The slow rate of adoption of savings and transaction accounts opened under financial inclusion programmes is a challenge for policymakers across the globe. We use the experience of a large-scale financial inclusion programme in India—Pradhan Mantri Jan Dhan Yojana (PMJDY)—to model the adoption of simple banking products, namely deposit, withdrawal, fund transfer and ATM usage. Based on account-level transaction data, we create a dataset to capture the adoption of each product over time. We use this data to estimate three growth models: Bass, Gompertz and logistic. The Bass model which is based on diffusion of innovations theory is found to be the best fit across products. The role of social influence in the diffusion of the four products studied is found to be much lower when compared to other products in subsistence marketplaces. Social network effect is comparatively stronger in the adoption of fund transfer and ATM. We discuss the mechanisms underlying this phenomenon and contend that information transfer by word of mouth and network effects are the most probable reasons for the observed adoption behaviour.

在金融普惠计划下开立的储蓄和交易账户的采用速度缓慢，这对全球的决策者来说都是一个挑战。我们利用印度大型金融普惠计划的经验（Pradhan Mantri Jan Dhan Yojana（PMJDY））来模拟简单银行产品的采用，即存款，提款，资金转移和ATM使用。基于帐户级别的交易数据，我们创建了一个数据集来捕获随时间推移每种产品的采用情况。我们使用此数据来估算三种增长模型：低音，贡珀兹和逻辑模型。发现基于创新理论扩散的Bass模型最适合所有产品。与生存市场中的其他产品相比，发现在所研究的四种产品的传播中，社会影响力的作用要低得多。在采用资金转移和ATM时，社交网络的作用相对较强。我们讨论了这种现象的潜在机制，并认为通过口耳相传的信息传输和网络效应是观察到的采用行为的最可能原因。