Banking Sector Playbook
Banks are financial intermediaries. They borrow money (deposits, bonds) at a low rate and lend it (loans) at a higher rate. The spread is their profit. | Inc...
How banks make money
Banks are financial intermediaries. They borrow money (deposits, bonds) at a low rate and lend it (loans) at a higher rate. The spread is their profit.
Revenue structure
| Income type | % of Total Income | Description |
|---|---|---|
| Net Interest Income (NII) | 65–80% | Spread between interest earned and interest paid |
| Fee & Commission Income | 10–20% | Loan processing fees, credit card fees, bancassurance |
| Treasury Income | 5–15% | Trading gains on govt bonds, forex. Volatile. |
🗣️
"Bank ka kaam simple hai — sasta paisa lao, mehenga udhaar do. Fark milk ke jaisa margins."
The 6 metrics that define a bank
1. GNPA & NNPA (Asset Quality)
GNPA% = (Gross Non-Performing Assets) / (Gross Advances) × 100
NNPA% = (Gross NPA - Provisions) / (Net Advances) × 100
| Metric | Excellent | Acceptable | Danger |
|---|---|---|---|
| GNPA % | < 1.5% | 1.5–3.0% | > 5% |
| NNPA % | < 0.5% | 0.5–1.5% | > 2% |
Why it matters: Bad loans are the #1 killer of banks. If borrowers don't repay, the bank loses capital.
2. Net Interest Margin (NIM)
NIM = (Net Interest Income) / (Average Interest Earning Assets)
| Bank Type | Typical NIM |
|---|---|
| Large private (HDFC, ICICI) | 3.5–4.5% |
| PSU banks (SBI, PNB) | 2.5–3.5% |
| Small finance banks | 5–8% |
Higher NIM = better spread. But very high NIM may mean risky lending (micro, unsecured).
3. Cost-to-Income Ratio
Cost to Income = (Operating Expenses) / (Total Income)
| Range | Interpretation |
|---|---|
| < 40% | Very efficient (technology-led, low branch costs) |
| 40–50% | Industry norm for well-run banks |
| > 55% | Bloated cost structure (typical for mid-tier PSUs) |
4. CASA Ratio
CASA% = (Current Account + Savings Account Deposits) / (Total Deposits)
CASA deposits are the cheapest source of funds (0–4% interest vs 6–7% for term deposits). Higher CASA = lower cost of funds = better NIM.
| Bank | CASA % | Advantage |
|---|---|---|
| HDFC Bank | ~45% | Strong retail franchise, low CoF |
| SBI | ~45% | Massive branch network, salary accounts |
| IndusInd | ~38% | Improving but still below leaders |
5. Provision Coverage Ratio (PCR)
PCR = (Cumulative Provisions) / (Gross NPA)
PCR > 70% = conservative (enough buffer for surprises). PCR < 60% = risky (if more loans go bad, there's not enough cover).
6. Credit Cost
Credit Cost = (Provisioning Expense) / (Average Advances)
Should be < 2% for good quality banks. Rising credit cost = rising bad loans.
Red flags for banking stocks
Red flags to watch
- GNPA spike without macro explanation — management hiding problems?
- NIM declining for 3+ quarters — competitive pressure or risky portfolio runoff
- CASA declining — depositors leaving (trust issue?)
- Cost-to-income rising — operational inefficiency
- Consistently changing auditors or qualified audit reports
- Large exposure to one sector (e.g., infra loans in 2015-17 devastated PSU banks)
- Tech platform outages — for digital-first banks, platform reliability is a risk
Valuing Banks: The PB-ROE framework
Standard PE works poorly for banks because:
- Earnings are volatile (one bad quarter of provisioning wipes out profit)
- Banks are leveraged by definition (D/E of 10x is normal)
Use PB (Price/Book) combined with ROE:
| Bank Quality | Expected PB | ROE Range |
|---|---|---|
| Best-in-class (HDFC Bank) | 2.5–3.5x | 15–17% |
| Good quality (ICICI, Kotak) | 2.0–3.0x | 13–16% |
| PSU banks (SBI, BoB) | 0.8–1.5x | 8–12% |
| Stressed / turnaround | 0.5–1.0x | < 8% |
The PB-ROE scatter
Plot PB vs ROE for all banks. Banks above the regression line are "expensive" relative to profitability. Banks below are potentially undervalued.
Case study: HDFC Bank vs Yes Bank
| Metric (FY19) | HDFC Bank | Yes Bank |
|---|---|---|
| GNPA % | 1.4% | 3.2% (later found to be much higher) |
| NIM | 4.3% | 3.3% |
| CASA | 42% | 30% |
| PCR | 72% | 43% |
| ROE | 16% | 10% |
| PB | 4.0x | 1.5x |
Yes Bank appeared "cheaper" (lower PB). But look at metrics — low NIM, low CASA, very low PCR. Within 12 months, hidden NPAs emerged and the stock collapsed 90%.
Lesson: For banks, "cheap on PB" without quality metrics is a value trap.
🏢 Apply: See it on InvestingSamjho
Banking analysis exercise
- Compare 3 banks: 1 private, 1 PSU, 1 small finance bank
- Pull NII, provisions, and operating expenses from P&L
- Calculate NIM, Cost-to-Income, Credit Cost for each
- Plot PB vs ROE — which bank offers the best value for quality?
Summary
Banks are fundamentally about asset quality, cost of funds, and efficiency. Always check GNPA, NIM, CASA, and PCR before looking at valuation. A "cheap" bank with bad asset quality is a trap.
1. Banking — How Banks Make Money
What you will learn
- The 3 revenue streams of an Indian bank
- How NII (Net Interest Income) dominates bank earnings
- Why treasury income is volatile and misleading
- How to read a bank P&L differently from a manufacturing company
Banks are not normal businesses
A bank's raw material is money. It borrows money at a low rate (deposits) and lends it at a higher rate (loans). The spread is the profit.
This sounds simple, but the entire complexity of banking comes from managing the risks in this spread — credit risk (borrower doesn't repay), liquidity risk (deposit withdrawal before loan matures), and interest rate risk (rate changes squeeze the spread).
Revenue stream 1: Net Interest Income (NII) — 65-80% of income
NII = Interest Earned on Loans & Investments - Interest Paid on Deposits & Borrowings
NII is the bread and butter of banking. Everything else is supplementary.
Where does interest income come from?
| Source | Typical Yield | Size (% of interest-earning assets) |
|---|---|---|
| Retail loans (home, auto, personal) | 8-12% | 40-55% |
| Corporate loans | 8-10% | 20-35% |
| SME/MSME loans | 10-14% | 10-20% |
| Government securities (SLR) | 7-7.5% | 18-23% (mandated by RBI) |
Where does interest expense come from?
| Source | Typical Cost | Size (% of total deposits) |
|---|---|---|
| Savings accounts | 2.7-3.5% | 25-35% |
| Current accounts | 0% | 8-15% |
| Term deposits (FDs) | 6-7.5% | 45-60% |
| Borrowings (bonds, RBI) | 6.5-8% | 5-15% |
The CASA advantage
Current Account (0%) + Savings Account (3.5%) are the cheapest sources of funds. A bank with 45% CASA ratio pays much less for deposits than one with 30% CASA — directly boosting NII.
Revenue stream 2: Fee & commission income — 10-20%
Non-interest income that doesn't depend on the spread:
| Fee Type | Example | Trait |
|---|---|---|
| Loan processing fees | 1-2% of loan amount | One-time, tied to disbursement |
| Credit card fees | Annual fees, interchange | Recurring, high-margin |
| Bancassurance commissions | Selling insurance via branches | Growing fast, capital-light |
| Transaction banking | Cash management, trade finance | Sticky, relationship-driven |
| Wealth management | Advisory fees, MF distribution | Premium segment |
Why fee income matters: It requires no capital, carries no credit risk, and diversifies revenue. Banks growing fee income faster than NII are building better business models.
Fee income leaders (FY24 approximate)
| Bank | Fee Income % of Total | Strength |
|---|---|---|
| HDFC Bank | ~20% | Cards, retail, bancassurance |
| ICICI Bank | ~22% | Retail + corporate transaction banking |
| Kotak | ~25% | Wealth management, broking |
| SBI | ~12% | Large base but capital-heavy |
Revenue stream 3: Treasury income — 5-15% (volatile)
Banks hold government bonds (SLR requirement) and trade them. When interest rates fall, bond prices rise — banks book trading profits.
| Rate Environment | Treasury Impact |
|---|---|
| Rates falling | Bond prices rise → trading gains → treasury income high |
| Rates rising | Bond prices fall → mark-to-market losses → treasury income negative |
| Rates stable | Minimal trading gains, only coupon income |
Don't trust treasury gains
Treasury income is volatile and non-recurring. A bank reporting "record profits" driven by treasury gains is not genuinely more profitable — rates will reverse.
How to strip treasury gains
Look at Pre-Provision Operating Profit (PPOP) excluding treasury:
Core PPOP = NII + Fee Income - Operating Expenses
This is the true operating performance of the bank. Ignore treasury.
The bank P&L structure
| Line Item | Formula |
|---|---|
| Interest Income | Interest on loans + investments |
| Interest Expense | Interest on deposits + borrowings |
| Net Interest Income (NII) | Interest Income − Interest Expense |
| Other Income | Fee income + treasury income |
| Total Net Income | NII + Other Income |
| Operating Expenses | Employee cost + technology + branch costs |
| Pre-Provision Operating Profit (PPOP) | Total Net Income − Operating Expenses |
| Provisions | Money set aside for bad loans |
| Profit Before Tax | PPOP − Provisions |
| Tax | |
| PAT | PBT − Tax |
🗣️
"Normal company ka profit = revenue - cost. Bank ka profit = spread - cost - provisions. Provisions hi sabse unpredictable cheez hai."
✅ Checklist
- ○I understand the 3 revenue streams (NII 65-80%, fees 10-20%, treasury 5-15%)
- ○I can explain why CASA ratio affects NII
- ○I know to strip treasury gains and focus on Core PPOP
- ○I understand the bank P&L structure differs from manufacturing
Key takeaway
Banks earn from 3 streams: NII (the core spread business), fee income (capital-light and growing), and treasury (volatile — ignore for quality assessment). Focus on NII growth and Core PPOP to judge a bank's true operating performance.
2. Banking — Asset Quality Deep Dive
GNPA and NNPA — the formulas and what they really measure Slippage ratio vs stock NPA — flow vs level 4 ways banks can hide bad loans How to calculate total...
Full sector playbook is a Pro feature
Sign up to continue learning — unlock the full path, in-depth lessons, projects, and every tool on the platform.
3. Banking — NIM, CASA & Efficiency Metrics
NIM — what drives it up and down CASA ratio — the cheapest funding advantage CosttoIncome ratio — operational efficiency How rate cycles affect all three met...
Full sector playbook is a Pro feature
Sign up to continue learning — unlock the full path, in-depth lessons, projects, and every tool on the platform.
4. Banking — Provisioning & Credit Cost
How provisioning works and why it's the biggest P&L wildcard Credit cost formula and benchmarks Provision Coverage Ratio (PCR) — conservative vs aggressive b...
Full sector playbook is a Pro feature
Sign up to continue learning — unlock the full path, in-depth lessons, projects, and every tool on the platform.
5. Banking — Valuing Banks (PB-ROE Framework)
Why PE ratio fails for banks The PBROE valuation framework The Gordon Growth PB derivation Case study: HDFC Bank vs Yes Bank (FY19) — the value trap lesson |...
Full sector playbook is a Pro feature
Sign up to continue learning — unlock the full path, in-depth lessons, projects, and every tool on the platform.