Using Psychology to Encourage Good Behaviour

Millions of people affected by the economic slowdown and job losses are looking at the prevailing low interest rates to borrow from banks and NBFCs. A majority of them happen to be first-time borrowers. While the potential growth in retail loan is good news for financial institutions, it is also full of risks and opens them up to potential fraud, defaults and NPAs, as most of these ‘new to credit’ borrowers are low on understanding of financial concepts and processes. Hence, not only do lenders need to nurture their relationship with first-time borrowers, they must also take responsibility to ensure the borrowers repay on time, making it a meaningful association for both.

To encourage timely repayments, lenders need to know the psychology of their borrowers. Behavioural science offers several insights into this space. Experts suggest people do not always behave rationally or in their best self-interest. Hence, it is worthy for lenders to understand what drives people to make timely loan repayments.

Lending is a data-intensive business. Financial institutions collect a lot of data from their borrowers to assess their creditworthiness. With the power of big data and analytics, artificial intelligence and machine learning, banks and NBFCs can analyse it to understand subtle patterns. This knowledge enables them to create customised ‘nudges’ to influence timely EMI repayments.

Loan repayment behaviour is a function of many factors, including employment, family responsibilities, social status, etc. A one-size-fits-all approach does not help in effective collections and recovery. AI introduces human element into the process, thereby improving customer response. It allows lenders to drive personalised communication with different customer segments at appropriate times. Lenders also receive additional personal insights from the customers to steer their conversations more effectively.

Lenders cannot assume that merely supplying facts and rational arguments would inspire customers to be conscious of repayment. Instead, they should design collection strategies to reflect how people respond, based on past behavioural patterns. With AI and ML, financial institutions can draw ‘choice architecture’ to trigger desired behaviour. Such architectures are about making the right offer to the right borrower at the right time. Lenders can frame suitable choices for their debtors by effectively segmenting them and adapting their engagement and collection strategy to match their needs. For instance, there could be fewer but more effective choices for customers to help them make the right decision. However, behavioural economists have shown that too many decisions decrease people’s motivation to decide. As a result, they are shortsighted and often put off decisions. Too many options, combined with the complexity of making a choice, create a formidable barrier.

Another area of tech-led behavioural intervention is helping people keep their word. People tend to keep promises they have made. Thus, collection systems can leverage this behaviour to help borrowers pledge to pay. In case the system senses a likelihood of default, it can remind the borrower of this pledge. As a result, they are more likely to repay.  

Despite stringent credit assessment and compliance norms, it is challenging for financial institutions to control the entire journey of borrowers. With the help of AI, they can analyse massive chunks of data to foresee if some loans are on the verge of becoming stressed assets. Such projections can help them expedite their recovery efforts and cut losses. Such a system, with a customer behavior-driven approach towards debt collections, will continuously improve learning from previous data patterns, and become a strategic pillar of the debt recovery process.

The author is the CEO and co-founder of Datacultr

DISCLAIMER: Views expressed are the author’s own, and Outlook Money does not necessarily subscribe to them. Outlook Money shall not be responsible for any damage caused to any person/organisation directly or indirectly.