Recognizing Value Stock Candidates: P/E, P/S, and P/B

Why a ratio exists at all

Video coming soon

This lesson explains the idea in full without it.

If you've applied for a loan, you've already been on the receiving end of this. A lender doesn't ask how much debt you carry, because the answer alone tells them nothing. They ask how much debt you carry relative to what you earn — your debt-to-income ratio.

Applicant Two earns more and owes more. Bigger in both directions. Look only at income and she's the stronger borrower; look only at debt and she's the weaker one. The ratio settles it: every dollar she earns is carrying 40 cents of debt, against 25 cents for Applicant One. Higher income, tighter squeeze.

Figure

That's the move you're about to make on companies. A ratio doesn't tell you a company is good. It tells you how one part of it sits against another, on a scale you can hold next to a different company entirely.

Which is also the limit of the thing. Nobody sensible buys a stock because a ratio was low. Ratios narrow a list of thousands down to a handful worth real work. They start the analysis. They don't finish it.

Two ways to read the same number

A ratio by itself is a fact with no context. It only means something next to another ratio, and there are two things you can hold it next to.

Time-series analysis compares a company's ratio to its own past. Take net profit margin — the share of every sales dollar that survives as profit. If UVWXYZ kept 6 cents of each dollar three years ago and keeps 9 cents now, something in that business got better. If it went the other way, you want to know why before you go further.

Cross-sectional analysis compares a company's ratio to a similar company, or to the average for its industry. If UVWXYZ keeps 9 cents per dollar and every rival keeps 4, UVWXYZ is doing something its competitors can't. That's the kind of clue that's invisible when you look at a company alone.

Use both. A ratio improving over time might still be dismal for the industry. A ratio that beats every peer might be sliding year after year. Neither view is complete without the other.

Financial statements can yield hundreds of ratios, grouped into families: profitability (does it make money), liquidity ratios (can it pay its bills soon), solvency (can it survive its debts), efficiency (how well it uses what it has), and valuation (what are you paying for all of the above).

This lesson is about the last family. A valuation ratio puts the stock's price on top and something the business actually produces on the bottom. Three of them do most of the work for value investors: price against earnings, price against sales, and price against what the company owns.

Price-to-earnings: what you pay per dollar of profit

Start with profit, since that's what you own a claim on.

Earnings is a company's profit. Spread across every share, it becomes earnings per share (EPS) — profit divided by shares outstanding.

Fictional UVWXYZ earns $2,000,000 in a year and has 1,000,000 shares. EPS is $2.00. Each share is entitled to two dollars of that year's profit. Investors watch EPS across several years, because a rising EPS is the baseline for expecting more later.

Put price over that and you get the price-to-earnings (P/E) ratio:

P/E  =  price per share  ÷  earnings per share

UVWXYZ trades at $30. Its P/E is $30 ÷ $2.00 = 15. Read it as a price tag: you're paying $15 for each $1 of annual profit.

Now the same arithmetic on a rival. Fictional ABCDEF also earns $2,000,000 on 1,000,000 shares — EPS of $2.00, identical to UVWXYZ — but trades at $60. ABCDEF's P/E is 30. Same profit, double the price. Buyers of ABCDEF are paying twice as much for the same dollar of earnings.

There's no universal number that counts as low, which is why the comparison matters more than the value. A P/E of 15 is unremarkable in one industry and remarkable in another. A P/E only means something next to something else: UVWXYZ against its own history, UVWXYZ against the companies it competes with, UVWXYZ against the broad market. Those baselines move — the market's typical P/E is higher in some decades than others — so look up where the market and the industry sit today rather than carrying a rule of thumb around. Most stock screeners and financial data sites publish current index and industry P/Es for free.

Trailing and forward

You'll hear P/E quoted two ways, and the difference is which earnings went into the denominator.

That second row is the whole point. A forward P/E is not a measurement — it's a measurement divided by a prediction. When a forward P/E looks attractively low, it's low because someone expects earnings to rise. If those earnings don't show up, the ratio was never real. Forecasts are useful and forecasts are wrong all the time. Know which one you're holding.

The crack in it

Earnings are the output of accounting decisions — when revenue gets recognized, how fast assets get depreciated, what counts as an expense this year versus next. Those choices are legal, and they're choices. A company making aggressive ones can report earnings that flatter it, which pushes its P/E down and makes it look like a bargain it isn't.

You can't fix that from outside the company. What you can do is refuse to let one ratio carry the whole argument — which is why the next two exist.

Price-to-sales: what you pay per dollar of revenue

Revenue, or sales, is the money coming in the door before any expense is subtracted. The price-to-sales (P/S) ratio compares the stock price to annual sales per share:

P/S  =  price per share  ÷  annual sales per share

UVWXYZ sells $20,000,000 a year across 1,000,000 shares — $20 of sales per share. At $30, its P/S is 1.5. You're paying $1.50 for each $1 of annual sales. Lower means paying less per dollar of revenue, so value investors generally lean low here too.

It looks redundant next to the P/E. It isn't, and the reason is worth sitting with. Sales and earnings are separated by everything the company spends. Put the two rivals side by side:

ABCDEF is cheaper on sales and more expensive on earnings. Both are true, and together they say something neither says alone: ABCDEF moves three times the revenue per share and converts none of that advantage into profit. Something between the top line and the bottom line is eating it — costs, competition, a business that has to spend heavily to sell anything. UVWXYZ sells less and keeps more of it.

That is what fundamental analysis is, mostly. Not one revealing number. Two ordinary numbers that disagree.

The crack in it

Sales swing. A company's revenue can move sharply from one year to the next — or one quarter to the next — on a big contract, a lost customer, a product cycle, a soft economy. And because sales sit in the denominator, the P/S moves whether or not anything happened to the price.

Say UVWXYZ's sales fall from $20,000,000 to $12,000,000 next year. The stock hasn't moved; it's still $30. Sales per share is now $12, so the P/S goes from 1.5 to 2.5. Nothing about the stock changed. It just got 67% more expensive by this measure, because the measure moved underneath it.

That cuts the other way too — a strong year can make a stock look cheaper on sales without a single thing improving about its price. So read the P/S across several years rather than trusting one. A single year's P/S tells you about that year, and years are noisy.

Price-to-book: what you pay per dollar of what's owned

P/E and P/S both come off the income statement — the record of what flowed through the business over a period. The last ratio comes from the balance sheet, which is a snapshot of what the company owns and owes right now.

Book value is assets minus liabilities: everything the company owns, less everything it owes. It's also called shareholders' equity, and the name explains the concept — it's the slice that would belong to shareholders if the company shut down tomorrow, sold everything, and paid off every debt.

If that sounds familiar, it should. It's home equity. Your house is worth $250,000 and you owe $180,000 on it, so your equity is $70,000. Same arithmetic, bigger balance sheet.

The price-to-book (P/B) ratio compares the stock price to book value per share:

P/B  =  price per share  ÷  book value per share

Figure

UVWXYZ holds $25,000,000 in assets and owes $10,000,000, so book value is $15,000,000 — $15 per share across its million shares. At $30, its P/B is 2.0. You're paying two dollars for each dollar of equity in the business. ABCDEF, at $60 a share on $10,000,000 of book value, has a P/B of 6.0. Value investors generally prefer a lower P/B: less paid per dollar of what the company actually owns.

Watch book value over time, too. The book value growth rate is the percentage change in shareholders' equity from year to year. A business steadily adding to its equity is building something. One whose equity erodes year after year is consuming what it has, and that shows up here before it shows up in a lot of other places.

The P/B is worth having precisely because it looks at a different thing. P/E and P/S ask what the business produced over the last year. The P/B asks what it has. It's most informative between companies of similar age and size in the same industry — a young company and a hundred-year-old one accumulate equity on entirely different clocks.

The cracks in it

Two of them, and they're serious.

Book value ignores intangibles. A brand people will cross town for, a patent portfolio, software written years ago, a reputation that wins contracts — these can be the most valuable things a company has, and the balance sheet barely acknowledges them. So companies whose worth is mostly ideas will show a small book value and a high P/B almost by construction. That's not evidence they're overpriced. It's evidence the ratio can't see what they're made of.

Book value reflects accounting policy. What an asset is carried at on the balance sheet is the result of rules and choices — original cost less depreciation, generally, not what the thing would fetch today. Two companies holding genuinely identical assets can report different book values because they recorded them differently. Factory equipment written down to nearly nothing might run fine for another decade. The number is real; it just isn't the same thing as economic value.

Neither flaw makes the P/B useless. It makes it one of three, which was always the arrangement.

The same three ratios on a real company

UVWXYZ and ABCDEF were invented so the arithmetic would stay clean. Here's what happens the moment it isn't.

The Campbell's Company — the soup — trades under the ticker CPB. For its fiscal year 2025 it reported diluted earnings per share of $2.01 under GAAP, the standard accounting rulebook every U.S. public company files by. (Diluted means the share count behind the division also includes shares the company is already committed to creating — through stock compensation and the like — so the profit is spread across every share that could soon exist, not just the ones that do.) The same report offers a second figure: adjusted EPS of $2.97, which is earnings with items management considers one-time — an impairment charge, restructuring costs, and the like — added back. Both figures describe the same fiscal year. On July 31, 2026, the stock closed at $21.98. Now do the division you just learned:

The other two ratios, same company, same price. Fiscal 2025 net sales were $10,253 million across 300 million diluted shares — $34.18 of sales per share — so the P/S is $21.98 ÷ $34.18 = 0.64. The most recent balance sheet before the snapshot (May 3, 2026) shows total shareholders' equity of $4,028 million; the release never prints book value per share, so you compute it yourself — about $13.47 on the roughly 299 million diluted shares that release reports, a count that had drifted just below fiscal 2025's 300 million — and the P/B comes out near 1.6.

One more thing worth watching in real life. Is a P/E of 7.4 low? Low enough to catch a value screener's eye — and the warning above said a low P/E usually has a reason. Here you can see one in the open: at the snapshot date, Campbell's own published guidance projected adjusted earnings to fall 23% to 26% for fiscal 2026. The denominator everyone was dividing by was one the company itself said was shrinking. That's not a footnote to the low ratio. It usually is the low ratio.

None of this says whether CPB was a bargain. That question needs the next two lessons, and honestly answering it needs more than three divisions. What the real numbers add is the part UVWXYZ couldn't teach: every ratio you'll ever look up was built from figures somebody chose, starting with which earnings number went in the denominator. Later in this unit you'll run a full valuation where choosing between that $2.01 and that $2.97 is the first step of the model.

Where this leaves you

You now have three prices for the same stock. What you pay per dollar of profit, per dollar of sales, and per dollar of what the company owns. Each is a real fact about the business. Each is also blinkered — earnings bend to accounting, sales lurch year to year, book value can't see an idea.

Read together, and against the right comparisons, they narrow a market of thousands to a list of a few worth genuinely studying. That's the job. A candidate isn't a conclusion — the next lesson takes what's on the list and asks what the business is actually worth.

Key takeaways

• A financial ratio divides one number by another so companies of wildly different sizes can be compared on the same scale — the same reason a lender looks at your debt against your income instead of your debt alone.
• A ratio only means something in comparison: against the company's own history (time-series) and against similar companies or its industry (cross-sectional). Both, or neither.
• P/E is what you pay per dollar of annual profit; P/S per dollar of sales; P/B per dollar of the company's assets minus its debts. Value investors lean toward low readings on all three.
• A low P/E is not the same as cheap. It usually means the market expects those earnings to fall, and it's often right. A ratio starts your analysis; it never ends it.
• Every one of these ratios has a blind spot. Earnings reflect accounting choices — a real company reports a GAAP and an adjusted EPS, so it carries two P/Es before you've decided anything — sales swing hard year to year, and book value ignores brands, patents, and everything else you can't put on a balance sheet. Use all three because each covers for the others.

• A ratio is one number about a company divided by another. Dividing throws away how big the company is and keeps how its numbers sit against each other, so a tiny company and a giant one can be compared at all.
• A ratio on its own means nothing. It only says something next to another one — the same company in earlier years, or other companies doing the same kind of work. You need both comparisons, not one.
• The three in this lesson ask three different questions: what you pay for each dollar the company earns, for each dollar it sells, and for each dollar of what it owns after its debts. Bargain hunters want all three to come out low.
• A low number does not mean cheap. Usually it means people expect the company to earn less soon — and often they turn out to be right. This is where your work starts, never where it ends.
• Each of these three has something it cannot see. The profit number depends on choices somebody made, sales lurch around from year to year, and what a company "owns" leaves out things like a famous name or a great idea. That's why you use all three.