PE ratio is the most famous valuation metric in investing. It's also completely useless for banks, infrastructure companies, and any business with heavy debt. Here's what analysts use instead — and when to use each.
When PE Ratio Fails You
The PE ratio (Price / Earnings) is elegant in its simplicity. But it breaks down in predictable ways:
Problem 1: Banks Banks don't have "cost of goods sold" the way manufacturers do. Their business is borrowing money cheaply and lending it expensively. Earnings swing dramatically based on how much they set aside for bad loans (provisioning). A bank's PE can look artificially low when it's under-provisioning — and look frighteningly high when it's aggressively building loan loss reserves. Neither reflects the true value of the business.
Problem 2: Infrastructure companies (Adani Ports, NTPC, ONGC) These businesses invest enormous amounts in long-lived assets — power plants, ports, pipelines. Accounting depreciates those assets over 20-30 years, creating large non-cash charges. High depreciation suppresses reported earnings dramatically. PE looks expensive; the business may not be.
Problem 3: Loss-making companies (Zomato in FY2022) Negative earnings = undefined PE. You literally cannot apply the ratio.
Problem 4: Heavily indebted companies Two companies might have identical operational earnings (EBITDA), but one has ₹50,000 crore in debt and the other has none. Their PE ratios will look similar, but the actual value proposition for equity shareholders is very different.
For each of these situations, there's a better metric: P/B Ratio for banks and asset-heavy businesses, EV/EBITDA for companies where debt levels differ significantly.
Book Value — What Is the Company Actually Worth?
Before we talk P/B ratio, you need to understand book value.
Book Value = Total Assets − Total Liabilities
It's the accounting value of the company if you sold everything it owned and paid off everything it owed. What's left is the shareholders' equity — also called net worth or book value.
Book Value Per Share = Book Value / Total Shares Outstanding
For a bank like SBI: Total assets are ₹60+ lakh crore (loans, investments, buildings). Total liabilities are deposits, borrowings, bonds. The difference is shareholders' equity — roughly ₹3-4 lakh crore.
Book value represents the minimum floor a company theoretically has in a liquidation scenario. It's not what the company is worth as a going concern — but it's an anchor.
Why book value matters more for some businesses: For a bank, the balance sheet IS the business. Loans are assets. Deposits are liabilities. The spread between them generates profit. Book value captures the size and quality of this business in a way earnings cannot.
For a software company like TCS, book value is almost meaningless — the real assets are intellectual property, client relationships, and talented employees. None of those appear at fair value on the balance sheet.
P/B Ratio — Formula, Sector Benchmarks, Interpretation
P/B Ratio = Stock Price / Book Value Per Share
Or equivalently: Market Capitalisation / Total Book Value
What it tells you:
- P/B of 1.0 → market values the company exactly at its book value (what you'd get in liquidation)
- P/B > 1.0 → market believes the company is worth more than its accounting value (premium for future earnings, brand, moats)
- P/B < 1.0 → market values the company below book (either a bargain or a business with structural problems)
Why P/B works for banks: A bank with P/B of 2.5 trades at 2.5× the value of its equity capital. For investors, this represents the "franchise value" — how much premium are you paying for the quality of the bank's loan book, its management, its brand, its CASA ratio?
Current benchmarks for Indian banks:
- HDFC Bank: P/B ~3.0-3.5 (premium for quality)
- SBI: P/B ~1.2-1.5 (PSU discount, but considered cheap vs history)
- Kotak Mahindra Bank: P/B ~3.5-4.0 (premium franchise)
- PSU banks broadly: P/B 0.8-1.5
The gap between SBI's P/B (1.3) and HDFC Bank's P/B (3.3) reflects the market's assessment of loan quality, management efficiency, CASA ratios, and bad loan risk. It's a quantified expression of the quality premium.
When P/B Below 1 Is a Bargain (And When It's a Trap)
P/B below 1 means the market values the company below what it's worth in liquidation. Sounds like a bargain. Sometimes it is. Often it isn't.
Genuine bargain scenario: A quality PSU bank temporarily trades at P/B 0.9 during a market panic. The loan book is clean, capital ratios are healthy, but sentiment is negative. Buying at P/B < 1 here gives a margin of safety.
Value trap scenario: A company's book value is inflated by assets that aren't worth what accounting says. Old factories worth ₹500 crore on the books but actually worth ₹100 crore. Loans on a bank's books that are actually bad (non-performing) but haven't been written down yet. In these cases, the "cheap" P/B is an illusion — book value itself is overstated.
The IL&FS lesson: IL&FS subsidiaries had book values that looked substantial. But the assets backing those book values were infrastructure projects that were never going to generate expected cash flows. The real book value — adjusted for realistic asset values — was far lower. P/B that looks cheap can be a trap if the underlying assets are impaired.
Enterprise Value — The "Acquisition Price"
Now let's move to the second major metric: EV/EBITDA.
EV = Market Capitalisation + Total Debt − Cash & Cash Equivalents
Market cap tells you what the stock market values the equity of the business. But if you were to acquire the entire company — buy all shares AND take on all its debt — you'd pay the enterprise value, not just the market cap.
Why subtract cash? If a company has ₹10,000 crore in cash on its balance sheet, you'd get that cash when you buy the company. So the "true" acquisition cost is reduced by the cash you're acquiring.
Example:
- Adani Ports market cap: ₹2,60,000 crore
- Total debt: ₹55,000 crore
- Cash: ₹12,000 crore
- Enterprise Value: ₹2,60,000 + ₹55,000 − ₹12,000 = ₹3,03,000 crore
EV is the true cost of acquiring control of the business — equity + debt, minus the cash you'd recover.
EV/EBITDA — The Debt-Neutral Valuation Ratio
EV/EBITDA = Enterprise Value / EBITDA
This ratio tells you: how many years of EBITDA would it take to pay off the enterprise value?
Why EV/EBITDA is superior to PE for debt-heavy companies:
Two companies, identical EBITDA of ₹1,000 crore:
- Company A: No debt, market cap ₹10,000 crore → EV/EBITDA = 10x → PE ≈ 14x
- Company B: ₹8,000 crore debt, market cap ₹6,000 crore → EV/EBITDA = 14x → PE ≈ 10x
On PE, Company B looks cheaper. But Company B has ₹8,000 crore in debt that Company A doesn't. The equity holder in Company B faces far more risk — interest payments, refinancing risk, potential dilution if the company needs to raise capital.
EV/EBITDA captures this. Company A at 10x is cheaper than Company B at 14x. PE was misleading you.
EV/EBITDA benchmarks for India:
- IT Services: 20-30x (high margins, zero debt)
- FMCG: 25-40x (premium for consistency)
- Infrastructure / Utilities: 12-18x
- PSU Banks: Not applicable (use P/B)
- Private Banks: Not applicable (use P/B)
- Auto: 8-14x
- Pharma: 15-25x
What the Numbers Still Can't Tell You
P/B tells you what you're paying relative to book value. EV/EBITDA tells you what you're paying relative to operating earnings. Neither can tell you whether the quality behind those numbers is real.
A bank's P/B of 1.2 looks cheap — but is the loan book actually clean, or are NPAs being masked? An infrastructure company's EV/EBITDA looks reasonable — but is management actually executing on projects, or are delays building up quietly?
This is where StockMirror's earnings page analysis fills the gap. When you open any company at stockmirror.in, the Earnings Quality section flags whether results were clean or impacted by one-time items. The Management Confidence signal and the Prepared Remarks vs Q&A divergence show whether management sounds assured when analysts press them on the numbers — or evasive. These are the signals that determine whether the valuation you calculated actually holds up.
Summary: PE vs P/B vs EV/EBITDA — When to Use Each
| Metric | Best For | Fails For |
|---|---|---|
| PE Ratio | Consumer, FMCG, IT, Pharma | Banks, Infrastructure, Loss-making |
| P/B Ratio | Banks, NBFCs, Asset-heavy cos | Asset-light businesses (IT, software) |
| EV/EBITDA | Comparing across debt levels, Infrastructure, M&A analysis | Banks (different capital structure) |
The practitioner's shortcut:
- Is it a bank/NBFC/insurance? → Use P/B
- Is it infrastructure or telecom (high debt, high depreciation)? → Use EV/EBITDA
- Is it loss-making? → Use Price/Sales or forward EV/EBITDA
- Everything else? → Start with PE, cross-check with EV/EBITDA
Never use just one metric in isolation. The best investors triangulate: if PE looks cheap, P/B looks cheap, AND EV/EBITDA looks cheap versus the sector — that's a genuine valuation signal. If only one metric looks cheap, dig deeper.
See the earnings quality signals behind any valuation → stockmirror.in
Next in the series: The Efficiency Story — ROE, ROCE & ROA
Frequently Asked Questions
What is the P/B ratio and what is a good value?
Price-to-Book (P/B) ratio = Market Price per Share ÷ Book Value per Share. Book value is assets minus liabilities. P/B below 1 means the stock trades below its net asset value — potentially cheap or distressed. P/B of 1–3 is typical for most sectors. Banks are most commonly valued on P/B because their earnings are tied directly to the quality and size of their balance sheet.
What does a P/B ratio below 1 mean?
It means the market values the company below its net assets on paper. This can be a genuine buying opportunity — or a signal that the market expects asset write-downs, hidden liabilities, or sustained low returns. The test: check ROE. A company with P/B below 1 and ROE above 12% is often genuinely undervalued. P/B below 1 with ROE below 8% is usually a value trap.
When should you use EV/EBITDA instead of PE ratio?
Use EV/EBITDA when comparing companies with different debt levels, for industries where depreciation significantly distorts PE (infrastructure, manufacturing, telecom), and for companies with losses at net profit level but positive EBITDA. It is the standard metric in M&A analysis because it shows what an acquirer pays for the full business including debt.
What is a good EV/EBITDA ratio for Indian companies?
8–12x EV/EBITDA is considered reasonable for most Indian sectors. Below 6x may indicate undervaluation. Above 15x typically reflects a growth premium. Specialty chemicals and consumer brands often trade at 18–25x due to pricing power and compounding margins. Always compare to sector peers and the company's own historical range.
Why can't you use PE ratio to value banks?
Banks earn their income from the spread between deposit and lending rates — interest income is their core operating line, not a financing cost. PE ratio treats interest as a financing expense (below EBIT), which makes the earnings figure structurally incomparable for banks. P/B and Return on Equity are the standard valuation metrics because they measure how efficiently the bank deploys its capital.
Disclaimer: This article is for educational purposes only and does not constitute investment advice. Always do your own research before making investment decisions.