Skip to main content

Valuation Multiples (P/E, EV/EBITDA, P/B, P/S)

Updated Jun 24, 2026 at 2:35pm

Research Draft High 1,303 words

A valuation multiple is a ratio of a company's price (or enterprise value) to some measure of its size or output — earnings, cash flow, book equity, or revenue. The premise is relative valuation: rather than discounting a company's future cash flows from scratch (intrinsic/DCF valuation), you price an asset by what the market pays for "comparable" assets per unit of a chosen fundamental. The core tension is that a multiple compresses an entire firm's growth, risk, profitability, and capital structure into a single number — making it fast, intuitive, and easy to communicate, but also easy to misuse when the companies being compared differ on exactly those buried variables. As Aswath Damodaran frames it, a multiple is "a starting point, not an end point."

How they're calculated

The four most common multiples split into equity multiples (numerator = price or market cap, claim = equity holders only) and enterprise multiples (numerator = enterprise value, claim = all capital providers). Damodaran's first rule is definitional consistency: equity value goes over an equity flow; firm/enterprise value goes over a pre-financing flow.

  • P/E (Price/Earnings) = price per share / earnings per share (or market cap / net income). An equity multiple. Trailing P/E uses the last 12 months of reported EPS; forward P/E uses analyst-estimated forward EPS. Undefined when earnings are negative.
  • EV/EBITDA = Enterprise Value / EBITDA, where EV = market cap + total debt + preferred + minority interest − cash & equivalents, and EBITDA = earnings before interest, taxes, depreciation, and amortization. An enterprise multiple.
  • P/B (Price/Book) = market cap / shareholders' equity (book value). An equity multiple, also called price-to-book or the inverse of book-to-market.
  • P/S (Price/Sales) = market cap / revenue (equity-side); a cleaner variant is EV/Sales, which is consistent because revenue is a pre-financing figure.

How they're used in practice

The standard workflow follows Damodaran's four steps: (1) define the multiple consistently; (2) examine its cross-sectional distribution — multiples are skewed and bounded at zero, so the median is more reliable than the mean; (3) understand the fundamentals that drive it (P/E rises with expected growth and falls with risk; EV/EBITDA falls with reinvestment needs); and (4) find genuinely comparable firms and control for differences.

Each multiple has a natural domain:

  • P/E is the default for mature, profitable companies and is the most widely quoted by far. Best within a single sector with similar growth and leverage.
  • EV/EBITDA is the professional's cross-company tool because it is capital-structure neutral — debt is added to the numerator and interest is excluded from the denominator, so two firms with very different leverage become comparable. It also neutralizes differing depreciation policies, which matters for capital-intensive industries and cross-border comparisons. Standard in M&A and leveraged-buyout analysis.
  • P/B is most meaningful where assets are marked near fair value and book equity is economically real — banks, insurers, and asset-heavy financials. It is the metric behind the academic value factor.
  • P/S (or EV/Sales) is the fallback when earnings are negative or erratic — early-stage growth firms, cyclical troughs, turnarounds. Revenue is harder to manipulate and more stable through cycles than earnings, which is why Kenneth Fisher built a strategy around it in Super Stocks (1984).

A common refinement is to relate a multiple to its driver: the PEG ratio (P/E ÷ earnings growth rate) adjusts P/E for growth so that fast-growers and slow-growers can be compared.

Adoption, debate & evidence

Multiples are the most widely used valuation method in practice — they dominate equity research, M&A, and IPO pricing because they are quick and reflect current market sentiment. But "widely used" is not "reliably predictive," and the evidence differs sharply by multiple and by horizon.

  • Aggregate-market P/E (CAPE / Shiller P/E): The cyclically-adjusted P/E (price ÷ 10-year inflation-adjusted average earnings) has genuine long-horizon predictive power. Campbell and Shiller's 1988 work ("Stock Prices, Earnings, and Expected Dividends," Journal of Finance; and the dividend-price-ratio paper in Review of Financial Studies) showed long-run smoothed earnings/valuation ratios carry information about future returns; later commentary (e.g. Advisor Perspectives) reports CAPE explaining a large share of 10-year-ahead return variation, though the specific R² figures vary by sample and are not settled. Crucially, the same research is emphatic that CAPE is near-useless for market timing — there is little relationship between valuation and short-term returns, and it can stay "expensive" for many years (Kitces). Critics also note accounting-rule changes and one-off earnings shocks (e.g. the 2008–09 trough) distort cross-era CAPE comparisons.
  • P/B and the value factor: Fama and French (1992–93) found book-to-market (the inverse of P/B) explains cross-sectional returns beyond market beta, and high book-to-market ("value") stocks historically outperformed. However, this premium has weakened markedly since roughly 2007, with value lagging growth for a prolonged stretch. A leading critique is that book value increasingly omits intangibles (R&D, brand, software), making P/B a degraded proxy for economic capital in asset-light economies.
  • Single-stock multiples generally: A low multiple is not an edge on its own — it is frequently a "value trap," correctly signalling impaired growth or distress. Fisher himself warned that "terrible companies can have low PSRs."

The honest summary: multiples are excellent descriptive and communication tools and, in aggregate cyclically-adjusted form, carry real long-run information — but as standalone single-stock buy signals their measured edge is weak and regime-dependent.

Strengths & limitations

Strengths: fast, transparent, grounded in observable market prices, and require fewer explicit assumptions than a DCF. EV/EBITDA in particular enables clean comparison across leverage and depreciation regimes.

Limitations: every multiple silently assumes the comparables share the firm's growth, risk, and profitability. P/E is distorted by leverage, one-time items, and accounting choices, and breaks on negative earnings. EV/EBITDA ignores capital expenditure, so it flatters capital-hungry businesses (Charlie Munger's critique of "EBITDA" as "bullshit earnings"). P/B misleads for asset-light firms and for those with buybacks or write-downs that have hollowed out book equity. P/S ignores profitability entirely — high revenue with no path to profit looks deceptively cheap.

The #1 misuse is comparing multiples across companies (or eras) that differ on the very fundamentals the multiple suppresses — then declaring the lower number "cheap" without checking whether the discount is deserved.

Sources

Disputes flagged: (1) the value premium's persistence is genuinely contested post-2007; (2) CAPE's cross-era comparability is debated due to accounting changes; CAPE precision figures (e.g. "67% within ±1.37%") are from a single Advisor Perspectives analysis and should be treated as illustrative, not settled.