the National Pension Scheme (NaPS) to distinguish it from the civil servant-related pension reform. The second tier of NPS and NaPS (voluntary contribution and withdrawal) was also opened on 1 December 2009 for those possessing a Tier I account. The third major initiative relates to the launch of NPS-Lite on 1 April 2010 for the econo mically backward classes,2 as the NaPS was not quite successful in increasing the coverage.

The need for pension reform in India arises from the low coverage of the existing old age security programmes, demographic trends and ﬁscal constraints (Swarup 2008). Another important aspect in the context of economic development relates to the creation of pension wealth, and with our current demographic proﬁle, we can certainly reap substantial wealth gains provided more and more people are motivated to get into the fold at an early age. The NPS architecture consists of a Central Recordkeeping Agency and competing pension fund managers along with the NPS trust, custodian, Trustee bank, r etirement advisers. This system has evolved based on the study of experiences and systems in other countries (ibid).

After almost a decade of pension reforms, it would be appropriate to reﬂect on the achievements of the new initiatives in terms of the response of the people to the new pension schemes and other reform issues. After a brief discussion on the broad features of the

vol xlvi no 8

three new initiatives, we examine the response so far to the NPS, and some proposals to incentivise better participation. The article also addresses some aspects of investment procedure and risk management.

Response to Pension Schemes

DC schemes can be effectively implemented in a country if two essential conditions are satisﬁed (i) rudimentary capital markets, and (ii) considerable government regulations to prevent frauds and excessive risks (James 1997). With functioning capital markets attracting foreign capital, and a regulatory structure in place, India did launch a DC scheme in 2004, though several issues remain unresolved, as we will show a little later in the article.

Some of the positive features in the NaPS and NPS-Lite include a mandatory private pillar (Tier I) (with a contribution of Rs 1,000 from the government up to 2013-14) with a voluntary pillar (Tier II), an equal contribution to the mandatory private pillar for the civil servants, auto choice3 fund option to enable novices with inadequate knowledge and experience in capital market investment choices, and the selection of three fund managers (for civil servants) and six fund managers (for all) under strict regulations. Nevertheless, the pension systems are getting more complicated with several parallel kinds of pension schemes operating within the country. The entrants into civil service after 1 January 2004 are under the NPS, but the old DB system continues for those who had joined service before 2004 and for the armed forces which account for 50% of the central government liability. In quick s uccession there have been two new schemes, the NaPS and the NPS-Lite. This is likely to further complicate the already complex system.

One of the important objectives of pension reforms in India and the framing of the pension architecture is to enhance the coverage of old age security. We examine the extent of current coverage since the launch of the NPS by looking at the response to the NPS in three different categories – the NPS civil servants, the NaPS and the NPS-Lite (Figure 1, p 18). The data reveals that a large number of subscribers are employees of the central and state government for whom the scheme is mandatory. In the ﬁrst year, only 21,807 workers subscribed to the NaPS.

COMMENTARY

Figure 1: Number of Subscribers under Alternative New Pension Schemes (Number of Accounts) 1400000

1200000

1000000

800000

600000

400000

200000

0 NPS (Civil Servants) NaPS (All Indian Citizens NPS-Lite Total Excluding Civil Servants) (Economically Backward Classes) The NPS numbers include accounts created for the new entrants of states and central government mandatorily from their respective dates of introduction to the dates mentioned in the figure. The NaPS includes subscribers (not under economically backward classes) since 1 May 2009 to the dates mentioned. The NPS-Lite numbers include the pension accounts of economically backward classes from 1 April 2010 to the above-mentioned dates. Source: PFRDA (2010, 2011).

Figure 2: Proposed Auto-Choice Structure

(% of Funds in Different Asset Categories) 90

0

18 20 22 24 26 28 3032 34 36 3840 42 44 46 48 5052 54 56 5860 Age Source: NPS offer document.

The introduction of the contributory pension of Rs 1,000 was guaranteed (proscheme was primarily to widen the cover-vided the annuity payment is more than that age of the plan so that majority of the pop-amount). The Pension Fund Regulatory and ulation comes under this plan in future. Development Autho rity (PFRDA) has identi-However, it is clear from Figure 1 that in ﬁed “aggregators”5 for the collection and one year (1 May 2009 to 30 July 2010) transmission of funds from the target popuonly 0.005% of the working population4 lation to the NPS fund managers. Some has joined the NaPS voluntarily and the amount of media campaigning is underway,6 others have joined NPS by default as per and a revision of the minimum pension is the design of the scheme. A paltry 3,317 also presently under consideration. The ini-Tier II accounts have been opened be-tial response to NPS-Lite was not encouragtween D ecember 2009 and November ing, with only 1,476 a ccounts till 20 Novem2010, which reﬂects the actual response of ber 2010. Nevertheless, there has been a workers towards future savings. It can be steep rise after that taking the numbers to concluded that the response to the NaPS 17,306. As on 17 January 2011, 0.016% of has been lukewarm in the very ﬁrst year. the working population had joined the NaPS

The government responded promptly by or NPS-Lite voluntarily and it is interesting giving incentives both to the points of pres-that NPS-Lite is growing at a rapid pace. The ence (POPs) and the account holders. It plausible reason could be that it is a scheme declared a contribution of Rs 1,000 (up to pushed (and sponsored) by the government, 2013-14) to the account holders and enha-and with a minimum pension guarantee. nced the incentive of POPs to Rs 150 (from The number of Tier II a ccounts, which is the Rs 50) for each account. Further, the mini-key indicator of response to future savings, mum contribution in NPS-Lite was reduced has increased to 4,189 only. to as low as Rs 100 per month and if paid up Although it seems that NPS-Lite has been to the retirement fully then a minimum successful at least in terms of coverage, the performance of the NaPS as regards widening of coverage is not adequate. The coverage can increase if some incentives/disincentives are incorporated in this structure. These might be in the form of employers’ contribution (like employees’ provident fund (EPF) or General Pension Fund (GPF) for formal sector), guaranteeing minimum pension after retirement for all the schemes, moderate disincentive rates and loans against pension fund to a certain limit (like loans against provident funds). In order to reach the goal of “pension for all”, the government will have to introduce “beveridgean pension” for all the elderly (a scheme similar to the “zero pillar” of the World Bank). This is to f ormally include the section of the population who are too poor to afford future savings. The funding procedure can be decided by the government taking into account the future economic and demographic scenario.

Investment Procedure and Risk Management

The auto-choice investment option conceived of under the NPS needs a careful reexamination. This option is structured on an age-based risk taking behaviour, framed to serve the purpose of those with little or no ﬁnancial knowledge. It has been observed from other countries’ experiences that once an auto-choice option is available, maximum number of subscribers opt for it (Sadhak 2009). This creates an imbalance for the availability of funds with the demographic transition of a country (if the auto-choice is adjusted with age as an exogenous factor). The auto-choice option under the NaPS is presented in Figure 2. Further, in order to get a clearer picture of the demographic transition the age distribution of people from the 2001 census is depicted in Table 1.

Now, if the current NaPS is successful, huge funds will be available for equity (E) compared to credit risk bearing ﬁxed instruments (C) and government securities (G). This might create an imbalance in the market. Further, as the demographic transition takes place, the share will increase in the upper cohorts leading to a larger availability of

Table 1: Age Distribution of the Working Population

(2001 Census)

Age	Population in Millions	Share in the Working Age (%)
20-34	247.4	51
35-49	173.8	36
50-59	64.2	13

Source: Socio-economic Tables, Census of India 2001.

february 19, 2011 vol xlvi no 8

Economic & Political Weekly

COMMENTARY

funds under C and G. This will happen assuming that a majority of the working population would opt for auto-choice either due to lack of proper ﬁnancial knowledge or simply convenience. It is likely that the low yielding asset class C and G funds investment would be plenty resulting in an excess supply of loanable funds in future. For the present, it looks like huge funds will be available in asset class E which may also reduce the return (assuming excess supply). Thus the person undergoing this transition will suffer the most. One might argue that the NaPS has given good returns in 2009-10, with 26% in equities, 11% in corporate bonds and 5% in government bonds with a simple average of 12%. However, the returns are high as the coverage currently is low, but with the significant increase in coverage, the returns can be low due to the coexistence of the auto-choice option and demo graphic transition.

The preference towards government bonds is also likely to create more problems for the investment structure. Bonds typically give lower returns on investment. Despite this, the risk-averse subscribers (which are deﬁnitely the majority) are likely to invest in the asset class G and C which bear less risk compared to E. This could drive good investments out. Further this preference may lead to the selection of those fund managers which have larger government shareholdings such as State Bank of India and Unit Trust of India. Also subscribers might be attracted towards higher returns when a fund manager gains by investing in highly risky assets.7 Thus there is a need to look at the NaPS’ investment structure so that the subscriber gets an assured return on retirement.

The risk management system in NaPS and NPS-Lite is not group-based but individual-based. If the fund is group-based the risk gets divided among the group members. On the other hand decisionmaking, which is a problem in a groupbased model, can be taken care of by the fund manager. Nevertheless, the formation of such a group should be on some homogeneous criteria (same company, same income, etc). This is required in order to minimise conﬂicts in the group.

Minimum Pension

Pension was ﬁrst introduced in order to promote welfare among the aged workers.

Economic & Political Weekly

february 19, 2011

Ironically, the “welfare” connotation of the NaPS seems highly questionable as it does not even guarantee a minimum pension. The absence of this provision will amount to a great deal of welfare loss to the investors under the NaPS. From the point of view of the government, the scheme would deﬁnitely help reduce the ﬁscal liability, but from the workers’ perspective, the scheme does not serve its purpose at all since the entire risk is borne by them. Further, the returns are largely determined by the behaviour of the annuity markets and the economy in general and are therefore largely undesirable in the context of pension payments. Chile has taken care of this problem by introducing a guaranteed pension (which is predetermined by the government) in case of a market failure. Although the government is allowing subscribers to shift from NaPS to NPS-Lite (provided they are eligible for it) there may be a case in which the NaPS subscriber gets no pension due to failure of markets but a contributor in NPS-Lite gets a minimum pension of Rs 1,000

– a scenario obviously not justiﬁed.

Conclusion

The poor response to the NaPS can be partly attributed to the uncertainty regarding the absence of assured pension. In order to deal with this, the introduction of a deﬁned beneﬁt component is strongly recommended. Further the investment procedure should not be individual-based but group-based and thus it is suggested that a centralised authority should take care of investments.

The NPS/NaPS/NPS-Lite together are setting a playﬁeld to replace the EPF. However the EPF (with 45 million subscribers and a corpus of Rs 2,620 billion) which currently offers a lower rate of return (8.5%) than the NaPS, guarantees a pension (although it might be low). Thus the contributory scheme needs to be revamped thoroughly before it replaces the EPF. One might mention NPS-Lite here, but this is only for the poor workers. Also it requires a contri bution of Rs 3 per day and there is a large proportion of poor in India who will prefer current consumption to future consumption.

Pension payments are primarily aimed at promoting welfare and the existing contributory scheme should not lose sight of

vol xlvi no 8

this objective. The above analysis reveals that pension reforms in India have a long way to go before they serve the interests of both the employer and employee and not of the employer alone.

Notes

1 A deﬁned beneﬁt pension system is one where the pension amount is predetermined either by a lump sum amount or by a predetermined formula. On the contrary, a deﬁned contribution system refers to a pension system where contributions are deﬁned by a lump sum amount or percentage of salary. However beneﬁts in this system depend sometimes on the returns of assets and sometimes are predetermined.

2 Members of self help groups, afﬁnity welfare groups, and welfare schemes of the state governments which have joined the NPS are eligible to be registered with NPS-Lite as a subscriber. For more details (http://pfrda.org.in/writereaddata/ linkimages/NPS-Lite_SOP_Registration1805021791. pdf), last viewed on 8 February 2011.

3 The fund accumulated through contribution is proposed to be invested in three asset categories Equities (E), Corporate Bonds (C), and Government Bonds (G). However if the subscriber is not able to decide how much is to be contributed into each account he/she can opt for auto-choice, in which the percentages are determined by age. It is assumed that a young person can take more risk in the ﬁnancial market, however as she grows old the capability to take risk goes down. Also, it is assumed that if through the auto-choice mode one suffers loss in the beginning of her life and for several years, she can make it up by becoming more risk averse and investing in government or less risky bonds.

4 According to 2001 Census, the working population is 440 million.

5 Aggregators shall be intermediaries identiﬁed and approved by PFRDA, to perform subscriber interface functions under NPS-Lite in respect of their constituent groups. These will be entities already in existence having continuous functional relation ship with a known customer base for deli very of some socio-economic goods/services. For more details, see http://www.pfrda.org.in/writereaddata/ linkimages/Regulationas%20for%20 Aggregator%20Under%20NPS%20Lite-2010_ Final7422072029.pdf, viewed on 8 February 2011.

6 Chile used a powerful media campaign for its pension reforms in 1981, where the ministers went into a fullﬂedged campaign to promote the contributory pension scheme with individual retirement accounts.

7 In order to check this phenomenon, Chile in the initial years of the reforms restricted investment in equities.

References

James, E (1997): “New Systems for Old Age Security-Theory, Practice and Empirical Evidence”, Policy Research Working Paper No-1766, Poverty and Human Resource Division, World Bank, May.

PFRDA (2010, 2011): “Subscriber Registered under NPS”, data available at http://www. pfrda.org.in/ writereaddata/linkimages/NPS%20Subscriber_ Status_30%20July%20108331710362.pdf http:// www.pfrda. org.in/writereaddata/linkimages/ NPS_Subscriber_ 20_November_10612721500.pdf http://pfrda.org.in/writereaddata/linkimages/ NPS%20Subscriber_Status_17%20Jan%20 117647498080.pdf

Sadhak, H (2009): “Does Not India Need a Default Option in the New Pension System?”, Economic & Political Weekly, XLIV, 14 November, 59-68.

Swarup, D (2008): “Role for Public Policy, Regulation and Business Strategy in Bridging India’s Pension Coverage”, keynote address at the 8th Annual IIEF Pension Policy Conference, 30 April, available at http://pfrda.org.in/writereaddata/ linkimages/IIEF%20Conf_30_04_08471758665. pdf, viewed on 8 February 2011.

19

Referees Consulted in 2009 and 2010

1	Balakrishnan, Pulapre	25	James, K S	49	Pati, Biswamoy
2	Banerjee, Ashok	26	Jayaraj, D	50	Prabhakara, M S
3	Banerjee, Prathima	27	Jejeebhoy, Shireen	51	Raghunandan, D
4	Baru, Rama	28	Jodhka, Surinder	52	Ramana, M V
5	Baskar, Vijay	29	John, Mary	53	Ramaswami, Bharat
6	Beena, P L	30	Jose, A V	54	Ravi, C
7	Bhatia, Bela	31	Kannabiran, Kalpana	55	Ray, Ranjan
8	Chakraborty, Pinaki	32	Katju, Manjari	56	Reddy, D N
				57	Reddy, V Nagi
9	Chandrasekhar, C P	33	Khera, Reetika
				58	Sen, Pronab
10	Chattopadhyay, Paresh	34	Krishna Kumar, T
				59	Sen, Tapas
11	Chattopadhyay, Saumen	35	Krishnaji, N
				60	Sharma, Jyotirmaya
12	Chaudhuri, Sudip	36	Lanjow, Peter
				61	Singh, Gurharpal
13	Chikarmane, Poornima	37	Mani, Sunil
				62	Subramanian, Arvind
14	Das, Mausumi	38	Mathew, Sebastian
				63	Suryanarayana, M H
15	Dasgupta, Chirashree	39	Meenakshi, J V
				64	Swaminathan, Padmini
16	Deaton, Angus	40	Mishra, Banikanta
				65	Thapar, Romila
17	Deshpande, Satish	41	Mishra, Tilottama	66	Thomas, George
18	Gazdar, Haris	42	Mohan, Dinesh	67	Unni, Jeemol
19	Ghate, Chetan	43	Mukherjee, Chandan	68	Vanaik, Achin
20	Ghose, Ajit	44	Murgai, Rinku	69	Venkatachalapathy, A R
21	Goldar, B N	45	Nagaraj, R	70	Vijayshankar, P S
22	Heredia, Rudi	46	Narayana, D	71	Vyasulu, Vinod
23	Himanshu	47	Omkarnath, G	72	Wilson, Kalpana
24	Jain, Tarun	48	Panchu, Sriram	73	Yadav, Yogendra
20				february 19, 2011	vol xlvi no 8	Economic & Political Weekly