Pensions are, in a sense, a necessary by-product of a rich economy. But what will it take to sell the idea to the rural poor? Especially when their income (never particularly substantial) is seasonal, increasing at harvest time and with demand in the cities for construction-related labor. What are the inducements that can convince them to invest for a future forced upon them by the changing social structure?

Olivia S. Mitchell, a Wharton professor of business economics and public policy and executive director of the Pension Research Council, and Anita Mukherjee, a professor at the Wisconsin School of Business at the University of Wisconsin-Madison, set out to answer these questions in a research paper titled, “Assessing the Demand for Micropensions among India’s Poor.” They chose India as the subject country because it is an ideal setting to study the market for micropensions, or pension plans designed for low-income individuals; the country’s new pension system is designed to reach informal sector workers. A micropension product — Swavalamban — has been applicable to all citizens in the unorganized sector who have joined the National Pension Scheme since 2011. This scheme was funded by grants from the government. It has been replaced with the Atal Pension Yojana, in which all subscribing workers below the age of 40 are eligible for pension of up to Rs. 5,000 ($74) per month after turning 60. In the Atal Pension Yojana, for every contribution made to the pension fund, the central government also co-contributes 50% of the total contribution or Rs. 1,000 per annum, whichever is lower, to each eligible subscriber account, for a period of five years. The minimum age of joining the Atal Pension is 18 and the maximum is 40. The age of exit and the start of the pension is 60.

There was a significant need for a pension system in India. “According to the government of India’s Planning Commission (2014), nearly 30% of the country’s 1.2 billion population lives below the poverty line (BPL),” the researchers write. “At the same time, according to the Population Research Bureau, the share of India’s BPL population age 60 or older is expected to increase from 8% in 2010 to 19% in 2050. Many of these older persons work in the unorganized sector and, as such, lack the identification and proof of employment documents required for accessing basic financial services. Nevertheless, current research estimates that about 80 million of these workers are capable of saving for retirement and the untapped savings are in the order of $2 billion.”

There are other systemic pressures at work. “In India, as in many developing countries, younger adults are moving from rural to urban areas for economic opportunity,” Mitchell and Mukherjee tell Knowledge at Wharton. “Often this means that parents are left behind in the rural areas and, though they may receive financial support from their children, this revolution in traditional family structure can make older people more vulnerable. As a result, older cohorts today may be more interested in a micropension product than they were in the past.”

“Our research shows that individuals broadly preferred a micropension plan that offers withdrawals starting at age 60, as well as partial withdrawals beforehand, to other variants that had different access features.”

Micropension Options

The experiment was conducted in two of the 71 districts in the central Indian state of Uttar Pradesh — Fatehpur and Siddharthnagar. Overall, the statistics are comparable to those of BPL populations. The average survey respondent was 43 years old, owned land, was illiterate and had minimal schooling. The two most common livelihood activities that the respondents engaged in were farming via cultivation of one’s own land (37%), and agricultural labor supplied to non-owned farms (34%). With respect to educational attainment, more than 60% had never attended school, while 21% had five to 10 years of formal schooling. Insurance access among the respondents was low, at 20% of the total sample population. But 66% held a life insurance policy. Saving penetration was relatively high, with 55% having access to a formal saving account. Respondents who had saved had an average balance of Rs. 3,000 in their accounts.

The study placed the existing pension plan as the baseline. An appropriate information and educational scheme was unfolded for the respondents. They were then asked two sets of questions. Group 1 was asked about variants 1B, 1C and 1D, and Group 2 was asked about variants 2B, 2C and 2D. The first variant (A) is the basic micropension product that was then being offered by the Indian government. The other variants included early withdrawal (1B), where the eligibility age was 55 instead of 60; a lower matching rate of 50% instead of 100% (1C); no early withdrawal (1D); delayed withdrawal, where the eligibility age was 65 instead of 60 (2B); a higher matching rate of 150% instead of 100% (2C), and option for full withdrawal at age 60 (2D).

“Our research shows that individuals broadly preferred a micropension plan that offers withdrawals starting at age 60, as well as partial withdrawals beforehand, to other variants that had different access features,” say the authors. “This is similar to the micropension currently on offer in India. One exception to this, not surprisingly, is that our respondents preferred an option that boosted government matches to their plan contributions.”

“Previous studies on the financial lives of the poor have documented that their incomes are irregular and highly seasonal. As a result, requiring them to pay significant sums in just a few payments could significantly reduce demand for the pension product.”

The study results included a few small surprises. Respondents were asked to rank their levels of trust in six institutions on a scale of one to five, with a level of one indicating a complete lack of trust and a level of five representing a very high level of trust. Banks topped the list with a score of 4.49. The government clocked in with 4.22, while non-governmental organizations (NGOs) at 2.55 and village councils (3.34) were regarded as relatively less trustworthy. “We do not know for certain why NGOs were less trusted relative to government entities, but it could be because they had a smaller presence in the areas we studied,” say the authors.

These results are informative about whether microfinance institutions or local governments are likely to be successful intermediaries in the micropension product. Since the government was viewed as a trusted entity, having government support for micropensions may have helped boost adoption and contributions, the researchers note.

The faith put in banks is understandable. “For some time, there has been a growing awareness of the benefits of secure banking, even in remote areas of India,” say the authors. “Moreover, technological improvements using audio cues and fingerprinting have helped expand banking to those who cannot read or write. The Jan Dhan Yojana plan was an important vehicle used to include many rural Indian families in the formal banking system. The Indian government’s demonetization policy has also spurred an interest in enhancing poor peoples’ access to banking, as it created cash constraints throughout the economy.”

Pointing out that India’s recent effort to eliminate larger banknotes was intended to crack down on the “shadow” economy,” the authors add: “It has prompted even poor and rural communities to take up mobile payment services. One example is Paytm, a phone-based system for transferring payments from a bank account to cover people’s everyday liquidity needs.”

Growing the Appeal

The paper has some advice for governments or other entities that are developing micropension products. The researchers write that an effective retirement savings device for the poor must take into account cash-flow needs, income seasonality, competing spending priorities and alternative investment options. They note that respondents to the study were among the poorest in their communities and relied heavily on income from agriculture.

“Previous studies on the financial lives of the poor have documented that their incomes are irregular and highly seasonal. As a result, requiring them to pay significant sums in just a few payments could significantly reduce demand for the pension product,” the researchers write. “For this reason, offering frequent opportunities for such individuals to contribute can be critical to the scheme’s success.”

“To grow [the appeal of micropensions], the focus should be on proper investments (inflation is currently around 10%) and policyholder retention.”

The ability to contribute frequently to an agent who makes door-to-door visits could also help explain why people were interested in micropensions even when making fixed deposits at an Indian bank would offer them high annual returns. “Our initial hypothesis in designing this survey experiment was that some respondents would exhibit a preference between early or late eligibility for withdrawal, and that we would be able to identify the heterogeneity driving these decisions,” the researchers write. “Instead, we found that with the exception of the high match variant, respondents were less willing to adopt or contribute to the alternatives to the baseline micropension product.”

Regarding the faith in the government, the authors elaborate: “In our study setting of rural Uttar Pradesh, one of India’s poorest states, individuals receive many benefits from the government such as ration cards for discounted groceries and free health care. We believe that this repeated and positive interaction with the government has engendered the high level of trust we found.” In addition, the country’s largest life insurance company, the Life Insurance Corporation of India, is also state-owned and enjoys a high level of trust. The authors note that the low levels of education and financial literacy found in the communities they studied highlight the need to provide a financial literacy program in conjunction with the micropension.

“We believe that the move toward digitized finance can facilitate automatic contributions to enhance the appeal of the micropension product in India,” the authors say. “Yet for a micropension plan to work for India’s poor, it must allow policyholders to contribute according to the seasonal incomes they earn while encouraging savings sufficient to provide meaningful support in old age.”

The paper finds that the Indian government’s current micropension product is appealing to the audience it is meant to reach, Mitchell and Mukherjee say. “To grow that appeal, the focus should be on proper investments (inflation is currently around 10%) and policyholder retention,” they add. “The Gates Foundation is also pushing innovations in digital finance [in India] and elsewhere in the developing world, as a means to help the poor do more to save, invest, borrow and mitigate financial risks.”