Austin Beck and Brady Grayson with Dr. Daniel Nielson, Political Science

Introduction

Uganda has one of the lowest savings rates among sub-Saharan countries, which tend to have the lowest savings rates in the world. Between 2002 and 2008 the private savings average was at 14.6% compared to the African average of 22.1% (African Development Bank). Low savings rates have a negative impact on the rural poor, who face the worst credit terms and have the least access to credible financial institutions.

In this study we test the effectiveness of a new savings account, the instant bonus account, to incentivize Ugandans to save more money and to save for longer periods of time. The instant bonus account offers an immediate bonus to the participant’s account upon deposit. This bonus is equal to the amount of interest that would be accrued by the initial deposit amount over a six month savings period. The difference between a traditional savings account, which accrues interest over time, and the instant bonus account is this front loaded interest at the beginning of the savings period. If the participant chooses to withdraw their money before the six month savings period has expired, they lose the entire upfront bonus.

The instant bonus account is based on the salience of an immediate reward and the power of loss aversion to encourage household savings

Using a randomized control trial to test this account, we find no difference in expressed interest between the instant bonus account and a normal savings account. We do however find differences in interest in the instant bonus account when we limit the data to male subjects, female enumerators, people in an urban region, and to subjects over the age of 35. Participants were also invited to open either account, but by the conclusion of the study none had. Using our survey data we discuss possible reasons why no one opened an account.

Methodology

A team of researchers comprised of BYU students and local Ugandans collected the data for this study. The Ugandan enumerators were recruited from the Uganda Cooperative Savings and Credit Union and the local scouting organization.

The data for this experiment consists of survey responses from Ugandan residents. During the survey, subjects were offered one of the two savings accounts and given a pamphlet containing information on the account. Survey questions included family background information, individual financial information and other personal information. In total, the research team was able to offer 2,227 accounts and was able to gather survey data on most of the individuals that were offered an account.

The outcome measures used to help us determine the effect of loss aversion on savings include expressed interest in the account and whether people actually set up an account and saved.

Results

As of four months after the completion of the experiment no subject has actually opened an account. This means the only outcome available to us is expressed interest in the account. We look at the impact of being offered the instant bonus account on interest in setting up the account by doing a simple t-test. We find that the interest expressed by individuals offered the instant bonus account was not significantly different than interest expressed by those offered the control account.

Using a logit regression we find that treatment effect varies based on different variables in our data. Specifically, we find that the gender of the subject, the gender of the enumerator, the age of the subject, and being located in an urban region are all statistically significant. All of these variables (being a male subject, having a female enumerator present, being older than 35, and living in an urban region) decreased a subject’s interest in opening an account.

Discussion

An important implication of this study is an answer to the question: Why were people not interested in opening either account, especially with financial institutions that cater to the poor? Data collected in our survey reveal that barriers such as monthly maintenance fees, fees for letters from civic leaders, and a lack of trust of financial institutions contribute to people disinterest in opening accounts.

Because of these barriers when trying to save, we assume that the power of loss aversion and immediate gratification (as incorporated into the instant bonus account) were not salient enough to the people offered accounts. These barriers contribute to the outcomes of zero accounts being opened and the insignificant difference between the instant bonus account and control account. We believe that if the barriers were eliminated, both immediate gratification and the power of loss aversion would have an effect on subjects when they are offered accounts and our results would be different.

Conclusion

Although we found the instant bonus account did not have a significant effect on people’s interest in opening an account, we believe our study provides a valuable contribution to the field of microsavings. Through this study we learned that characteristics such as age, gender, and location influence people’s financial decisions. We also discovered the need to lower the barriers that Ugandans face when seeking to save. In the future, these findings can aid the development of microsavings products to assist the poor in obtaining a higher standard of living