The Reserve Bank of India has cautioned the country’s non-bank finance companies against getting over-reliant on algo-based credit models or they may regret later.
“It is crucial to recognize that rule-based credit engines are only as effective as the data and criteria upon which they are built. Overreliance on historical data or algorithms may lead to oversights or inaccuracies in credit assessment, particularly in dynamic or evolving market conditions,” RBI Deputy Governor Swaminathan J said in a speech on May 15. The speech was uploaded on the central bank’s website on Wednesday.
Algorithm-driven lending models use various indicators, ranging from cash flow data to home addresses, to swiftly generate approvals for personal loans.
India’s shadow lenders need to maintain a vigilant approach regarding their strengths and weaknesses, coupled with ongoing scrutiny and validation of credit scoring systems, Swaminathan said, adding it is essential for supervised entities to regularly adjust rule engines and models based on real-time insights and evolving situations.
Periodic validation, either internally or externally, is crucial to ensure the continued relevance of these models. I urge the heads of Risk and Internal Audit to give special focus to this mandate, the deputy governor said.
Swaminathan J urged risk function leaders to focus on the business models being adopted to ensure their ongoing viability. He also said it is important to periodically review the portfolio mix to mitigate the potential accumulation of risks, such as concentration risk.
“Over reliance on such products may bring grief at some point in time later,” he said.
Furthermore, it’s evident that the risk thresholds set for certain product categories or segments, such as unsecured lending, in some entities, are unreasonably high for long-term sustainability. I trust that risk managers will conduct a thorough evaluation of any potential risks accumulating in their portfolios.
RBI Governor Shaktikanta Das had also earlier this year warned against ‘algorithm-based’ lending by banks and said the country’s lenders must take guard against depending on algorithms and artificial intelligence to review customers for loans. Model based, algorithm lending can “lead to a potential crisis,” Das had warned.
Reuters had reported in March citing sources that the RBI is intensifying its efforts to curb “excessive enthusiasm” in retail lending, amid growing concerns about escalating risks to the financial system.
In recent months, the Reserve Bank of India has implemented stricter regulations on high-risk lending practices to enhance financial stability. In November, it increased the capital requirements for unsecured lending by banks. Additionally, it directed lenders to divest investments in alternate investment funds or face substantial provisioning costs against those assets.
