I refer to the article, “How to bring Indonesia’s great unbanked masses into the fold of financial inclusion” (January 5), where agent banking was proposed as a “creative, hi-tech” method to address financial inclusion of the unbanked population in Indonesia.

With all due respect, I would like to share some observations, drawn from China’s experience, to urge caution.

While the rise of fintech or financial technology has offered more convenient and cost-effective banking services, it is predicated on one assumption – that people fully comprehend and trust banking services offered online.

Sadly, this is not the case for the unbanked/underbanked population with minimal experience with banking services, which have been deprioritised by traditional financial institutions. Therefore, human intermediaries become their stepping stone to innovative banking services.

For example, one such intermediary in China hosts a three-tiered network of individual agents targeting customers in less-competitive “below Tier 3” cities, and rural areas. Their agents reach out to and help consumers apply for consumer loans provided by online lenders whose credit decisions are driven by data without any paperwork.

Some big tech firms also opt to use agents. For instance, the fintech arms of China’s biggest internet giants have cooperated with telecom operators to provide instalment loans to customers purchasing cellphones at their stores, with in-store agents facilitating the loan application process.

The agent banking model is no different in nature to the agent insurance model which has been in place for centuries, with the aim of reducing the information gap and promoting trust between institutions and customers. However, some specific concerns exist for the agent-banking model.

First, data privacy: agents would inevitably have access to sensitive client information while interacting with customers. Though there might be penalties imposed in case of data breaches, agents are either less capable of safeguarding client information or even willing to leverage the information on hand to promote other services, for which they can earn commission. In either case, the risk is borne by the financial service provider, since the customers involved would ultimately become their clients.

Second, control and compliance: agents do not operate with the same incentives and constraints as regulated institution financial institutions, and thus there would be challenges in ensuring they comply with the applicable regulatory requirements.

Ultimately, agent banking is just another way to recruit and serve customers, one in which technology is supplemented by human agents for maximised efficiency. Sustainable financial inclusion must be driven by both innovation and efficiency.