Financial Inclusion Through Voice Alerts

How to cite

Pande, Rohini; Field, Erica; Rigol, Natalia; Schaner, Simone; Troyer-Moore, Charity. 2018. "Financial Inclusion Through Voice Alerts." AEA RCT Registry. AEARCTR-0002917. https://doi.org/10.1257/rct.2917-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Rohini Pande* , Erica Field , Natalia Rigol , Simone Schaner , Charity Troyer-Moore

* Primary investigator

Registry ID
AEARCTR-0002917
DOI
https://doi.org/10.1257/rct.2917-2.0
First registered
Last updated
Status
On going
Trial dates
2018-01-01 – 2019-10-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/2917

Abstract

Over the last few years, financial inclusion has been a key priority for the Government of India, evident in the country's push to universalize access to basic "no frills" bank accounts through the Pradhan Mantri Jan Dhan Yojana (PMJDY). While there has been progress towards providing basic banking products to the masses, the path to universal financial inclusion is still fraught with major challenges. Even among holders of accounts with positive balances, many accounts remain dormant. Low financial literacy among the rural poor, especially women, implies the newly banked often lack a transparent, reliable, and low-cost method to access information their accounts, especially direct benefit transfers (DBT) made by the government into citizens' accounts. As a result, awareness of transfers made into accounts is low. This lack of information about accounts impedes account holders' ability to effectively use their bank accounts and adversely affects trust in formal banking channels. Low awareness also increases the likelihood that corruption can occur at the last-mile, since even biometrically-linked identification tools do not preclude all forms of funds leakage and fraud.

With the aim of addressing these challenges, we test the value of an Interactive Voice Response (IVR) system to provide individuals with important information needed to successfully engage with the formal financial system. We propose testing the IVR system through a randomized control trial to assess the degree to which delivery of account information, particularly benefits from government schemes, via IVR helps integrate rural individuals into the formal banking system, increases knowledge and use of bank accounts, and whether IVR-based information provision speeds this process.

Intervention

IVR phone calls will deliver information about account activity and balances to account holders.

Experimental design

This study includes females who bank at local kiosks using bank accounts provided through one of India's largest public sector banks. These local kiosks are operated by agents who provide basic banking services. We will send regular IVR phone calls to a randomly selected group of female account holders (the treatment group). These phone calls will provide the account holders information on their account activity and account balances for the accounts that they have at the kiosks. For another group of female account holders, we will monitor account activity but will not send IVR calls. We will use data on account activity before, during and after the intervention to compare outcomes for the treatment and control group. Randomization is at the individual level, and the sample is stratified by variables correlated with our main outcomes of interest: district, phone ownership, frequency of visits to a kiosk bank, and level of trust in saving money in the kiosk bank. We expect to examine heterogeneity based on these indicators, and notably trust in the banking system, as well. These indicators were collected in the baseline for the study, conducted in March and April 2018.

Design details

Randomization method
Randomization was completed using Stata with fully replicable code.
Randomization unit
Randomization is at the individual level, and the sample is stratified by variables correlated with our main outcomes of interest: district, phone ownership, frequency of visits to a kiosk bank, and level of trust in saving money in the kiosk bank.
Clusters
This RCT is randomized at the individual level.
Observations
900 account holders across two districts
Arms
900 account holders across two districts; the sample is equally divided across treatment and control.
Primary outcomes
Account activity Average Bank Account Balance Number of visits to the bank
Secondary outcomes
Trust in the banking system Demand for mobile phones Outcomes related to fraud and leakages

Cited by

No citation data available for this registration yet.

Keywords