Metrics
Price-to-book explained: what book value counts, what it misses, and when the ratio works
What price-to-book measures, why Apple trades at 46.55 times book and General Motors at 1.12, how payouts shrink book value, and where the ratio still earns its place.
The figures and scores in this guide are those of October 14, 2026. The comparison pages it links to are refreshed more often, most recently with closing prices of November 9, 2026, and the scoring rules were revised on November 10, 2026, so a figure or a score on a linked page can differ from the one quoted here.
Price-to-book compares a company's market value with its book value — the shareholders' equity on its balance sheet, which is everything the company owns minus everything it owes. A price-to-book of 1 means the market values the company at exactly what its accounts say the owners' stake is worth. Microsoft, with $442.39bn of equity and a market value of $3,962.9bn, trades at 8.96 times book. It is one of the four measures our valuation score votes on, and it is the oldest of them, from an era when a company's value lay mostly in factories, land and inventory that a balance sheet could count. All figures below use the latest balance sheet in each company's filings, with market values at the 14 October 2026 close.
The range, from 52.82 to 1.12
| Company | Price/book | Return on equity |
|---|---|---|
| Palantir | 52.82 | 38.19% |
| Apple | 46.55 | 148.75% |
| Eli Lilly | 33.59 | 102.44% |
| Nvidia | 23.29 | 117.21% |
| Coca-Cola | 11.74 | 44.64% |
| Microsoft | 8.96 | 34.04% |
| Meta | 7.24 | 29.85% |
| Exxon Mobil | 2.81 | 12.38% |
| JPMorgan | 2.38 | 17.79% |
| Disney | 1.71 | 7.85% |
| Ford | 1.31 | −14.82% |
| Bank of America | 1.19 | 11.23% |
| General Motors | 1.12 | 6.03% |
Read down the two columns together and a broad pattern appears: the companies that earn more on their equity tend to trade at higher multiples of it, though not in strict order — Palantir's P/E of 171.77 puts it at the top on price alone, and Ford, which lost money, trades above General Motors. That broad link is the most useful thing price-to-book can tell you, and it is also why the ratio on its own says so little about whether a share is cheap.
What book value leaves out
Accounting puts things on the balance sheet when they are bought, at what they cost. What a company builds for itself mostly never appears there. Coca-Cola's brand and Nvidia's chip designs were created through years of marketing and research, and those costs were charged against profit as they were spent rather than recorded as assets. So the balance sheet of a company whose value lies in brands, code or know-how understates what it owns, often by a wide margin.
That is why businesses built on intangibles trade far above book. Palantir at 52.82 times book and Coca-Cola at 11.74 are not necessarily expensive on that account; their book value simply measures a small part of what the market is paying for. A purchased asset is different: when a company buys another, it records the target's brands, technology and other identifiable assets at fair value, and anything paid above that as goodwill, so an acquisitive company carries more of its intangible value on the balance sheet than one that grew on its own. Two similar businesses can therefore show quite different book values depending on how they were built.
How buybacks and dividends shrink the book
The second distortion comes from returning cash. Every dollar a company spends on dividends or buying back its own shares leaves the balance sheet, and when a company returns more than it earns for years, its book stays small even as the business grows. Apple is the clearest case: a market value of $5,005.2bn against equity of $107.52bn puts it at 46.55 times book, a figure that reflects that long history of payouts.
The result is that a company with a long record of large payouts can look extremely expensive on book value without its price being any higher relative to its earnings. Return on equity rests on the same small base, which is why Apple's comes out at 148.75%; our guide to ROE versus ROIC explains why return on invested capital is the cleaner comparison for companies like these.
What a price near book value says
At the other end of the table, General Motors trades at 1.12 times book and Ford at 1.31. For businesses like these — capital-heavy, with most of their value in plants, vehicles and loans to customers — book value is a more meaningful anchor, and a price close to it says the market expects the assets to earn roughly what investors require of them. The returns column is mixed: General Motors earned 6.03% on its equity and Ford lost money, yet Ford trades the higher of the two. Our carmakers study looks at what lies behind those returns.
Disney, at 1.71 times book with a return on equity of 7.85%, sits in similar territory, while Meta at 7.24 times book earns 29.85%. The gap between them in price-to-book is mostly a gap in what each earns on its capital, not a judgement that one is cheap and the other dear.
The link to earnings
Price-to-book is roughly the P/E multiplied by the return on equity, since price over earnings times earnings over equity leaves price over equity. Exxon Mobil shows it working: a P/E of 22.78 and a return on equity of 12.38% give about 2.8, against a measured price-to-book of 2.81. The match is only approximate in general, because return on equity uses the average equity over the year and earnings per share the average share count, while price-to-book uses the latest balance sheet and current shares; for a company whose equity moved a lot during the year, such as Apple, the two can sit far apart. The equation still makes the main point: a high price-to-book is what a high return on equity looks like at an ordinary P/E.
Where the ratio earns its place
Banks are where price-to-book matters most, because a bank's assets are mostly loans and securities, recorded at amounts that follow their value far more closely than a brand or a factory does, so its book value is a real measure of its capital. JPMorgan, earning 17.79% on its equity, trades at 2.38 times book; Bank of America, earning 11.23%, trades at 1.19. Our guide to comparing bank stocks works through how the two numbers fit together.
It also matters when book value disappears. Lucid ended its latest period with negative equity of −$1.06bn, so a price-to-book would be a negative number with no sensible reading, and we leave it blank — one of the rules set out in the numbers we refuse to print.
How our score uses it
Price-to-book is one of four valuation votes, with EV/EBITDA, price-to-sales and free cash flow yield, and the cheaper company on each measure wins that vote. Because the vote counts the ratio as it is, it can favour a company for reasons that have nothing to do with value. In Nvidia against AMD, AMD wins the price-to-book vote at 13.86 against 23.29, largely because Nvidia's far higher returns lift its multiple of book. Nvidia wins the other three votes — EV/EBITDA, price-to-sales and free cash flow yield — and still takes valuation; the six-factor score explained sets out every rule.
Read well, price-to-book is a measure of returns as much as of price: use it to compare capital-heavy businesses and banks, read it next to the return on equity, remember that buybacks and home-grown intangibles push it up, and never treat a high figure for an asset-light company as proof that the share is dear.
Comparisons that use these companies
Related guides
- Revenue growth vs EPS growth: why they diverge and which to trustHow three-year growth rates are calculated, why earnings can grow ten times faster than sales or shrink while sales rise, and what Salesforce, Meta and Tesla show.
- Dividends vs buybacks: how companies hand cash back, and how to judge itDividend yield, payout ratio and buyback yield explained with real companies, why a high yield can be a warning, and how to tell whether a payout is sustainable.
- Balance-sheet risk: net debt, interest cover and the current ratioThe three numbers that show whether a company can carry its debt, how to read each one, and what they reveal about Pfizer, Intel, Disney and Palantir.
TickerVerdict provides factual data comparisons for informational purposes only. Nothing here is investment advice or a recommendation to buy or sell any security. Figures may be delayed; verify with your broker before investing. Capital is at risk.