Sector Guides
How to Analyze PSU Bank Stocks in India: CASA & GNPA
Team QuartrlyPSU bank earnings hinge on the spread between what a bank earns on loans and pays on deposits (NIM), how cheaply it funds itself (CASA ratio), and how much of its loan book has gone bad (GNPA/NNPA). State Bank of India, Bank of Baroda, and Canara Bank — which together hold a large share of India's ~60% PSU-owned banking assets — report these numbers every quarter, and the post-IBC (Insolvency and Bankruptcy Code) cleanup of legacy corporate bad loans has made the trend in these metrics, not just the absolute level, the thing that moves these stocks. Because PSU banks also carry priority sector lending mandates that private peers don't, their "healthy" ranges run structurally lower on NIM and CASA than a private bank's.
Key Takeaways
- NIM (Net Interest Margin) above 3% indicates healthy lending spreads for PSU banks; compression below 2.5% signals margin pressure
- CASA Ratio above 40% provides low-cost funding advantage and protects margins during rate cycles
- GNPA below 4% and NNPA below 1% are current benchmarks for well-managed PSU banks post-IBC cleanup
- PCR above 75% indicates adequate provisioning buffer against future loan losses
- Rising SMA-2 accounts serve as an early warning indicator for future NPA slippages
Quick Reference
| Metric | Definition | Healthy Range | Warning Sign |
|---|---|---|---|
| NIM | Net interest spread on earning assets | > 3.0% | < 2.5% or compressing |
| CASA Ratio | Low-cost deposits to total deposits | > 40% | < 35% or declining |
| GNPA | Bad loans to total advances | < 4% | > 5% or rising |
| NNPA | Bad loans net of provisions | < 1% | > 2% |
| Slippage Ratio | New bad loans during period | < 1% | > 2% |
| Credit Cost | Provisions as % of advances | < 0.75% | > 1.5% |
| PCR | Provisions covering bad loans | > 75% | < 60% |
| RAM Share | Granular lending proportion | > 55% | Heavy corporate exposure |
| CD Ratio | Loans to deposits | 75%-80% | > 90% |
| SMA-2 | Accounts overdue 61-90 days | Stable/declining | Sharp increase |
| RoRWA | Profit on risk-adjusted assets | > 1.0% | < 0.8% |
Understanding PSU Bank Metrics
PSU banks differ from private sector banks in several important ways. Government ownership brings implicit sovereign backing but also mandates like priority sector lending (40% of advances) and financial inclusion targets. These obligations can impact profitability metrics when compared to private peers.
The post-IBC era has fundamentally improved PSU bank asset quality. Legacy bad loans from infrastructure and corporate lending have been substantially resolved. Investors now focus on the sustainability of this improvement through metrics like slippage ratio, credit cost, and the composition of the loan book toward granular RAM (Retail, Agriculture, MSME) advances.
Net Interest Margin (NIM)
What it is: Net Interest Margin is calculated by dividing net interest income (interest earned minus interest paid) by average interest-earning assets. It is expressed as a percentage and represents the core spread a bank earns from its lending operations.
Why it matters: NIM is the primary indicator of a bank's lending profitability. For PSU banks, which face pricing constraints from priority sector lending and competition for quality borrowers, maintaining NIM above 3% indicates effective asset-liability management.
What good looks like: PSU banks typically operate with NIMs between 2.8% and 3.5%, lower than private banks due to their loan mix. SBI reported domestic NIM of 3.09% in Q2 FY26. NIMs above 3% indicate healthy pricing power without excessive risk-taking.
Red flag: NIM compression exceeding 20 basis points QoQ may indicate rising deposit costs or competitive pressure on lending rates. Conversely, a sudden NIM spike without corresponding growth may signal riskier lending to higher-yielding segments.
Example from earnings call:
"Our domestic NIMs for the quarter improved by 7 basis points quarter-on-quarter to 3.09%, driven by repricing of deposits and operating leverage." — SBI Q2 FY26 Earnings Call
CASA Ratio
What it is: CASA Ratio measures the proportion of Current Account and Savings Account deposits to total deposits. Current accounts pay zero interest while savings accounts pay regulated rates (currently around 2.7%-3%), making CASA the cheapest source of funds for banks.
Why it matters: A high CASA ratio lowers the bank's cost of funds, providing a structural advantage in maintaining NIMs. PSU banks like SBI leverage their extensive branch networks and customer trust to maintain strong CASA franchises.
What good looks like: CASA ratio above 40% is considered strong for PSU banks. SBI maintains CASA above 40% due to its brand strength and distribution reach. Private banks like HDFC Bank operate at 35%-40% post-merger.
Red flag: CASA ratio declining while loan growth accelerates indicates reliance on expensive bulk deposits or certificates of deposit. This funding mix pressures margins and may not be sustainable during tight liquidity conditions.
Example from earnings call:
"We have maintained CASA ratio at more than 40%. Our strength in low-cost liabilities is derived from our brand and customers' trust in SBI." — SBI Q2 FY26 Earnings Call
Gross Non-Performing Assets (GNPA)
What it is: GNPA percentage is calculated by dividing gross non-performing assets by gross advances. Per RBI classification, an asset becomes non-performing when interest or principal remains overdue for more than 90 days. GNPA represents the total stock of bad loans before accounting for provisions.
Why it matters: GNPA indicates the overall health of the loan portfolio and the effectiveness of credit underwriting. Post-IBC, PSU banks have significantly cleaned up legacy NPAs, making current GNPA levels more representative of ongoing credit quality.
What good looks like: For PSU banks, GNPA below 4% is currently considered healthy. Top performers like SBI operate at 2%-3%. A decade ago, PSU bank GNPAs exceeded 10%, so current levels represent substantial improvement.
Red flag: GNPA rising for two or more consecutive quarters, or GNPA above 5%, warrants investigation into specific segments (corporate, agriculture, retail) driving deterioration.
Net Non-Performing Assets (NNPA)
What it is: NNPA is calculated as (Gross NPAs minus Provisions) divided by (Gross Advances minus Provisions). It represents the bank's actual exposure to bad loans after accounting for provisions already set aside.
Why it matters: NNPA shows the residual risk on the balance sheet. A bank can have moderate GNPA but low NNPA if adequately provisioned. NNPA is the more relevant metric for assessing potential future P&L impact from bad loans.
What good looks like: NNPA below 1% is the current benchmark for well-managed PSU banks. Top performers maintain NNPA below 0.7%. SBI and Bank of Baroda have achieved these levels post-cleanup.
Red flag: NNPA above 2% indicates either inadequate provisioning or severe asset quality issues. The gap between GNPA and NNPA should be examined alongside PCR.
Slippage Ratio
What it is: Slippage Ratio measures fresh additions to NPAs during a period divided by standard advances at the beginning of that period. Unlike GNPA (which is a stock measure), slippage is a flow measure showing new asset quality deterioration.
Why it matters: Slippage ratio is a more timely indicator than GNPA because it captures recent credit quality trends. Consistently low slippages indicate disciplined underwriting and early warning systems are working effectively.
What good looks like: Slippage ratio below 1% is considered excellent for PSU banks in the current environment. SBI reported 0.84% slippage ratio, indicating tight credit monitoring.
Red flag: Slippage ratio above 2%, or rising agricultural slippages blamed on "seasonality" for multiple consecutive quarters, suggests structural underwriting issues rather than temporary factors.
Example from earnings call:
"Slippage Ratio for Q1 FY25 improved by 10 bps YoY and stands at 0.84%." — SBI Q1 FY25 Earnings Call
Credit Cost
What it is: Credit Cost is calculated as provisions for loan losses divided by average advances, expressed as a percentage. It represents the actual P&L impact of asset quality issues during the period.
Why it matters: Credit cost directly impacts profitability. A bank with high operating profit but high credit cost will show weak net profit. In the post-IBC era, declining credit costs have been a primary driver of PSU bank earnings recovery.
What good looks like: Credit cost below 0.75% is the current guidance from well-managed PSU banks. SBI reported 0.38% credit cost, meaning almost all operating profit flows to the bottom line. Bank of Baroda targets sub-0.75% credit cost.
Red flag: Credit cost consistently lower than slippage ratio over multiple quarters may indicate under-provisioning. The bank may be deferring recognition of losses.
Example from earnings call:
"Credit cost is 0.55%, which is within our guidance. We aim to maintain credit cost below 0.75%." — Bank of Baroda Q2 FY26 Earnings Call
Provision Coverage Ratio (PCR)
What it is: PCR is calculated as provisions held against NPAs divided by gross NPAs, expressed as a percentage. It measures how much of the bad loan book is already covered by funds set aside.
Why it matters: High PCR provides a buffer against future losses. If a fully provisioned loan defaults completely, the P&L impact is minimal. PCR above 70% is required by RBI for resolution of stressed assets under certain frameworks.
What good looks like: PCR above 75% is considered adequate, with 90%+ indicating a fortress balance sheet. SBI reported PCR of 75.66%. Bank of Baroda and Canara Bank maintain PCR above 90%.
Red flag: PCR below 60% indicates potential for future earnings pressure as provisions need to be built up. This is particularly concerning if GNPA is also elevated.
Example from earnings call:
"We have a well-provided NPA book with PCR at 75.66%. This is the first time we crossed 75% cover." — SBI Q2 FY26 Earnings Call
RAM Advances (Retail, Agriculture, MSME)
What it is: RAM represents the share of Retail, Agriculture, and MSME advances in the total loan book. This segmentation reflects a strategic shift away from large corporate lending toward granular, diversified exposure.
Why it matters: PSU banks historically suffered losses from concentrated corporate exposures (infrastructure, steel, telecom). RAM lending distributes risk across millions of smaller borrowers, reducing the impact of any single default.
What good looks like: RAM share above 55% of total advances indicates a well-diversified book. Bank of Baroda maintains approximately 61% RAM share. Canara Bank reported RAM credit of ₹6.71 lakh crore with retail credit growth of 29%.
Red flag: Rising retail slippages despite high RAM share indicates the "safe" segment is souring. This removes the diversification benefit and suggests broad-based credit quality issues.
Example from earnings call:
"The bank's focus on retail, agriculture, and MSME segments showed strong results, with RAM credit reaching ₹6,71,141 crore. Retail credit surged 29.11%." — Canara Bank Q2 FY26 Earnings Call
Credit-Deposit Ratio (CD Ratio)
What it is: CD Ratio is calculated as total advances divided by total deposits. It measures how much of the deposit base is deployed as loans versus held as investments or reserves.
Why it matters: CD Ratio indicates both growth appetite and liquidity management. Banks need to balance loan growth (which drives NII) against maintaining adequate liquidity buffers.
What good looks like: CD Ratio between 75% and 80% is considered optimal for PSU banks. This allows reasonable loan growth while maintaining liquidity. Some private banks operate at higher CD ratios (80%-85%).
Red flag: CD Ratio above 90% indicates potential liquidity stress. The bank may struggle to fund further loan growth without expensive bulk deposits or may face challenges during deposit outflows.
Special Mention Accounts (SMA)
What it is: SMA is a classification system for accounts showing early signs of stress. SMA-0 accounts are overdue 1-30 days, SMA-1 are overdue 31-60 days, and SMA-2 are overdue 61-90 days. After 90 days, accounts are classified as NPA.
Why it matters: SMA data provides a leading indicator of future slippages. Rising SMA-2 accounts in the current quarter will likely become NPAs in the next quarter. RBI mandates SMA reporting to enable early intervention.
What good looks like: Stable or declining SMA-2 as a percentage of advances indicates effective early warning and collection systems. No specific benchmark exists, but quarter-over-quarter trends matter.
Red flag: Sharp increase in SMA-2 accounts, particularly in specific segments (unsecured retail, MSME), signals incoming NPA pressure within 90 days.
Return on Risk-Weighted Assets (RoRWA)
What it is: RoRWA is calculated as net profit divided by risk-weighted assets. Unlike ROA (which uses total assets), RoRWA adjusts for the risk profile of the asset base, providing a more accurate measure of capital efficiency.
Why it matters: RoRWA shows how effectively a bank generates returns relative to the risk it carries. Banks can improve RoRWA by either increasing profitability or optimizing their asset mix toward lower risk-weight categories.
What good looks like: RoRWA above 1% is considered healthy for PSU banks. Top performers achieve 1.2%-1.5%. Private banks typically show higher RoRWA due to better operating efficiency.
Red flag: RoRWA below 0.8%, or declining trend over quarters, indicates either weak profitability or inefficient capital allocation toward higher risk-weight assets.
Special Considerations: The Write-Off Strategy
PSU banks frequently use technical write-offs as a balance sheet management tool. When a loan is fully provided for (100% PCR), the bank may write it off, removing it from the visible loan book. The borrower still owes the money, and the bank continues recovery efforts.
When recovery occurs on a written-off account, the entire amount flows as income to the P&L since the loan was already fully provided for. This creates an earnings smoothing mechanism where banks can report strong quarters by recognizing recoveries from past write-offs.
Investors should examine whether reported "recoveries" or "negative credit costs" come from operational improvements or from the accumulated pool of past write-offs. The latter, while legitimate, does not indicate sustainable earnings improvement.