The composition of savings groups makes all the difference to their effectiveness.
Could you manage your life without being able to save or borrow money from a bank or any other financial institution? Could you do it if your income was US$1.90 per day? Despite the recent dramatic expansion of microfinance and microcredit, to this day about two billion people worldwide are without access to financial services. Many are ultra-poor, living under the equivalent of US$1.90 per day. They face formidable challenges trying to accumulate savings, especially when they are confronted by financial difficulties such as sickness or a bad harvest.
Yet there is a promising, locally-based financial movement for the poorest of the poor. In parts of the world without direct access to banking, savings groups have emerged as a way to ensure that people can save and borrow. In 2014, an estimated 10.5 million households worldwide were members of savings groups, a tenfold increase relative to 2008. There is hope that they can become the most important local financial institution in developing countries.
To understand why this matters, just remember that, even though banks and financial institutions have existed since at least 2000 BC, it wasn’t until the birth of credit cooperatives in the 19th century that most people in Europe and North America gained access to basic financial services. This shows that well-functioning local financial institutions are a key element in the process of economic development. Getting them right is therefore of utmost importance.
A savings group is a group of people who save with and borrow from each other. The group meets weekly for a period, usually a year. Initially, the group establishes its rules of operation, most important of which are how long the group will be in operation and the interest rate that will be charged on loans. After that, each week, those who can afford it put some money in the safe. Those who need a loan can ask to borrow from the common safe based on the conditions established initially. At the end of the period of operation, the money left in the safe is redistributed among the members in proportion to how much each person saved. That is, each member receives back his or her savings plus a fraction of the interest payments collected by the group.
Several factors contribute to the popularity of savings groups. They can make finance accessible to the illiterate. For example, when a person puts money in the safe, her or his passbook is stamped a number of times. If a stamp indicates 10 cents and the saver puts in 40 cents, she or he gets four stamps in their passbook. Keeping track of how much each person saved is therefore a matter of counting the number of stamps. Furthermore, savings groups do not need any form of external financial support. The only support they need is training and monitoring. They are inexpensive to set up and can reach remote and marginalised populations.
Successful savings groups
Despite the widespread popularity of savings groups, little is known about how their composition affects their fortunes. A member’s ability to borrow and to earn a return on savings depends on how much the other members of the group borrow and save. Therefore, a reasonable question is how the composition of a group impacts its ability to meet the group’s financial needs.
To answer this question, my co-author Alfredo Burlando and I ran a large randomised control trial in Uganda and subsequently wrote “Does group inclusion hurt financial inclusion? Evidence from ultra-poor members of Ugandan savings groups”, published in the Journal of Development Economics. The goal of our study is to better understand savings groups and hopefully improve how they function. We found that the savings group composition is, indeed, an important determinant of the group’s functioning. Specifically, we find that loanable funds are scarcer in poorer groups, which disproportionately ends up affecting the poorest members. This means that including many ultra-poor in the same savings group may reduce the ability of the group to provide financial services to these very same ultra-poor.
Our study was carried out from 2013 to 2014 in collaboration with a large USAID-financed anti-poverty programme called SCORE. SCORE officers identified households which are vulnerable because of their low socio-economic status, and then facilitated the creation of savings groups that include these vulnerable participants as well as other members of the local community. Relative to the average member of the local community, households identified as vulnerable have lower income, are more likely to have a disabled member, host an orphaned child or have a member with a chronic disease, and report eating fewer daily meals (0.8 fewer, for an average of 1.1 meals a day).
In 90 villages, SCORE identified 14 vulnerable households for inclusion in a savings group. We randomly assigned these 14 households either to a single savings group, or to two separate savings groups. In both cases, membership in the savings groups was open to self-selected members of the community who are, on average, better off relative to our target population. Because the total membership was capped at 27, we obtained groups with higher or lower average socio-economic status, which we call 50% vulnerable groups and 25% vulnerable groups, respectively. We then carefully analysed the evolution of individual savings and borrowing for all the members of the savings groups created for our study.
At the halfway point of the savings cycle, we found that the more vulnerable groups generated 21 percent fewer savings and loaned 33 percent less than the groups with fewer ultra-poor. This difference was driven by the behaviour of the vulnerable members of the groups, who saved 23 percent less and borrowed 48 percent less when placed in a 50% vulnerable group rather than a 25% vulnerable group. These effects however faded out towards the end of the savings group cycle.
Our results also indicate that wealthier members of the group are more likely to be on the lending side, while poorer groups are more likely to be on the borrowing side. When many ultra-poor are in the same group, fewer loanable funds are generated. As a consequence, these same ultra-poor are unable to borrow from the group.
Our research shows that a balance between people of different poverty levels must be maintained in the composition of savings groups. This is an important factor in whether a group is able to provide financial services – and the associated accumulated wealth – to its members and thus help the ultra-poor out of extreme poverty.
Andrea Canidio is the Stone Fellow of the INSEAD James M. and Cathleen D. Stone Centre for the Study of Wealth Inequality and an Assistant Professor of Economics at the IMT School for Advanced Studies, Lucca, Italy.