When P/E breaks: the other multiples

By the end of this lesson

Five alternative multiples and exactly when each one is the right choice
Why enterprise value is the honest way to compare indebted companies
How to pick a multiple to fit the business rather than the habit

Picture this

You wouldn't weigh a truck on a kitchen scale

P/E is a superb instrument for a stable, profitable, moderately-levered company. Point it at a loss-making start-up, a bank, or a company carrying five times its equity in debt, and it produces a number that is either meaningless or actively misleading.

That's not a flaw in P/E. It's a flaw in using one instrument for every measurement.

Price to book — for banks and asset-heavy businesses

P/B

P/B = Market cap ÷ Shareholders' equity = Share price ÷ Book value per share

Reads as: how many rupees am I paying for each rupee of net assets on the books? Only meaningful where book value means something — banks, NBFCs, insurers, and heavily asset-based businesses.

For a bank, the balance sheet essentially is the business: loans and deposits, marked at realistic values, updated constantly. So book value is a genuine measure of scale. For a software company whose real assets are people and code, book value is close to fiction — which is why P/B is nearly useless there.

Worked exampleWhy P/B must be read with ROE

Two banks. P/B alone would tell you the wrong story.

Bank A — P/B: 2.9×
Bank A — ROE: 18%
Bank B — P/B: 0.7×
Bank B — ROE: 5%

Bank A: pays 2.9× book for equity earning 18% → implied earnings yield 18/2.9 = 6.2%

Bank B: pays 0.7× book for equity earning 5% → implied earnings yield 5/0.7 = 7.1%

But B's 5% ROE is below its cost of capital — its book value is arguably worth LESS than stated

So → P/B and ROE must always be read together. A bank earning 18% on equity deserves to trade well above book; one earning 5% deserves to trade below it. A low P/B with a low ROE is not a discount — it is the market correctly pricing poor returns.

EV/EBITDA — comparing companies with different debt loads

Enterprise value

EV = Market cap + Total debt − Cash

Market cap: the price of the equity
+ debt: you inherit the obligations when you buy the company
− cash: you also get the cash, so it reduces the true cost

EV is what it would actually cost to buy the whole business outright — equity and the debt that comes with it. It's the number an acquirer thinks in.

Worked exampleTwo companies that look identical and aren't

Both have a market cap of ₹1,000 Cr and EBITDA of ₹150 Cr.

Company A — market cap: ₹1,000 Cr
Company A — debt / cash: ₹0 / ₹200 Cr
Company B — market cap: ₹1,000 Cr
Company B — debt / cash: ₹900 Cr / ₹50 Cr

Company A: EV = 1000 + 0 − 200 = ₹800 Cr → EV/EBITDA = 800/150 = 5.3×

Company B: EV = 1000 + 900 − 50 = ₹1,850 Cr → EV/EBITDA = 1850/150 = 12.3×

So → Same market cap, same EBITDA — and B costs more than twice as much for the same operating profit, because a buyer inherits ₹900 Cr of debt. P/E would have missed this entirely. This is the single best reason to know EV.

Worth knowing

Where EV/EBITDA earns its place

Capital-intensive businesses — cement, telecom, hotels, infrastructure — where depreciation policies differ so much between companies that net profit isn't comparable. EBITDA sits above depreciation and financing, so it compares operations directly.

The caveat matters though: EBITDA is not cash flow. It ignores the very real cost of assets wearing out. For a business that must spend heavily every year just to stand still, EBITDA systematically overstates what owners actually receive.

The rest of the toolkit

Which multiple, and when

P/E
Use it when: Stable, profitable, moderate debt
Watch out for: One-offs, cyclicals, loss-makers
P/B
Use it when: Banks, NBFCs, insurers, asset-heavy
Watch out for: Meaningless where assets are intangible; always pair with ROE
EV/EBITDA
Use it when: Comparing across different debt levels; capital-intensive sectors
Watch out for: Ignores depreciation and capex — flatters asset-hungry businesses
EV/EBIT
Use it when: Same as above, but honest about depreciation
Watch out for: Less widely quoted, so fewer ready comparisons
P/S or EV/Sales
Use it when: Loss-making or early-stage companies with real revenue
Watch out for: Revenue without a path to profit is worth very little
Price / free cash flow
Use it when: Mature, cash-generative businesses
Watch out for: Lumpy capex makes single years unreliable — average several
Dividend yield
Use it when: Mature payers; income-focused holdings
Watch out for: A high yield often means the market expects the dividend to be cut
PEG
Use it when: Growth companies, as a rough sanity check
Watch out for: Depends entirely on a growth forecast; garbage in, garbage out

Multiple	Use it when	Watch out for
P/E	Stable, profitable, moderate debt	One-offs, cyclicals, loss-makers
P/B	Banks, NBFCs, insurers, asset-heavy	Meaningless where assets are intangible; always pair with ROE
EV/EBITDA	Comparing across different debt levels; capital-intensive sectors	Ignores depreciation and capex — flatters asset-hungry businesses
EV/EBIT	Same as above, but honest about depreciation	Less widely quoted, so fewer ready comparisons
P/S or EV/Sales	Loss-making or early-stage companies with real revenue	Revenue without a path to profit is worth very little
Price / free cash flow	Mature, cash-generative businesses	Lumpy capex makes single years unreliable — average several
Dividend yield	Mature payers; income-focused holdings	A high yield often means the market expects the dividend to be cut
PEG	Growth companies, as a rough sanity check	Depends entirely on a growth forecast; garbage in, garbage out

PEG — price/earnings to growth

PEG = P/E ÷ earnings growth rate (%)

A company at 30× growing 30% has a PEG of 1.0. Below 1 is traditionally read as reasonable, above 2 as demanding. Treat it as a conversation-starter, not a valuation — it is only as reliable as the growth number you feed it, and it quietly assumes growth continues at that rate.

Where people lose money

The high dividend yield that is a warning

Yield is dividend ÷ price. If the price collapses, the yield spikes. A stock showing 11% yield is usually not generous — the market is telling you it expects the dividend to be cut, because the earnings behind it are going.

Check two things before trusting a yield: is the dividend covered by free cash flow, and has the payout been maintained through a bad year? An uncovered dividend funded by borrowing is a countdown, not an income stream.

Choosing a multiple in practice

1Is it a bank, NBFC or insurer? → P/B with ROE, plus P/E. Never EV-based.
2Is it loss-making but growing with real revenue? → EV/Sales, and a hard look at the path to profitability.
3Is debt large or very different from peers? → EV/EBITDA or EV/EBIT, not P/E.
4Is it capital-intensive with heavy depreciation? → EV/EBIT, and check free cash flow separately.
5Is it mature and cash-generative? → Price to free cash flow, and dividend yield.
6Is it stable and normally profitable? → P/E is fine, and simplest is best.

The one thing

Use at least two multiples, always. When they agree, you have a view. When they disagree, the disagreement is itself the finding — go and understand it.

In one screen

1P/B belongs to banks and asset-heavy businesses, and is only meaningful next to ROE.
2EV = market cap + debt − cash: the real cost of buying a business, and the only fair way to compare different debt levels.
3EBITDA ignores the cost of assets wearing out — never treat it as cash flow.
4A very high dividend yield is usually a warning about the dividend.
5Use two multiples; treat any disagreement between them as information.

Check yourself

Nothing is scored or stored — the explanation after each answer is the point.

1/2

Two companies both have ₹2,000 Cr market cap and ₹300 Cr EBITDA. One has zero debt; the other has ₹1,500 Cr of debt. What does EV/EBITDA reveal?

2/2

A bank trades at 0.6× book with ROE of 6%. What's the most likely explanation?

Next lesson →

Why cheap is sometimes the most expensive

← Previous

P/E: how many years to get your money back

← All lessons

By the end of this lesson

Five alternative multiples and exactly when each one is the right choice
Why enterprise value is the honest way to compare indebted companies
How to pick a multiple to fit the business rather than the habit

Picture this

You wouldn't weigh a truck on a kitchen scale

That's not a flaw in P/E. It's a flaw in using one instrument for every measurement.

Price to book — for banks and asset-heavy businesses

P/B

P/B = Market cap ÷ Shareholders' equity = Share price ÷ Book value per share

Reads as: how many rupees am I paying for each rupee of net assets on the books? Only meaningful where book value means something — banks, NBFCs, insurers, and heavily asset-based businesses.

Worked exampleWhy P/B must be read with ROE

Two banks. P/B alone would tell you the wrong story.

Bank A — P/B: 2.9×
Bank A — ROE: 18%
Bank B — P/B: 0.7×
Bank B — ROE: 5%

Bank A: pays 2.9× book for equity earning 18% → implied earnings yield 18/2.9 = 6.2%

Bank B: pays 0.7× book for equity earning 5% → implied earnings yield 5/0.7 = 7.1%

But B's 5% ROE is below its cost of capital — its book value is arguably worth LESS than stated

EV/EBITDA — comparing companies with different debt loads

Enterprise value

EV = Market cap + Total debt − Cash

Market cap: the price of the equity
+ debt: you inherit the obligations when you buy the company
− cash: you also get the cash, so it reduces the true cost

EV is what it would actually cost to buy the whole business outright — equity and the debt that comes with it. It's the number an acquirer thinks in.

Worked exampleTwo companies that look identical and aren't

Both have a market cap of ₹1,000 Cr and EBITDA of ₹150 Cr.

Company A — market cap: ₹1,000 Cr
Company A — debt / cash: ₹0 / ₹200 Cr
Company B — market cap: ₹1,000 Cr
Company B — debt / cash: ₹900 Cr / ₹50 Cr

Company A: EV = 1000 + 0 − 200 = ₹800 Cr → EV/EBITDA = 800/150 = 5.3×

Company B: EV = 1000 + 900 − 50 = ₹1,850 Cr → EV/EBITDA = 1850/150 = 12.3×

Worth knowing

Where EV/EBITDA earns its place

The rest of the toolkit

Which multiple, and when

P/E
Use it when: Stable, profitable, moderate debt
Watch out for: One-offs, cyclicals, loss-makers
P/B
Use it when: Banks, NBFCs, insurers, asset-heavy
Watch out for: Meaningless where assets are intangible; always pair with ROE
EV/EBITDA
Use it when: Comparing across different debt levels; capital-intensive sectors
Watch out for: Ignores depreciation and capex — flatters asset-hungry businesses
EV/EBIT
Use it when: Same as above, but honest about depreciation
Watch out for: Less widely quoted, so fewer ready comparisons
P/S or EV/Sales
Use it when: Loss-making or early-stage companies with real revenue
Watch out for: Revenue without a path to profit is worth very little
Price / free cash flow
Use it when: Mature, cash-generative businesses
Watch out for: Lumpy capex makes single years unreliable — average several
Dividend yield
Use it when: Mature payers; income-focused holdings
Watch out for: A high yield often means the market expects the dividend to be cut
PEG
Use it when: Growth companies, as a rough sanity check
Watch out for: Depends entirely on a growth forecast; garbage in, garbage out

Multiple	Use it when	Watch out for
P/E	Stable, profitable, moderate debt	One-offs, cyclicals, loss-makers
P/B	Banks, NBFCs, insurers, asset-heavy	Meaningless where assets are intangible; always pair with ROE
EV/EBITDA	Comparing across different debt levels; capital-intensive sectors	Ignores depreciation and capex — flatters asset-hungry businesses
EV/EBIT	Same as above, but honest about depreciation	Less widely quoted, so fewer ready comparisons
P/S or EV/Sales	Loss-making or early-stage companies with real revenue	Revenue without a path to profit is worth very little
Price / free cash flow	Mature, cash-generative businesses	Lumpy capex makes single years unreliable — average several
Dividend yield	Mature payers; income-focused holdings	A high yield often means the market expects the dividend to be cut
PEG	Growth companies, as a rough sanity check	Depends entirely on a growth forecast; garbage in, garbage out

PEG — price/earnings to growth

PEG = P/E ÷ earnings growth rate (%)

Where people lose money

The high dividend yield that is a warning

Choosing a multiple in practice

1Is it a bank, NBFC or insurer? → P/B with ROE, plus P/E. Never EV-based.
2Is it loss-making but growing with real revenue? → EV/Sales, and a hard look at the path to profitability.
3Is debt large or very different from peers? → EV/EBITDA or EV/EBIT, not P/E.
4Is it capital-intensive with heavy depreciation? → EV/EBIT, and check free cash flow separately.
5Is it mature and cash-generative? → Price to free cash flow, and dividend yield.
6Is it stable and normally profitable? → P/E is fine, and simplest is best.

The one thing

Use at least two multiples, always. When they agree, you have a view. When they disagree, the disagreement is itself the finding — go and understand it.

In one screen

1P/B belongs to banks and asset-heavy businesses, and is only meaningful next to ROE.
2EV = market cap + debt − cash: the real cost of buying a business, and the only fair way to compare different debt levels.
3EBITDA ignores the cost of assets wearing out — never treat it as cash flow.
4A very high dividend yield is usually a warning about the dividend.
5Use two multiples; treat any disagreement between them as information.

Check yourself

Nothing is scored or stored — the explanation after each answer is the point.

1/2

Two companies both have ₹2,000 Cr market cap and ₹300 Cr EBITDA. One has zero debt; the other has ₹1,500 Cr of debt. What does EV/EBITDA reveal?

2/2

A bank trades at 0.6× book with ROE of 6%. What's the most likely explanation?

Next lesson →

Why cheap is sometimes the most expensive

← Previous

P/E: how many years to get your money back

← All lessons