The Need for an Automated NPA System

The Need for an Automated NPA System


Recently, RBI issued a detailed guideline to all Scheduled Commercial Banks and Small Finance Banks to have an appropriate IT system in place for NPA identification, income recognition, provisioning, and generation of related returns in an automated way. The banks are advised to put such a system in place by June 30, 2021. Key highlights of the guideline are:

  1. Asset classification and provisioning should be done based on pre-defined rules in the system as per the regulatory stipulations.
  2. Income recognition/derecognition of NPAs accounts should also be system driven.
  3. All types of credit facilities including investments, irrespective of size, sector, or type of limits, should be covered in the system.
  4. The system should have the ability to upgrade/downgrade the accounts without any manual intervention.
  5. Asset classification should be updated on a daily basis. The system should be able to generate a classification report at any given point in time with the actual date of classification.
  6. In cases where the NPA system is maintained outside the Core Banking System (CBS), the borrower accounts should be updated in CBS in an automated way.
  7. In exceptional cases where manual intervention/system-override is required, it should follow two-level authorization (maker/checker) and proper audit trail/logs should be maintained.
  8. Any changes to the data, parameters should be updated from the front-end, not backend.
  9. The system should have proper User Access Management and access to the solution should be provided on a “need to have/least privilege” basis for all users.

Over the years, either proactively or post-facto RBI observations, few banks have started using a dedicated NPA Management system. However, majority of banks still rely on core systems for asset classification with high levels of manual intervention. Furthermore, provisioning of accounts and return generations are performed manually in spreadsheets. The said approach is inefficient for managing NPAs for banks with a diversified product portfolio and a large customer base. RBI’s latest guidelines not only addresses the current inefficiencies of the entire process of asset classification, income recognition, and provisions of various asset classes as per the IRAC norms, but also paves the path for an error-free, streamlined reporting in the stressed economic scenario.

Based on our experience, below are some of the current problem statements and possible solutions:

  1. Problem Statement: Inefficiencies of multiple source systems

Over years of operations and launching newer products, banks have ended-up with disparate sources for products (Business Loans, Housing Loans, and Cards, etc.). The efficacy of these systems to perform NPA Calculations and keep up to date with the changing regulations is always a question mark.

Possible Solution 1: Core Systems to upgrade/maintain IRAC related configurations

Possible Solution 2: Perform IRAC related calculations in an independent system


  1. Problem Statement: Unavailability of unified customer level view

IRAC norms require asset classification at borrower level (not facility-wise) thereby identify the worst classification across facilities/accounts for a borrower. A classic problem arising out of facilities for the same borrower in multiple systems. Even if all systems are capable of identifying NPAs at the account level, a consolidated view across facilities is required for worst classification, which results in the manual process for NPA identification, by collating information across sources

Possible Solution 1: Load customer/account level information from peripheral systems into a CBS capable of performing customer unification and maintaining IRAC configurations

Possible Solution 2: Load customer/account level information from all sources into a third-party solution capable of performing customer unification and maintaining IRAC configurations

  • Problem Statement: Lack of accuracy in reporting

In the current challenging economic scenario, RBI’s reporting requirements have not only increased but also are under constant scrutinizing lenses. The entire process of asset classification, applying provisions, and reporting is a manual error-prone process leading to multiple iterations over an elongated timeline. RBI is moving towards a daily identification/reporting of NPAs, which could be a near impossibility with current team sizes at banks

Possible Solution 1: Core Banking Systems to provide reporting and intermittent manual intervention capabilities (with complete audit trail/workflows)

Possible Solution 2: A robust third-party system, with automated classification workflows and capabilities of manual intervention as and when required, with workflows and audit trails

NPA regulations are dynamic in nature as evident with the recent tinkering with moratorium guidelines to ease the stressed situation. Banks relying on excels or only CBS for their NPA management have continuously come out their comfort zones to apply newer classification and provisioning norms within aggressive timelines. Moreover, with a view of reverting back to normalcy, it is also pertinent to manage two books from an audit perspective. Such frequent regulatory changes highlight the need for an automated system. Automation also improves the transparency of processes and policies with the auditors and regulators, with any divergence or disagreements easily traced and reported.

The rationale behind mandating banks to adopt NPA automation is to make sure that the entire process of classification of assets, income recognition, and applying provisions is performed in a seamless, secure, and transparent manner with minimal intentional manual intervention. A robust and automated NPA system not only helps in proper and timely identification of NPAs but also helps in generating reliable and quality information with regards to the asset quality for effective decision making for both banks as well as the regulator.

Tags: No tags

Add a Comment

Your email address will not be published. Required fields are marked *