Peer effects and women's demand for financial services in Peru

This study will develop two experiments using nudges to increase the use of financial services by Peruvian women microentrepreneurs who are currently clients of two financial institutions. The delivery channel for the messages to these clients takes advantage of peer effects which seem important for these entrepreneurs; a fictional character with characteristics that resonate with the entrepreneurs will deliver personalized communication to promote saving and use of credit.

For the trial, Whatsapp text messages and audios, based on a fictional character, will be sent to randomly-selected clients over a two-month period, while a control group will receive innocuous, placebo messages.

Primary and secondary outcome measures will be measured one month after the intervention was completed.

This study will develop two experiments using nudges to increase the use of financial services by Peruvian women microentrepreneurs who are currently clients of two financial institutions. In Institution "A", of the 13,000 clients who have a programmed savings product, but only between 4% and 7% save periodically. In Institution "B", only 42% of clients who received a loan took out a second loan in the subsequent 12 months, despite being pre-approved for the loan.

Semi-structured in-depth interviews were carried out with the clients of the partner financial institutions with the objective of identifying the “pain points” and the main behavioral biases that women face when related to savings and use of credit. Based on the finding of the qualitative study on the importance of the peer effects for business development, we decided to develop a delivery channel for messages to these clients that takes advantage of this peer effect. A fictional character was created with the intention of generating confidence in businesswomen by having a life story and identity similar to them; this character will deliver personalized communication, phrased in a way that emulates a relevant peer, with a similar story and a similar language to the entrepreneurs, in order to promote saving and use of credit.

Eight Whatsapp text messages and audios the fictional character will be sent to randomly-selected clients over a two-month period, while a control group will receive innocuous, placebo messages. The eight messages will be of four types:i) a message of transparency, whose objective is to inform women that the messages they will receive are part of a research project that seeks to enhance the growth and development of their businesses, but also gives women the opt out, i.e., not to receive future messages if they prefer not to participate in the experiment; ii) two introductory messages that seek to validate the character as an interlocutor-peer of the women who will participate in the experiment; iii) eight messages with content, which are specifically those that seek to influence the behavior of women to confront the behavioral biases identified in the qualitative study: self-efficacy and present bias; and iv) a farewell message, the objective of which is to inform the women that no more messages will be sent.

The following principles shaped the design of the messages:
• Self-contained: messages should not redirect clients to any other place on the web. While women have smartphones, but they may not have postpaid plans, so everything must be kept in the realm of WhatsApp.
• Accessible: they should be in formats that women consume on a daily basis: text and audio.
• Empathic: they should take into account social inequalities and the specific needs that women have for financial products.
• Convenient: they should be constructed to be read or listened to while working; women do not have much time available to read or listen to messages.

In all messages, simple, direct and clear language must be used.

Primary and secondary outcome measures will be measured one month after the intervention was completed.

2024-02-26

2024-04-22

For Institution A, primary outcomes include the percentage of women who have a savings account and the amount saved.

For Institution B, primary outcome measure is the percentage of clients (also disaggregated by gender) who take out pre approved second loans.

For Institution B,

These primary outcome measures will come from administrative data from the financial institutions.

For Institution A, secondary outcome measures include: i) intention to save, ii) whether or not women have a saving target; and whether a saving plan was shared with another individual.

The secondary outcome measures for Institution A will be collected by a survey conducted at endline.
No secondary outcome measures will be examined for Institution B because there are not resources to conduct an endline survey that would allow additional outcomes to be analyzed.

The experimental design for the two financial institutions will be quite similar.

First, the sample will be stratified by region to ensure representativity across Peru.

For both financial institutions, the unit of randomization is the financial advisor. Advisors will be randomly assigned to either the treatment or control groups.

For Institution A, financial advisors organize their clients into groups ranging in size from 10 to 25 members; advisors communicate with these groups via Whatsapp. The universe of advisors eligible for the experiment will have: i) at least 20 groups of clients, and ii) a minimum of two years working for Institution A.

Given the relatively small size of groups, our strategy will be to include all members of a group in the sample; as noted above, since advisors are randomly assigned to either the treatment or control group, there is no possibility of one group of a given advisor being in the treatment group and another group from this same advisor being in the control group. (This avoids the potential of contamination of the control groups via information accidently leaked by the advisor.)

From this universe, a sample of approximately 588 financial advisors will be randomly selected to participate in the experiment. Of these, half will be randomly assigned to the treatment group and half to the control group.

For Financial Institution B, the unit of randomization is also the financial advisor. Group size is larger than in the case of Institution A. Groups may have up to 250 members, and the average group size is about 150. From the universe of financial advisors, approximately 108 will be selected to participate in the experiment, with half being randomly assigned to the treatment group and half to the control group. Given the large group size, not all group members from a given advisor will be included in the treatment or control group; this participation will be capped at 30 individuals in order to ensure the participation of a sufficiently large group of advisors.

Randomization done in office by computer.

Financial advisor (see description of experimental design above)

Yes

For Financial Institution A, there will be 588 clusters.

For Financial Institution B, there will be 108 clusters.

For Financial Institution A, there will be 10,000 observations. For Financial Institution B, there will be 16,000 observations.

For each financial institution, there will be only one treatment arm.

For Institution A, minimum detectable effect size is an absolute increase in savings of 2 percentage points, from 4% to 6% This is a 50% increase. For Institution B, minimum detectable effect size is an absolute increase in use of credit of 5 percentage points, from 50 to 55%. This is a 10% increase.