What P/B measures and why it matters
P/B can be derived from fundamental valuation: P/B = (Return on Equity − growth) / (Cost of equity − growth). This formula reveals that P/B is fundamentally driven by the gap between a company’s return on equity (ROE) and its cost of equity. A company earning its cost of equity exactly should trade at P/B = 1. A company earning ROE above its cost of equity (a high-ROIC franchise) should trade at P/B > 1. A company destroying value (ROE below cost of equity) should trade below book value. This is why banks trading at 0.5× book are not necessarily cheap — if their cost of equity is 10% but they are earning 5% ROE (a common situation for European banks post-2008), the “correct” P/B may be well below 1.
When P/B is the right multiple
P/B is most informative for financial companies — banks, insurance companies, and investment trusts — where the balance sheet is the business. The primary assets (loan books, investment portfolios) are marked to market or valued close to economic value, making book value a meaningful reflection of liquidation value. Bank analysts spend significant time on tangible book value (removing intangibles and goodwill) and tangible book value per share growth as the primary equity value measure. P/B is also appropriate for holding companies, property companies (where underlying assets are independently valued), and mining companies (where the value of proved reserves can be benchmarked against market cap).
When P/B is misleading
For asset-light and intangible-heavy businesses, book value is essentially meaningless. A software company spends its capital on engineers and R&D — under GAAP, most of this is expensed immediately, leaving minimal asset base on the balance sheet despite substantial economic value. Amazon’s book value grossly understates its economic worth because its most valuable assets (AWS infrastructure advantage, Prime ecosystem, logistics network) are either expensed or not separately capitalised. A P/B of 10× for such a company is not expensive if ROE is 50%+ and cost of equity is 10%. Applying a P/B framework that implies “below 1× is cheap, above 5× is expensive” uniformly across sectors is one of the most common valuation mistakes.
“A low P/B ratio is not value. It is an invitation to investigate whether book value is real, whether the business can earn its cost of equity, and whether the discount is permanent or cyclical.”
What this means for you
The Fama-French HML factor (high book-to-market value stocks outperforming low book-to-market growth stocks) is essentially the academic formalisation of P/B-based value investing. The empirical evidence suggests low P/B stocks have historically outperformed — but this premium has been weaker and more contested since 2007, as the economy has shifted toward intangible-heavy businesses where P/B is a poor signal. For investors today, the most useful application of P/B is in financial sector analysis: UK and European bank stocks trading at 0.6–0.9× tangible book may genuinely be cheap if ROE is recovering toward cost of equity, or may represent structural value destruction if the business model is under permanent threat from fintech and margin compression.