Principal investigator: Eduarda Miller de Figueiredo
Author: Silvia Prina
Paper title: Banking the poor via savings accounts: Evidence from a field experiment
Location of the Intervention: Nepal
Sample Size: 1.118 households
Sector: Financial
Primary Variable of Interest: Monetary assets, non-monetary assets, and total assets.
Type of Intervention: Savings account offer with minimal transaction fees.
Methodology: Random experiment
Poor families often lack access to formal financial services, such as a savings account, making it necessary to adopt expensive and riskier alternative strategies to save money. A randomized experiment conducted with poor families in Nepalese slums showed that access to a savings account with minimal transaction costs (i.e., zero fees) and physical proximity to a local bank branch can help poor families better manage their resources. Therefore, it is suggested that basic savings accounts positively affect the behavior of these families.
- Policy Problem
The poor are willing and able to save, but lack access to savings accounts or banking services of any kind (Demirguc-Kunt and Klapper, 2012). Therefore, they save informally, that is, they keep money at home, buy livestock and durable goods (Rutherford, 2000; Dupas and Robinson, 2013). The literature also points out that providing access to financial services for the poor appears to increase income and reduce poverty (Aportela, 1999; Bruhn and Love, 2009).
The author's objective in this study was to examine the impact of offering a savings account with minimal transaction costs, i.e., zero fees and high proximity to a bank branch. This is because, according to the literature, reducing transaction costs and improving trust in banking institutions leads to increased use of savings products by poor families (Karlan et al., 2014).
A comparison of savings account usage with accounts offered in other interventions demonstrated that poor families seem to value a savings product associated with low transaction costs. Distance from the bank branch was mentioned by Brune et al. (2014) as one of the reasons for the low utilization of savings accounts. Furthermore, Banerjee and Duflo (2011) suggest that high fees also discourage usage.
- Implementation and Evaluation Context
Formal financial access in Nepal is very restricted, with only 26% of households having a bank account. This access is concentrated in urban and wealthy areas. The main reasons reported in the national survey for the lack of a bank account are transaction costs, distance from banking institutions, and complicated deposit and withdrawal procedures. Only 37% of households who had an account and savings in the previous year stated that they had deposited money into the account (Ferrari et al., 2007).
The experiment took place in 19 slums surrounding Pokhara.[1], which are located on the outskirts of the city or in semi-rural and rural areas, being further away. Only 17% of the households in the sample used in the study have a bank account. The families in the sample earn, on average, $3 per day. 18% of the sample were members of some type of Revolving Savings and Credit Association (ROSCA).[2] and 54% belonged to some microfinance institution or savings cooperative at the beginning of the study.
- Policy/Program Details
A baseline survey was conducted in May 2010 in each location, with all women aged 18-55 participating in the survey, totaling 1.118 households. Both the initial survey and the final survey in June 2011 included information on household expenses to help understand the role played by supply and demand factors in explaining the acceptance and use of the account.
Administrative data from GONESA bank, a non-governmental organization (NGO) operating in the study area, regarding the use of savings accounts were also used. This data includes the date, branch location, amount of each deposit and withdrawal, and the reason for withdrawal from all accounts in the treatment group.
The families in the sample are highly vulnerable to shocks, with 41% indicating they had experienced a negative shock to external income during the previous month and 43% dealing with loans, of which 17% were from family and friends, 17% from a loan shark, and 9% from other sources.
- Method
Of the total households included in the survey, 567 women were randomly assigned to the treatment group, receiving the option of opening a savings account at a local bank branch. The control group was not given this option.
The sample therefore includes families with female heads of household who were, on average, 37 years old. On average, they had less than 3 years of schooling, 90% of respondents were married or living with a partner, and the average family size was 4-5 people. The average weekly family income was around US$24.
Therefore, of the 1.118 households included in the final sample and the 567 that had the opportunity to open a savings account, 84% opened the account and 80% actively used it. Active use was defined as having at least two deposits in the first year after the account was offered.
Intention-to-treat (ITT) effects were estimated, where the main dependent variables are monetary assets, non-monetary assets, and total assets. Some basic characteristics were included to control for and fixed village effects because randomization occurred within the village.
- Main results
The results demonstrated that acceptance and active use of the account are positively related to having a bank account and negatively related to obtaining income from entrepreneurial activity. Furthermore, most transactions made during the study period were deposits, where the average amount deposited in a week was about 8% of the average weekly family income.
The administrative database showed that the main reasons for withdrawing money were to pay for a health emergency (17%), to buy food (17%), to pay off a debt (17%), to pay school fees and materials (12%), and to pay for festival expenses.[3] (8%). The results of the estimates demonstrated a positive effect of gaining access to a savings account in monetary assets.
The average effect of being assigned to the treatment group on the amount spent by the household on health, education, meat and fish, festivals and ceremonies, dowries, and other expenses was estimated. The results demonstrated that financial access has a positive and statistically significant effect on expenses for education, meat and fish, and festivals and ceremonies. Families in the treatment group spent, on average, 20% more on education than families in the control group. Furthermore, regression results showed higher investments in human capital for the treatment group than for the control group.
- Lessons in Public Policy
In general, if granted access to a basic account with minimal transaction costs, poor families use it very frequently. Where access to a savings account appears to help poor families better manage their resources, thus improving their financial situation.
Reference
PRINA, Silvia. Banking the poor via savings accounts: Evidence from a field experiment. Journal of development economics, v. 115, p. 16-31, 2015.
[1] Second largest city in Nepal.
[2] Rotating Savings and Credit Associations (ROSCAs).
[3] Teej festival; Dashain festival, Tihar festival, Maghe Sankranti, New Year according to Nepal calendar and Dumji festival.
