«D I R E C T I O N S I N D E V E LO P M E N T Human Development Public Disclosure Authorized The Elderly and Old Age Support in Rural China Challenges ...»
The key determinant of the real value of the individual account portion of the pension will be the rate of return earned on contributions during the accumulation phase. Figure 6.2 presents a simulation of the benefit level from an individual account based on a total annual contribution of 360 renminbi (RMB) in present-value terms (that is, just under three times the current minimum combined contribution from the individual plus a government match) and based on the following assumptions: (a) the scheme is launched in 2010, (b) rural mortality remains unchanged, (c) the retirement age is increased to 65, (d) benefits are price indexed, (e) the real discount rate is 2.5 percent, and (f) the wedge between real rate of return and real wage growth is −1.0 percent over the first 10 years and 1.0 percent afterward. Figure 6.2 shows the relatively modest individual account accumulation that will likely result under this set of plausible assumptions with a retirement age that is notably higher than that currently adopted in the national rural pension pilot.
Of course, the national pilot’s total replacement rate will not be based 116 The Elderly and Old Age Support in Rural China Figure 6.2 Present Value of Individual Account Monthly Benefit by Years of Contributions Given a 360 RMB Total Annual Contribution yuan / month (PV)
Source: World Bank forthcoming.
Note: Assumes (a) current rural mortality; (b) price indexation; (c) retirement age of 65; (d) real discount rate of
2.5 percent; (e) the wedge between real rate of return and real wage growth is −1.0 percent over the first 10 years and 1.0 percent afterward; and (f ) the scheme begins in 2010. PV = present value.
solely on benefits paid from individual accounts. But the example underscores the importance of ongoing consideration of the total replacement rates the authorities will seek from their rural pilot and urban residents’ schemes and the contribution toward the level of income replacement that can realistically be expected from individual accounts under the current minimum required contributions and government match.
An alternative—and, in the view of this book, preferred—approach would be for the central government to provide a guaranteed rate of return on the individual accounts. The guaranteed rate of return could be set at the national growth rate of the gross domestic product (GDP). This approach would offer a workable compromise between the competing objectives of security and adequacy of returns.
Although it may be too complicated at this stage, an option for the future would be to allow different investment rules for different cohorts of workers participating in the individual account scheme. For older workers, having conservative investment rules makes sense because it limits exposure to market risk and volatility and takes account of their lower degree of financial literacy. However, for younger workers, the compounded effects of rules for low rates of return on conservative investments are likely to be large, and their extended contribution histories will Issues for the Evolution of the Rural Pension System 117 allow for averaging market volatility across the life cycle. The authorities might consider using age-based portfolio default rules, which allow more aggressive investment for younger contributors that gradually becomes more conservative over the life cycle to focus on preserving the value of accumulations at retirement.
Another question that merits consideration is whether a reserve fund is desirable. The answer depends in part on the intended long-term role of the basic pension benefit and any future social pension for the elderly.
Such funds are more typically used to address fluctuations in returns or the exhaustion of benefits because of longevity. In principle, the concept of a reserve fund is attractive, and practice in places such as Zhuhai indicates interest.5 Management of funds in individual accounts and the appropriate level of the system for management also merit consideration. Current pilot schemes and past practice have generally involved management at the county level. However, combining funds from localities into a single pool that can be managed at higher levels to generate economies of scale in fund management has clear benefits. China’s experience with rural pensions cautions against localized management and investment of accumulations. A sensible option would be to have the National Social Security Fund (NSSF) manage funds in rural individual accounts, although considerable preparation would be required for the infrastructure to record and remit balances to the NSSF and to make transfers back to paying authorities as the system matures. More specifically, the new scheme could put the public subsidy for individual accounts into the NSSF and put contributions in the Agricultural Bank of China, the Postal Savings Bank, or other suitable financial institutions, with a uniform interest rate announced nationally and recorded in individual account passbooks. Although the option of contracting fund management to authorized asset managers might be considered in the longer run, it would require a much stronger regulatory and accountability framework than exists presently.6 With respect to the regulation of funded rural pension schemes, capacity is presently inadequate; lack of capacity has been an ongoing weakness of rural schemes in China (see World Bank forthcoming; Wu 2009). In addition, existing capacity to regulate even urban schemes is stretched.
Thus, rather than creating new institutions for regulating funded rural pensions, strengthening capacity in existing urban institutions to expand coverage to rural areas would seem more desirable. Taking such an approach would also help achieve critical mass in regulatory capacity.
118 The Elderly and Old Age Support in Rural China For the credibility and utility of the individual account in the medium term, permitting account holders to borrow a portion of their account accumulations at a specified interest rate and with legally binding repayment conditions is probably desirable. Eligible conditions might include the purchase of a home, certain types of health care costs, or costs for children’s education. In some systems, including 401(k) accounts in the United States, such “borrowings” are repaid at a designated interest rate by the individual. In other schemes, loans simply reduce the total accumulated balance. Given the volatility of nonwage incomes and the economic conditions of residents who work outside the formal sector, the opportunity to borrow from one’s individual accounts would be an attractive feature of the scheme; the structure of such a feature should be considered, ensuring, of course, that an adequate minimum accumulation will be available at retirement.
Rules for payouts must be developed. The options include annuitization, phased withdrawal, and lump-sum payouts. Previous rural schemes have used phased withdrawal, which would be adequate for the foreseeable future, but authorities should look toward eventually providing annuities instead. The lump-sum option, however, will likely be important for the transitional generation (which is discussed further below).
Some previous rural schemes in China provided for the lump-sum repayment of contributions to those without full contribution histories.
In some cases, repayment was made at the point of retirement (for example, in Beijing and Zhuhai); in others, repayment was made during the accumulation phase. Other schemes have given older cohorts the option of continuing to contribute for up to five years after reaching the normal retirement age.
Notional or Fully Funded Accounts for the Individual Account?
A bigger-picture question on the individual account portion of the rural pension scheme is whether the funding of public subsidies for individual accounts should be actual or notional.7 If investment of accumulations is restricted to bank deposits at fixed interest rates—which has been the case in urban schemes and most rural pilots to date—the more fundamental question is whether funded individual accounts or notional defined contribution (NDC) accounts are the preferred approach. The relative attraction of actual versus notional funding depends on a few key factors. The first is the investment rules for the individual account (including consideration of possible guarantees for rates of return), which were discussed earlier. If investments are constrained to fixed-interest Issues for the Evolution of the Rural Pension System 119 deposits in public sector banks, the effective difference between funded schemes and NDC schemes is significantly reduced. Second, the policy choices for the future of the urban system may be relevant, given the government’s desire for greater harmonization of urban and rural pension systems over time. If the urban system moves to an NDC design at some point, the attraction of notional funding for subsidies on individual accounts for the rural scheme increases. Counterbalancing these factors, however, are two significant issues that are specific to rural individual accounts. First, farmers lack awareness of the NDC concept and have a natural skepticism toward past scheme performance based on the promises of rural pension schemes that are not funded—“real money” would be a more intuitive concept for farmers. Second, and even more important from a structural viewpoint, the demographic transition in rural areas in the coming decades would argue against an NDC approach. On balance, therefore, pursuing a fully funded approach to the individual account, as is being done under the national pilot scheme, makes sense.
Portability between Schemes Transition issues will become important as the national rural pension scheme matures. The first type of transition is from old to new rural pension schemes within the same locality. Practice in China has varied, with areas such as Beijing allowing portability of funded accumulations from old to new schemes, whereas others, such as the province of Hunan, stipulate that participants must close their accounts in old schemes before starting afresh in new pilots.8 The national scheme appears to allow transfers of balances from prior schemes. The second (and ultimately more important) issue is portability between rural and urban schemes (or migrant-worker schemes located in urban areas where they exist). Portability will be critical to reaching the government’s stated goal of an integrated social security system by 2020. The ease of portability has varied according to scheme design and the compatibility of rural pilots with existing urban schemes. Suzhou, for example, has had a simple two-to-one rule for farmers wishing to transfer their social pooling rights from the rural to the urban scheme; transfer is easily done because the rural contribution base has been exactly half that of the urban system.
Beijing has also had provisions for portability, although it occurs only at the point of retirement. If a former farmer has accumulated enough years in the urban system at retirement, he or she will enjoy an urban pension (with an allowance for lower rural contributions), whereas his or her urban contributions will be credited to the rural pension scheme if the 120 The Elderly and Old Age Support in Rural China accumulation period in the urban system is fewer than 15 years. In principle, in all schemes, the funded portion can easily become portable.
Because the MDC design adopted in the national rural pilot has no social pooling, apportionment is not an issue.
Having a funded portion in the rural pension system may be useful for rural workers who move either to other rural areas without hukou or to urban areas. In principle, transferring balances from one scheme to another in the funded portion through totalization agreements between schemes should be straightforward. This approach raises the question, however, of whether to also consider pooling urban and rural accumulations in the provinces and combining fund management. Such an approach has merit, although it may require greater coordination than is possible in the short run.
The issue of portability raises a host of design and implementation questions that will need to be closely considered during the pilot phase of the rural pension system. When a worker moves, will the funds in his or her account move or will only his or her contribution records move?
How will the system account for accumulations in the rural system when workers move to the urban system? Finally, are pensions paid from various locations or from a single payment authority? These sorts of practical questions raise issues for record keeping, communication and information exchange between systems, and exchange of account information. In principle, standardized record-keeping and reporting formats for the new rural pension system (which have been developed and are being disseminated by the Ministry of Human Resources and Social Security) should be capable of addressing such practical questions. The harmonization of fund transfer and pension disbursement procedures would also be required. Again, some degree of centralization would be desirable to lessen administrative demands at the lower levels, although centralization might occur at the provincial level within national guidelines. These issues will require elaborated guidance from the central authorities over time.
International Experience with MDCs9 The MDC approach is relatively new but is being explored in a number of developing countries. Many OECD countries provide tax incentives to encourage workers and employers to contribute to voluntary private pension schemes. Most OECD countries provide incentives of at least 10 percent of contributions—the average is around 20 percent—although this provision is made through tax deductions for employers on such Issues for the Evolution of the Rural Pension System 121 contributions (Yoo and de Serres 2004). In the United States, for example, the value of forgone tax revenue for such deductions is equivalent to 1 percent of GDP. Evidence shows that such programs increase participation, although the composition of savings is affected rather than its overall level. What seems clearer is that tax exemptions increase saving among low-income people and others with low saving rates (Benjamin 2003;