*On June 8th and 9th, 2010, MicroSave and FAI sponsored a virtual conference – Reimagining Microfinance around the World: Implementing Lessons from Portfolios of the Poor – to share findings from yearlong financial diaries kept by villagers and slum dwellers in Bangladesh, India, and South Africa.
Session Three Summary (June 9, 2010)
Key Principles of Designing Financial Services
• Daryl Collins posed the question of whether product design for the poor needs to undergo a complete and utter re-think, or whether tinkering around the edges is sufficient.
• The conversation about how to formalize some of the informal mechanisms (such as savings groups) that seem to be working for poor families continued. While Village Savings and Loan Associations (VSLAs) and Savings and Loans Groups (SLGs) seem to be a sort of “savings club plus” with enhanced reliability (in theory), Daryl wondered whether there was evidence that they actually do enhance reliability in practice.
• Daryl followed up on the VSLA and SLG conversation and noted that they were largely absent in any of the three diary countries. Based on her observations and familiarity with other research, she suspected that there would be take up despite the existence of other centrally offered services. Daryl wondered whether people save more in VSLAs than in other more traditional types of financial devices, like a bank account. Citing research from the South African diaries, Daryl reported that respondents saved much more of their monthly income in savings groups than banks, despite holding accounts in both types of accounts. She wondered how these patterns shift when a new type of channel, such as m-banking is introduced into the picture.
• Jonathan Morduch offered his comments on the VSLA discussion and commended Daryl for her question about what happens when people have the choice to save/borrow in VSLAs versus in traditional banks. Jonathan cited observations from Tanzania where the population using VSLAs is distinct from the “banked” population. In Tanzania, the VSLA-users Jonathan saw were also in ROSCAs. This portfolio allowed for greater flexibility and is in accord with other parts of East Africa where it is common for people to be in several ROSCAs simultaneously. Jonathan cited a surprising element in a two-year VSLA in Tanzania where the dividend was divided based on cumulative savings, without regard for whether the savings were deposited in the first month or the 24th. While this arrangement made things simpler for the group, it seemed problematic for economists who weigh the “time value of money” heavily.
• Graham Wright had earlier raised a question about the levels of loss of savings in informal systems that diarists experienced. He referenced a MicroSave report that found that in Uganda, on average people lost 22% of the amount they had saved in the last year through informal devices. Joanna Ledgerwood pointed out that much has changed in the services available in rural Africa since the MicroSave study was published eight years ago. She noted that the members of modern savings groups (VSLAs) in Zanzibar in a DFS/FSDU study from 2006 had an average return of 53% on their savings in the year preceding the DFS/FSDU study, and no members reported losing money.
• The discussion of insurance that began on Day 1 continued. Daryl wondered if one of the reasons why we have yet to see a real surge in microinsurance is because we are putting too much energy into designing the right product, rather than executing the right marketing and systems to support that product. Jonathan agreed that marketing is really critical for microinsurance, but reiterated his thought from Day 1 that we also have a ways to go on improving insurance product design and pricing.
• Julie Zollman from the Fletcher School at Tufts University argued that instead of focusing on insurance, we should be focusing on closing the gap on precautionary savings because of the flexibility it offers in coping with the many kinds of devastating shocks a family can experience. Jonathan agreed that fostering long-term saving is high on the list of "next step" products, and referred to Chapter 6 of Portfolios of the Poorwhich examines the Grameen Pension Scheme (GPS), a contractual saving device that locks up saving for 5 or 10 years. GPS does a good job of enforcing discipline in saving, but isn’t flexible if families need to withdraw their savings early. Jonathan pointed out that the real trick is designing mechanisms with enough discipline and structure to encourage accumulation, coupled with flexibility to make them helpful in emergencies.
• Daryl commented that she thought that microinsurance has shown potential for success, despite a dearth of concrete examples of success. Daryl attributed this disconnect to problems with product design, citing “wildly” different pricing for burial insurance as a prime example. Daryl cited ease of payments and claims, simplicity of the product, trust in the institution, in addition to price that influence take-up. Daryl posted research from the South African diaries (http://www.financialdiaries.com/Funerals_and_Financing.pdf) to illustrate her point. She posed a question to the group and wondered if at least some of the lack of success could be attributed to poor marketing, execution, and support.
Responding to the Financial Needs of the Poor
• Anil Paul from MicroSave pointed out that saving is hard for poor families, and wondered if people successfully continue saving for their life cycle events even after realizing how long it takes to put together a large sum, or if they just save to meet their immediate or unforeseen needs. Jonathan responded that the financial diaries show people actively saving, but for periods under year. He point out that contractual savings can be helpful, since they are targeted toward a clear saving goal from the start, and they help households keep accumulating so the goal is met as quickly as possible.
Do Interest Rates Matter?
• Madhu pointed out that it’s challenging for an MFI to maintain itself as a market leader in price for long. Quality products come as institutions grow and diversify. Increased competition pushes many MFIs to compete on interest rates but over time most of their focus shifts to developing a convenient, service-oriented environment.
• Reflecting on a number of comments that asked “are interest rates important?,” Daryl flipped the question and hypothesized that evidence of interest rates becoming a higher priority suggests that other needs such as flexibility, convenience, safety, etc. are being more or less met. Further, is it safe to say when we observe clients prioritizing interest rates over other qualities, this is evidence that financial markets are “working better” for the poor?
• Kimberly Wilson from Tufts responded and suggested that interest rates become a priority when people actually understand what they are paying—when they have the basic data and knowledge to understand the arrangement.
• Daryl responded and supported efforts that require financial institutions to provide more information than just APR, and cited research from Portfolios of the Poor that found that customers often see interest charges as “fees” rather than percents. Daryl argued that it is not beyond the capacity of poor clients to ultimately understand APR and that understanding would also be beneficial to a well-functioning credit market. Additionally, Daryl’s research has found that when loan agreements include total interest paid, respondents benchmark loan “fairness” by whether the interest paid is double the principal of the loan. As this level is quite high and does not include the maturity of the loan, an APR would be handy if consumers started shopping around for rates on loans of different maturities. Daryl’s concerns centered on the trade offs between immediate consumer protection needs and future goals, stating that immediate needs should not force policymakers into thinking that poor individuals will never be able to calculate or have a need for APR. She concluded that teaching this tool should continue to be included in financial curricula.
Relevance of Market Research
• Helke Waelde from the University of Mainz posed the questions of whether there is a need for theoretical research to understand the key drivers in microfinance. Daryl agreed that theory in development economics has been left woefully wanting in the area of household financial management. Jonathan Morduch was able to point us to some new work, and recommended the following articles:
• "Financial Intermediation" in the latest Handbook of Agricultural Economics by Jonathan Conning and Christopher Udry
• Work by Ashok Rai, Jonathan Conning, Sendhil Mullainathan, and Karna Basu
• The second edition of The Economics of Microfinance, just released buy MIT Press
• Following up on Helke Waelde's comments, Kimberly Wilson from The Fletcher School of Law and Diplomacy at Tufts University submitted a question on the use of lotteries as a saving mechanism for the poor. Kimberly cited her research in Haiti and observations in Cambodia and India to frame her question. In light of the approach banks and MFIs have taken to harness the gambling impulse in middle and lower-income populations through prize-linked savings instruments, Kimberly wondered if this type of approach could be adopted for the poor.
• Daryl responded to Kimberly's comments and wondered, given what seems to be a typical interest in playing the lottery among low or even middle to low income communities, why lottery-linked products haven't been more "earth-shatteringly" successful. Citing her research in South Africa, Daryl noted that there wasn't substantial funding being funneled to lotteries or any sort of gambling efforts. The lottery-linked bank account introduced in the South African market was targeted at middle income South Africans, not the poor. And, while it was generating a lot of hype, deposits it was attracting were not significant.
• Kimberly followed up on Daryl's comments and reported that gambling is occurring in Haiti, Cambodia, Ghana, and India. She attributed the lack of awareness of these lotteries to the social stigma of gambling. Other reasons for limited take-up could be that the transformational lump sum may be hard to generate by banks with only 1-2 points of interest in play, or the fact that transactional costs are too high.
• Peter Wrede wondered why Brazilian insurance products that include a lottery, despite their success, have not been replicated elsewhere. One major disadvantage of life insurance is that you have to die to benefit, so having an additional trigger, such as winning the lottery to benefit is very attractive. The Brazilian life insurance product is designed with two triggers: the sum insured will be paid either on death or if the individual policy number is drawn in the periodic lottery (ex. If the last X digits of the policy number coincide with the last X digits of the national lottery lucky number of the week). This allows consumers to enjoy benefits and removes the waiting period to build up savings (if savings are part of the insurance).
• Jonathan Morduch asked Peter if the lottery element was mainly a marketing tool. Citing examples of other lotteries where the chance to win increases with savings level (like at BRI), Jonathan wondered if there was an equivalent feature in the Brazil product.
• Peter followed up and explained that the lottery had “always” been there and thus doubted whether it was used as a marketing tool. However, the lottery was not a “microinsurance” device. Peter did not believe that Brazil’s system included an element where the chance of winning increases with savings level.