Price-to-Book Ratio (P/B)

Book value is total assets minus total liabilities -- the accounting residual that belongs to shareholders. P/B divides the current share price by book value per share to tell you how many dollars the market is paying for each dollar of accounting net worth. At face value this sounds like a clean comparison. In practice, it is only useful when book value is a reasonable approximation of economic value -- a condition that holds in far fewer industries than most investors assume.

The core problem is that accounting standards require most assets to be carried at historical cost, not current market value. A manufacturer that bought land in 1985 carries it at 1985 prices. A software company that spent $500M building its platform expenses that spending as incurred -- it appears nowhere on the balance sheet as an asset, leaving book value artificially depressed. Meanwhile, an acquirer who paid $2B for a competitor will carry $1.5B of goodwill on its balance sheet -- an intangible that says nothing about the economic value of the business going forward. These distortions mean that P/B comparisons across different industries are nearly meaningless without adjustment.

Acquisitions compound the problem. Serial acquirers accumulate large goodwill balances that inflate total assets and therefore appear to reduce P/B. But goodwill is only as valuable as the business generating it -- if the acquired business underperforms, goodwill impairment charges write it down, sometimes catastrophically. Book value for acquisition-heavy companies is therefore particularly unreliable as a measure of tangible worth. Tangible book value (which excludes goodwill and other intangibles) is a more defensible starting point for these cases.

Banks and insurance companies are the natural home of P/B analysis because their balance sheets actually do approximate economic reality. Banks mark most assets to fair value under accounting rules -- loans are carried at amortized cost with reserves for expected losses, securities portfolios are largely marked to market, and the liability side (deposits, borrowings) is well-understood. When a bank trades at 0.7x book, you are paying 70 cents for a dollar of regulatory capital -- that is a concrete and meaningful statement about valuation.

The critical complement to P/B for financial institutions is return on equity (ROE). The justified P/B for any business is a function of its ROE relative to its cost of equity: a bank that earns 15% ROE on book equity deserves to trade above 1x book; a bank earning 6% ROE on equity when its cost of equity is 9% deserves to trade below 1x book. This ROE-to-P/B framework, sometimes called the residual income model, is the theoretical foundation for bank valuation. A low P/B is not automatically a bargain -- it may simply reflect the market's rational assessment that the bank will continue to earn below its cost of equity.

The academic value factor -- long cheap stocks, short expensive ones -- was historically defined using P/B in the foundational Fama-French three-factor model. But the value premium, measured this way, has been weak to negative since roughly 2007. This is not because value investing stopped working; it is because P/B became a less reliable proxy for value as the economy shifted toward intangible-intensive businesses where book value systematically understates economic worth.

A software company trading at 10x book is not necessarily expensive -- if its competitive moat, recurring revenue base, and unit economics justify a high return on the (very small) capital it deploys, the 10x P/B may be a fair or even cheap price. The lesson is not to abandon price-to-book but to use it contextually: it remains a useful primary signal in financial services, REITs, and asset-heavy industrials, and a nearly useless one in technology, biotech, and any sector where value creation is driven by human capital and intellectual property rather than physical assets.