NPA in Banking: Meaning, Types, RBI Rules & Real Examples (2026 Guide)

---

NPA in banking refers to a loan or advance where the borrower has stopped making payments for 90 days or more. In fact, the Reserve Bank of India classifies any such loan as a Non-Performing Asset. This single classification triggers a chain of consequences for both the lender and the borrower.

As of September 2025, India’s gross NPA ratio dropped to a historic low of 2.15%. However, this headline number masks important risks. Specifically, unsecured retail lending and MSME portfolios still show elevated stress. Moreover, private banks now face 76% of their slippages from unsecured products alone.

This guide covers everything a lending professional needs to know about NPA in banking. It explains the full form, the RBI’s classification rules, the three NPA types, provisioning norms, and real examples. Additionally, it connects NPA risk to bank statement analysis and early warning systems that help lenders prevent loans from turning non-performing.

What Is NPA in Banking? Full Form and Definition

NPA stands for Non-Performing Asset. In banking, an NPA is any loan or advance where the principal or interest payment remains overdue for more than 90 days. Before 2004, the threshold was 180 days. However, the RBI tightened this to 90 days to align with global prudential norms.

The classification applies to all types of credit. This includes term loans, cash credit, overdraft facilities, agricultural advances, and even bill discounting. If the borrower fails to pay for 90 days, the bank must reclassify the asset from “standard” to “non-performing.”

Why does this matter? Because the moment a loan becomes an NPA, the bank must stop recognizing interest income on it. Additionally, the bank must set aside provisions against potential losses. Both actions directly hit the bank’s profitability.

For NBFCs, the RBI applies a separate framework under the Scale-Based Regulation. However, the core 90-day overdue rule remains consistent. Consequently, any lender operating in India needs a clear understanding of NPA classification.

How the RBI Classifies Non-Performing Assets

The RBI’s Master Direction on Income Recognition, Asset Classification and Provisioning (IRAC norms) governs how banks classify loans. The framework divides all loans into two broad categories: performing assets (standard) and non-performing assets.

Within the NPA category, the RBI specifies three sub-classifications based on how long the asset has remained non-performing and how likely recovery is. Each subclass carries different provisioning requirements. Therefore, accurate classification directly affects a bank’s balance sheet.

The classification process is not optional. Banks must review their entire loan portfolio at least once per quarter. Furthermore, the RBI’s Asset Quality Review (AQR), introduced in 2015, requires banks to recognize NPAs transparently. Any attempt to evergreen loans or hide stress results in regulatory action.

Three Types of NPA in Banking: Sub-Standard, Doubtful, and Loss

Sub-Standard Assets

A sub-standard asset is a loan that has remained NPA for 12 months or less. At this stage, the bank still expects partial or full recovery. However, the risk of loss is elevated. The RBI requires banks to provision 15% of the outstanding amount for sub-standard assets. For unsecured portions, the provisioning rises to 25%.

Doubtful Assets

If a loan stays NPA for more than 12 months, it moves to the doubtful category. At this point, full recovery becomes uncertain. The RBI’s provisioning norms scale with duration. For doubtful assets up to 1 year, the provision is 25% of the secured portion. For 1–3 years, it rises to 40%. Beyond 3 years, the bank must provision 100% of the secured portion. Additionally, the unsecured portion always requires 100% provisioning.

Loss Assets

A lost asset is one that the bank or an auditor identifies as uncollectible. However, some nominal amount may still remain on the books. The RBI mandates 100% provisioning for loss assets. In practice, most banks write off these assets entirely to clean their balance sheets.

NPA Type	Duration	Provisioning	Recovery Outlook
Sub-Standard	NPA for ≤12 months	15% (secured); 25% (unsecured)	Partial to full recovery expected
Doubtful (1 yr)	>12 months as NPA	25% secured + 100% unsecured	Full recovery uncertain
Doubtful (1–3 yr)	1–3 years as NPA	40% secured + 100% unsecured	Recovery increasingly unlikely
Doubtful (>3 yr)	>3 years as NPA	100% secured + 100% unsecured	Minimal recovery expected
Loss Asset	Identified as uncollectable	100% of total outstanding	No recovery expected

Gross NPA vs Net NPA: What Lenders Need to Know

Two metrics define a bank’s asset quality: gross NPA and net NPA. Understanding the difference matters for anyone analyzing a lender’s health.

Gross NPA is the total value of all non-performing loans on the bank’s books. It includes every rupee of principal and interest that remains overdue. As of March 2025, the gross NPA ratio for Indian scheduled commercial banks stood at 2.31%—the lowest in 20 years.

Net NPA, on the other hand, subtracts the provisions the bank has already set aside. It reflects the actual unprotected exposure. The net NPA ratio for Indian banks dropped to 0.52% by March 2025. This gap between gross and net NPA shows that banks have built strong provisioning buffers.

Why does this distinction matter? Because gross NPA shows the scale of the problem. Meanwhile, net NPA shows how well the bank has prepared for potential losses. A high gross NPA with a low net NPA indicates disciplined provisioning. Conversely, a narrowing gap signals that the bank may be under-provisioned.

RBI Provisioning Norms for NPA in Banking

Provisioning is the money a bank sets aside to absorb expected losses from NPAs. The RBI mandates specific provisioning percentages based on asset classification.

For standard assets, the provision is just 0.4% (0.25% for agriculture and SME). However, the moment a loan becomes NPA, provisioning jumps sharply. Sub-standard assets require 15%. Doubtful assets require 25–100%, depending on duration. Lost assets require 100%.

The RBI also plans to introduce an Expected Credit Loss (ECL) framework. This approach, proposed in 2025, would require banks to provision for expected losses at the time of loan origination. In other words, banks would set aside money before a loan turns bad. This represents a significant shift from the current “incurred loss” model.

For NBFCs and fintech lenders, the provisioning burden directly impacts profitability. Therefore, preventing NPAs through better credit underwriting and early detection is far more cost-effective than provisioning after the fact.

Real Examples: How Loans Become Non-Performing Assets

Example 1: The MSME Borrower With Seasonal Cash Flow

A small manufacturer takes a term loan of ₹20 lakh with monthly EMIs. Business is seasonal, and cash flow dips sharply between July and September. The borrower misses two EMIs during the lean period. By the third missed payment (Day 91), the bank classifies the loan as NPA. However, a bank statement analyser reviewing 12 months of data would have flagged the seasonal cash flow pattern upfront. The lender could have structured the loan with a moratorium during lean months instead.

Example 2: The Salaried Borrower Who Lost Employment

A salaried professional takes a personal loan of ₹5 lakh. Six months later, the employer terminates their contract. Without income, the borrower defaults after three months. The loan becomes NPA. In this case, continuous monitoring of bank statement data would have detected the salary credit stoppage. This early warning could have triggered proactive restructuring before the 90-day threshold.

Example 3: The Fraudulent Application

A borrower submits a fake bank statement showing inflated salary credits. The lender approves a ₹10 lakh personal loan. The borrower defaults on the first EMI. Within 90 days, the loan is NPA. Moreover, recovery prospects are minimal because the borrower never had the declared income. An automated bank statement analyser with tampering detection would have caught the forgery before disbursement.

How Early Warning Systems Prevent NPA in Banking

The RBI now requires all public sector banks to maintain automated Early Warning Systems (EWS). These systems use approximately 80 triggers and third-party data to detect stress in borrower accounts before loans turn NPA.

For digital lenders, the most effective EWS layer is continuous monitoring of bank statements. Specifically, a bank statement analyser can track real-time signals like declining balances, bounced mandates, new EMI obligations, and irregular cash flows. These signals appear weeks or months before a formal default.

Here is how this connects to NPA prevention. When the system detects early stress, the lender can intervene proactively. Options include restructuring the loan, offering a moratorium, or initiating soft collection. Each intervention reduces the probability of the loan crossing the 90-day threshold.

According to the RBI, PSBs using comprehensive EWS have reduced their slippage ratio continuously for six years. Therefore, the data clearly support investing in detection infrastructure rather than absorbing NPA losses after the fact.

Key Takeaways

• NPA in banking stands for Non-Performing Asset. It refers to any loan where the borrower has not paid principal or interest for 90+ days.
• The RBI classifies NPAs into three types: sub-standard (NPA for up to 12 months), doubtful (NPA for over 12 months), and loss (identified as uncollectable).
• Gross NPA measures the total value of bad loans. Net NPA subtracts provisions. India’s gross NPA ratio hit a 20-year low of 2.31% in March 2025 and dropped further to 2.15% by September 2025.
• Provisioning norms range from 15% for sub-standard assets to 100% for loss assets. The upcoming ECL framework will require provisioning at origination.
• Real-world NPA cases often trace back to weak credit underwriting. Better bank statement analysis catches seasonal cash flow risks, employment changes, and document fraud before disbursement.
• Early warning systems that monitor bank statement data in real time reduce slippage rates. Prevention costs far less than provisioning.

Conclusion:

NPA in banking is not just a balance sheet problem. It is a credit underwriting problem. Every loan that turns non-performing traces back to a gap in assessment—an income that was overstated, a liability that was missed, or a behavioral pattern that went undetected.

India’s banking sector has made remarkable progress. The gross NPA ratio dropped from 11.46% in 2018 to 2.15% in September 2025. However, emerging risks in unsecured retail lending and MSME portfolios demand continued vigilance.

For lenders, the most effective NPA prevention strategy is better data at the point of origination. Specifically, thorough bank statement analysis catches the risks that credit scores miss. Combined with early warning systems and continuous monitoring, this approach stops loans from becoming NPAs—and that protects both the balance sheet and the borrower.

Frequently Asked Questions