The majority of the poor lack access to bank accounts and rely on informal savings mechanisms (Banerjee and Duflo 2007)1. There is little evidence on how poor households’ behavior changes when offered access to a traditional savings account. Would poor households open a basic savings account if given access to one? Would this access help them accumulate small sums into increased savings over time? In this ﬁeld experiment households in 19 poor neighborhoods in Nepal were randomly offered simple bank accounts with no fees at local bank branches. Results show that there is untapped demand for savings accounts. Access to the savings accounts increased monetary assets and total assets without crowding out other kinds of assets or formal savings in institutions. Households with this new ﬁnancial access also increased investments in education.
The potential benefits of a formal savings account are manifold and include improved ability to cope with shocks, asset accumulation and capacity to plan for the future. There has been promising though limited evidence to date on the benefits of access to savings accounts for the poor. Additionally, such studies have thus far focused on specific subsets of the population such as entrepreneurs or, for commitment savings studies, existing clients of a bank or microfinance institution. The current literature lacks studies that consider how generally poor households’ behavior changes when offered access to financial markets through a savings account.
This study takes place in Pokhara, Nepal in 19 poor neighbhorhoods commonly referred to as slums. The neighborhoods, which are actually permanent settlements, vary in population from 20 to 150 households. At the time of the study baseline, households in these areas had an average weekly salary of 1,600 Nepalese rupees, roughly $20 USD. Study participants were primarily involved in the agricultural and construction industries, engaged in activities such as collecting sand and stones, selling produce, and raising livestock. Researchers collaborated with a local NGO, Good Neighbour Service Association Nepal (GONESA), as it was offering savings products in new locations, to assess the impact on the population.
A baseline survey was administered to 1,236 households. Shortly thereafter, public lotteries were held to randomly offer new savings accounts at a local GONESA branch to half of the surveyed households. The other half served as a comparison group and was restricted from opening accounts.
The new bank branch offices in the 19 neighborhoods are open twice a week for three hours on an established schedule. Customers cannot make deposits or withdrawals outside these hours at the branches, but they can visit the bank's main office at the city center all days of the week. The bank accounts have no opening or transaction fees and pay 10% annualized interest rate to customers.
One year after the accounts were opened, an endline survey was administered to both the beneficiaries and comparison group participants to collect data on consumption levels, financial behavior, and asset accumulation, among other indicators. Administrative data were also collected from GONESA on savings account usage at the individual level which includes the date, location, and amount of every deposit and withdrawal.
This field experiment provides detailed evidence on the causal effects of access to a fully liquid bank account on savings and investment behavior. Results show first, that there is untapped demand for fully liquid savings accounts: 84 percent of the households that were offered the account opened one. Second, the poor can save: 80 percent of the households that were offered the account used it frequently, making deposits of about 8 percent of their weekly income 0.8 times per week, on average.
Third, a year after the start of the intervention, households with access to the GONESA savings accounts had 25% more monetary assets than households who did not have access. In addition, their total assets, which include monetary and non-monetary assets (consumer durables and livestock), were 12% higher than the ones of comparison households. Hence, the increase in monetary assets did not seem to come at the cost of crowding out savings in non-monetary assets.
Fourth, being offered access to a savings account strongly increased household investment in education. Furthermore, households with access to the savings account who had suffered health shocks in the previous month did not seem to suffer large changes in weekly income, suggesting that having savings may serve as a buffer against adverse circumstances.
Overall, findings suggest that if given access to a basic savings account with no fees, poor households save more than those who only have informal strategies at their disposal. They accumulate greater assets and invest more in education. These results highlight that savings accounts can be beneficial even when the households do not use the money saved for microenterprise development because they permit households to make productivity-enhancing investments in human capital.