Nava Ashraf of Harvard University Press designed a randomized control trial for a commitment savings product at a Philippine bank. The study involved a baseline survey of 1777 existing or former bank clients, with the product offered to a randomly selected subset of 710 individuals. After twelve months, average savings balances for the treatment group increased by 81 percentage points relative to the control group.
Use Cases
- Analyzing the impact of commitment devices on savings balances based on the 12-month treatment effect.
- Studying self-selection into commitment products based on time discounting preferences measured in the baseline survey.
- Modeling savings behavior using data from a randomized control trial with treatment and control groups.
- Investigating gender differences in financial product adoption based on the finding that women with specific discount rates were more likely to open the account.
Strengths
- Data was collected via a randomized control trial methodology, which suggests a strong causal inference design.
- The study includes a substantial baseline sample of 1777 individuals and tracked outcomes over a 12-month period.
- Findings include specific quantitative results, such as a 28.4% acceptance rate and an 81 percentage point increase in savings balances.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count for the final dataset is unknown, which may limit suitability assessment.
- Data may reflect geographic and institutional bias inherent to a single bank in the Philippines.
Provenance
- Source
- Nava Ashraf, Harvard University Press
- Collection Method
- Randomized control trial and survey implemented with a Philippine bank.
- Time Range
- The experiment included a baseline survey and a 12-month outcome period.
- Geography
- Philippines