Social Networks and Financial Inclusion in India

We choose a sample of villages in Jharkhand (India) and randomly assign them to one of four conditions. All households in a particular village, assigned to a condition, undergo the same treatment. The four conditions are:

1. Control: No intervention takes place.

2. Treatment 1: A standard financial literacy training is conducted by a partner non-profit organization (Swadhaar FinAccess) on financial products and services available to households.

3. Treatment 2: A standard financial literacy training is conducted, as in Treatment 1. In addition to this, a communication-central peer is incentivized to spread information among households in the village about selected financial products after the training.

4. Treatment 3: A standard financial literacy training is conducted, as in Treatment 1. In addition to this, a financially influential peer is incentivized to spread information among households in the village about selected financial products after the training.

2025-03-30

2025-04-30

We examine the effect of various financial inclusion interventions on the key dependent variables of financial attitudes, awareness, and behaviors of households living around the poverty line in rural India. Our unit of observation is a single household.

The dependent variables will be captured pre- and post-intervention using survey questions about the household's knowledge of available financial products (awareness), beliefs about products offered (attitudes), and adoption decisions (behaviors).

As part of the pre- and post-intervention surveys, we map communication networks in a village and capture households' beliefs about the financial influence of peers. We examine changes in beliefs and network structure as a secondary outcome of interest. In addition to this, we also survey financial behaviors of households when preparing for unexpected future income shocks.

To examine the effects of various types of financial literacy interventions, we conduct a pre-interventions survey to capture households' baseline financial attitudes, awareness, and behaviors. Villages are randomly assigned to treatments comprising different types of financial inclusion interventions, such that all households in a given village undergo the same treatment. After this, we conduct a post-intervention survey of households and capture financial attitudes, awareness, and behaviors. We examine if these variables change to measure the impact of various types of financial inclusion interventions.

As per RBI guidelines, standard financial literacy interventions conducted by partner organizations in India must share information about several financial products and services relevant for households around the poverty line. We identify three financial products from those mentioned in standard interventions and reinforce information about these products through a carefully chosen peer in each village. We choose three financial products that match the following criteria:
(i) a significant proportion of households in the chosen villages are eligible for said product
(ii) low baseline adoption of the product among a majority of the population surveyed
(iii) product is beneficial to households, given the pecuniary and non-pecuniary costs of adoption

We use the pre-intervention survey to map the existing communication networks and financial influence networks in our sample of villages. This is then used to create a ranked list of communication-central peers and financially influential peers within each village. We use a fixed assignment rule to identify a peer for each village who is either communication-central or financially influential (but not both).

Villages are randomly assigned to one of four conditions (one control or three treatment groups). Households in villages assigned to the treatment groups undergo the standing financial literacy training. In villages assigned to Treatment 2, a communication-central peer is incentivized to reinforce information about relevant products, while in villages assigned to Treatment 3, a financially influential peer is incentivized to reinforce information about relevant products. No intervention of any kind takes place in the control group.

Using the post-intervention survey, we measure changes in key dependent variables for a village as a causal effect of the intervention assigned to the village. We also study whether village networks change after the interventions. Changes in network structure are measured using changes in average connectivity, distribution, and centrality of each node pre- and post-intervention. For all our analyses, we will identify an appropriate regression specification, while controlling for other variables.

Sample: We have a sample of 100 villages in Jharkhand (India) where the partner non-profit organization is planning to begin operations. These can be mapped to 36 unique gram panchayats (GPs) in the districts of Khunti, Ranchi, Simdega, and Gumla. Due to security concerns, the Gumla district was eventually dropped, leaving us with 23 unique GPs. We further exclude 3 GPs due to geographical proximity (< 4 kms of each other) to mitigate spatial spillover concerns. For the remaining 20 GPs, we identify the headquarter village, giving us a sample of 20 villages. These villages are then randomly assigned to treatment or control conditions.

Villages are randomly assigned to treatment groups or a control group, via randomization by a computer.

Randomization takes place at the village level. Villages are randomly assigned to either a control group or to one of three treatment groups.

Yes

5 villages per cluster (that is, 5 villages are assigned to a condition), 4 clusters (conditions) in total.

Approximately 2000 households (~100 households surveyed per village, across 20 villages).

- 5 villages in control
- 5 villages in Treatment 1 (standard financial literacy training by non-profit organization)
- 5 villages in Treatment 2 (standard financial literacy training by non-profit organization and reinforcement of information by a communication-central peer)
- 5 villages in Treatment 3 (standard financial literacy training by non-profit organization and reinforcement of information by a financially influential peer)