Price-to-Book (P/B) Ratio Explained

The Price-to-Book ratio compares a company's market value to the accounting value of its equity. It's the oldest valuation metric in finance, used by Benjamin Graham himself, and it still matters today for banks, insurers, and asset-heavy businesses where the balance sheet is the business. For everything else, especially modern software and brand-driven companies, P/B is mostly noise.

What it measures

The formula:

P/B Ratio = Share Price ÷ Book Value Per Share

Or equivalently:

P/B = Market Capitalization ÷ Total Shareholders' Equity

Book value is what's left for shareholders on the balance sheet after subtracting all liabilities from all assets. It's an accounting construct: assets recorded at historical cost (sometimes marked to market, often not), minus debts and obligations.

If $JPM has a market cap of $610B and shareholders' equity of $340B, the P/B is approximately 1.8. Translation: the market values JPMorgan at 1.8 times the accounting equity sitting on its books.

A P/B of 1.0 means the market values the company exactly at its accounting net worth. Below 1.0 means the market thinks the assets are worth less than reported, the business is destroying value, or there's hidden risk on the balance sheet. Above 1.0 means the market sees value beyond what accounting captures, brand, intellectual property, future earnings power, or simply optimism.

A close variant is Price-to-Tangible-Book, which strips out goodwill and intangible assets. For acquisitive companies that have written huge goodwill onto their balance sheets, tangible book value is the more honest version. A bank with $100B of equity but $30B of goodwill from past M&A really has $70B of tangible equity backing its operations.

The metric's deep flaw, in modern markets, is that book value undercounts almost everything that drives value in today's economy. R&D gets expensed, not capitalized. Brands built over decades are recorded at zero. Software, code, network effects, customer relationships, none of it shows up. This is why $MSFT trades at 12x book and $V trades at 14x book; the book value isn't the business.

How to use it in practice

P/B is genuinely useful in a narrow set of cases: banks, insurers, REITs, asset managers, and capital-intensive industrials with hard physical assets. In these businesses, the balance sheet largely is the business, and book value approximates economic value reasonably well.

Banks are the textbook case. $JPM, $BAC, $GS, and other major banks are valued primarily on P/B, with the multiple expanding when ROE rises and contracting when credit cycles deteriorate. The relationship is mechanical: a bank generating 15% ROE deserves to trade above book value; a bank generating 5% ROE deserves to trade below book value. The simple formula many bank investors use is that P/B should roughly equal ROE divided by the cost of equity. A bank with a 12% ROE and a 10% cost of equity should trade at around 1.2x book.

Real numbers help. $JPM has traded between 1.4x and 2.0x book over the past decade, expanding when ROE pushes above 15%. $BAC has historically traded at lower multiples (0.8-1.4x) reflecting structurally lower returns. $GS, despite being widely considered the best investment bank, often trades around 1.0-1.4x book because its ROE is volatile and capital markets revenue is cyclical. The discipline of the metric forces investors to confront the relationship between returns and valuation.

Insurance follows similar logic. $BRK.B is valued partly on book value because, accounting-wise, Berkshire is a holding company. Buffett himself used P/B as a benchmark for years before pivoting to intrinsic-value language as Berkshire's business mix shifted toward operating companies and away from securities.

Sector benchmarks vary dramatically:

• Banks: 0.7-2.0x is the normal range. Below 0.7x flags distress. Above 2.0x flags exceptional returns or speculative excess.
• Insurers: 0.8-1.5x typically.
• REITs: 0.8-1.3x of NAV (a real-estate adjusted version of book).
• Asset-heavy industrials: 1.0-3.0x.
• Software, consumer brands, services: 5x+ is normal and the metric is largely meaningless.

For practical screening: P/B under 1.0 in financials demands an investigation, sometimes a hidden gem, often a value trap. P/B between 1.0 and 1.5 in financials with double-digit ROE is the sweet spot. P/B above 3.0 in non-financials is acceptable only with strong returns on capital.

Common mistakes

Using P/B on asset-light businesses. Pulling up a P/B screen and finding $MSFT at 12x or $V at 14x doesn't mean they're expensive. It means book value isn't the right yardstick. Software companies have minimal physical assets, and their accounting equity bears little relationship to their economic value. Skip P/B entirely for these names.

Ignoring goodwill bloat. A company that's done multiple large acquisitions can show a P/B of 1.5x while its tangible book P/B is 4x or higher. The goodwill represents premiums paid in past deals that may or may not have been worth it. For acquisitive companies, always check tangible P/B alongside reported P/B. If goodwill exceeds 30% of total assets, the standard P/B ratio is misleading.

Treating low P/B as automatically cheap in financials. A bank trading at 0.6x book usually deserves it. The market is pricing in either unrecognized credit losses, structural ROE problems, or governance issues. Genuine bargains at 0.6x book are rare; value traps are common. Always check the trajectory of ROE, the credit metrics, and recent provisioning before assuming the discount is unwarranted.

ACCE perspective

P/B is not in our core Value Score formula, which is a deliberate choice. The score weights trailing PE, forward PE, EV/EBITDA, price-to-sales, and FCF yield because these metrics are applicable across sectors. P/B works only for financials and asset-heavy industrials, so applying it universally would unfairly skew scores against modern asset-light businesses.

For the financials and insurers in our curated universe, we layer P/B in as a sector-specific overlay rather than a universal score input. A bank with a 1.2x P/B and 14% ROE scores well on our internal financial sector model. A software company with a 12x P/B is evaluated entirely on other metrics, and the P/B is ignored.

We also track tangible P/B separately for acquisitive companies. When a stock's reported P/B looks reasonable but tangible P/B is 3x higher, our financial model flags it for closer review. The gap between book and tangible book is one of the cleaner signals of accounting risk in our coverage universe. To find financials trading at reasonable book multiples with strong returns, use the Undervalued Quality preset filtered to the financial sector.