TL;DR. The price-to-book (P/B) ratio compares what the market is paying for a share to what the accountants say the company’s net assets are worth per share. A P/B below one implies the market values the business at less than its liquidation-style equity; a P/B of ten implies the market is paying almost entirely for future earnings, not the balance sheet. P/B is the sharpest tool in finance for valuing banks, insurers and REITs — and one of the bluntest for valuing software, brands and services businesses that carry very little on the balance sheet.
What “book value” actually means
Book value is an accounting output, not a market number. It comes straight from the balance sheet identity that every public company reports in its 10-Q and 10-K filings with the SEC: Assets minus Liabilities equals Shareholders’ Equity. The SEC’s investor education glossary defines shareholders’ equity as “the residual claim after debts have been paid.” That residual claim, divided by the number of common shares outstanding, is book value per share.
Nothing about that number tells you what the assets are actually worth if you had to sell them today. Property carried at 1975 purchase cost, goodwill from a decade-old acquisition, patents amortised to zero, and inventory frozen at cost are all in there at whatever GAAP says, not at what the market would pay. That gap between the accounting number and any real-world liquidation value is the central limitation of P/B.
The formula
P/B is calculated in two equivalent ways:
- Per-share: Market Price per Share ÷ Book Value per Share
- Aggregate: Market Capitalization ÷ Total Shareholders’ Equity
Both produce the same number. The per-share form is what appears on screens and financial data terminals. The aggregate form is what analysts actually use when they run screens, because it’s robust to share splits, buybacks and repurchase-driven share count changes across periods.
A worked example: a mid-size bank
Pretend a regional bank trades at $52 per share. Its most recent 10-Q reports total shareholders’ equity of $12.0 billion and 240 million diluted shares outstanding. Its book value per share is $12.0 billion ÷ 240 million = $50.00. Its P/B is $52 ÷ $50 = 1.04.
Now filter out the goodwill and other intangible assets on that balance sheet — say, $2.0 billion of them. Tangible book value per share becomes ($12.0 billion − $2.0 billion) ÷ 240 million = $41.67. Price-to-tangible-book (P/TBV) is $52 ÷ $41.67 = 1.25. That’s the number bank analysts actually quote, because tangibles are what would survive a wind-down.
Both numbers together tell a specific story: the market is paying only a modest premium to what shareholders would theoretically walk away with after paying off the depositors and other creditors. That premium is roughly the present value of future returns above the bank’s cost of capital. For a bank earning at or below its cost of equity, the premium collapses to zero and P/B trades near 1.0 — which is exactly where the regional-bank sector sits today.
Sector snapshot: where P/B lands across US industries
| Sector (US) | P/B ratio | Why the number lives here |
|---|---|---|
| Banks (Regional) | 1.14 | Assets and liabilities are mostly financial and marked close to fair value |
| Utility (General) | 1.81 | Rate-regulated returns anchor P/B close to book value |
| Air Transport | 2.84 | Heavy PP&E on the books but volatile earnings |
| Advertising | 4.55 | Human-capital business — little tangible equity |
| Pharmaceutical | 6.64 | Patents amortized quickly relative to actual economic value |
| Aerospace/Defense | 7.88 | Long-cycle backlog and buybacks compress book value |
| Biotechnology | 8.22 | Pipeline value dwarfs balance-sheet assets |
| Software (System & Application) | 9.14 | Intangible IP and recurring revenue not reflected in equity |
| Semiconductor | 13.31 | Fabless models plus AI-cycle returns on relatively small equity |
| Computers/Peripherals | 34.08 | Massive buybacks (Apple, Dell) have driven equity toward zero |
| Restaurant/Dining | 74.74 | Franchise-heavy models (McDonald’s, Yum!) plus buybacks — equity is a rounding error |
| Retail (Building Supply) | 132.20 | Home Depot has repurchased so much stock its equity is near zero |
| Total US market | 4.61 | Aggregate across 5,994 US firms |
| Total US market ex-financials | 5.10 | Financials pull the aggregate down because banks anchor near 1× |
Why the spread is that wide
Two forces explain most of that dispersion. The first is intangibles: GAAP requires most internally-developed brand, IP and R&D to be expensed, not capitalized, so a company like Microsoft or Pfizer carries almost none of its real economic value on the balance sheet. The second is buybacks: when a company repurchases shares above book value, it directly reduces shareholders’ equity. A decade of aggressive buybacks at premium prices is why Home Depot’s P/B looks like a data-entry error. Neither force is a signal about the business; both are artifacts of accounting rules meeting corporate finance policy.
The S&P 500’s P/B over time
Two observations from the long history matter. First, aggregate P/B is a slow-moving valuation signal — it does not compress rapidly the way P/E can when earnings collapse, because book value itself is relatively sticky. Second, when P/B does compress hard, it is almost always alongside a recession or credit event: the 1980 stagflation lows, the 2002 tech bust, and the March 2009 low of 1.78 all sat below the long-run mean of 3.16. Today’s reading near 5.8 is, by that reference, historically expensive on a book-value basis.
The P/B-ROE relationship analysts actually use
P/B is not supposed to be interpreted in isolation. The classic bank-valuation formula, which follows from the Gordon growth model applied to equity, is roughly:
P / B ≈ (ROE − g) / (r − g)
where ROE is return on equity, g is the sustainable growth rate of book value, and r is the cost of equity. Plug in a bank earning a 12% ROE, growing 3% and demanding an 11% cost of equity, and you get roughly (12% − 3%) / (11% − 3%) = 1.13× book. A bank earning below its cost of equity produces a P/B below 1.0 — which is exactly what happens to distressed banks in credit downturns. The framework is standard undergraduate finance and is walked through in most CFA curriculum readings on equity valuation.
Common mistakes
- Using P/B on asset-light businesses. A software company’s P/B tells you almost nothing because its book value tells you almost nothing. The real assets are engineers, code and customer contracts, and none of those sit on the balance sheet.
- Ignoring goodwill. A company that has made big acquisitions carries goodwill on its balance sheet. That goodwill is not a “real” asset in a wind-down. Serious analysts strip it out and use price-to-tangible-book, especially for banks.
- Assuming P/B below 1 is automatically cheap. Sub-1.0 P/B is often a signal that the market expects future losses to eat into book value — that’s the “value trap” case. Regional banks in early 2023 were sub-1.0 not because the market was wrong, but because it was pricing in unrealized losses on their held-to-maturity bond portfolios.
- Ignoring the buyback distortion. Buying back shares above book value reduces shareholders’ equity. A rising P/B from buyback shrinkage is not the same signal as a rising P/B from earnings growth.
- Treating book value as market value. The oil and gas company’s reserves are on the books at cost less depletion, not at spot commodity prices. The insurance company’s long-tail liabilities are on the books at actuarial estimates that can prove badly wrong.
Related concepts
Once P/B is on your dashboard, three neighbors are worth knowing. Price-to-tangible-book (P/TBV) strips out goodwill and other intangibles and is the standard for banks, insurers and financials. Enterprise value to invested capital (EV/IC) extends the same idea to the whole capital stack, treating equity and debt together. Return on equity (ROE) is the earnings-side complement to book value: a high P/B is only justified when ROE meaningfully exceeds the cost of equity. Together, P/B and ROE tell you both how much you’re paying for a dollar of equity and how productively the company deploys that dollar.
For companies where P/B is unhelpful — software, brands, services, consumer franchises — step over to EV/EBITDA, free cash flow multiples, or a full DCF. Every valuation framework has a domain of validity; P/B’s is banks and other genuinely asset-heavy businesses.
Sources
- Aswath Damodaran, NYU Stern, “Price and Value to Book Ratio by Sector (US)” — 5,994 firms, data as of January 2026: pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/pbvdata.html
- Multpl, S&P 500 Price-to-Book Ratio — historical series, mean 3.16, median 2.91, March 2009 low 1.78, July 2026 reading 5.80: multpl.com/s-p-500-price-to-book
- SEC Office of Investor Education and Advocacy, “Beginners’ Guide to Financial Statements” — balance sheet, shareholders’ equity, and the accounting equation: sec.gov — Beginner’s Guide to Financial Statements
- CFA Institute, Equity Valuation curriculum readings — residual income models and the P/B-to-ROE link: cfainstitute.org — refresher readings
Disclosure: This article was produced with AI assistance and reviewed before publication. It is for informational purposes only and is not investment advice.