Early Warning System: New Age Financial Risk Aversion

AI/ML-powered Credit Monitoring Systems rely on regulator prescribed EWS methodologies

Across both the public and private sector in India, banks are yet to overcome the hit from bad loans. The gross non-performing assets (GNPA) of scheduled commercial banks rose to 11.5% in March 2018. It dropped marginally to 10.8% in September 2018 – a sign that’s interesting for a few reasons.


The slight improvement of the banking stability indicator (BSI) in September 2018 was due to improved asset quality, but it has also been notice that profitability has eroded further. Irrespective of the marginal decline in GNPA, the rising trend of NPAs and loan related fraud has been a matter of intense regulatory debate.


Further, in addition to rising NPAs, the banking sector also observed a delay in detecting and reporting such frauds to the regulators. Due to such occurrences, the banking sector has been under pressure from regulators to adopt technology-driven proactive measures to tackle the menace of NPAs and loan frauds.


A Revolutionary Approach to NPAs: Early Warning Systems


Keeping frauds in mind, the RBI proposed Early Warning Systems to depict beforehand whether accounts are an asset or a liability. In support of this thought, D2K Technologies offers a fitting EWS product.


CRisMac Early Warning System: A set of automated processes for identifying risk at a nascent stage in the bank's loan portfolio. The rule based solution identifies borrowers at risk of distress or default well before they actually default. A host of Early Warning Signals have been identified depending upon the systems and processes already implemented in banks.


Early warning signals are categorized into – transactional, financial, non-financial, external, and statistical indicators – to identify emerging problems in credit exposure at an early stage. These consist of real-time business data from internal activities and big data from external factors such as forward-and-backward sector linkages, ESG data, and social media.


The automated rule based EWS identifies the traces of potential stress/default long before the occurrence of actual default, enabling the bank to initiate timely remedial measures. Hence, EWS facilitates the bank in strengthening health of the bank’s credit portfolio.