Valuation Multiples — P/E, EV/EBITDA, Companion Variables, and the Rate Sensitivity

The P/E ratio is the most widely used valuation multiple. But it is also the most misused — because the number means nothing without context. Damodaran derives P/E from the Gordon Growth Model to make the hidden assumptions explicit:

EV/EBITDA avoids the leverage distortion that makes P/E incomparable across companies with different capital structures. Enterprise value includes all capital (debt + equity − cash); EBITDA is pre-interest. Both are 'capital structure neutral,' making EV/EBITDA superior for comparing businesses with very different D/E ratios:

• Enterprise Value (EV) = Market capitalization + Total debt − Cash and equivalents + Preferred equity + Minority interest. EV represents the total cost to buy the entire business — the price a private equity buyer or strategic acquirer would pay.
• What EV/EBITDA embeds: like P/E, EV/EBITDA encodes growth, risk, and a third factor unique to it — capital intensity. Two businesses with the same EBITDA but very different CapEx requirements deserve different EV/EBITDA multiples: the high-CapEx business converts less EBITDA to FCFF (after CapEx deduction), so each EBITDA dollar is worth less.
• EV/EBITDA fundamental drivers (from McKinsey): EV/EBITDA is approximately = (1 − cash tax rate) × (1 − g/ROIC) ÷ (WACC − g). Companies with high ROIC, high WACC-to-growth spreads deserve high multiples. This formalizes the intuition: a business that earns 25% ROIC with 10% growth and 9% WACC should trade at higher EV/EBITDA than one earning 9% ROIC with 8% growth and 9% WACC (the second is barely creating value).
• Industry norms for EV/EBITDA: Software/SaaS: 20–50× (high growth, high ROIC, low capital intensity). Consumer staples: 12–18× (moderate growth, high ROIC, low volatility). Industrial manufacturing: 8–12× (moderate growth, moderate ROIC). Capital-intensive industries (utilities, telecom): 6–10× (low growth, high CapEx offset). Always compare within industry — cross-industry EV/EBITDA comparisons require explicit normalization.

The PEG ratio (P/E ÷ Growth rate) adjusts the P/E multiple for growth, creating a rough 'apples-to-apples' comparison across companies growing at different rates:

• PEG strengths: simple, intuitive, reduces the 'high growth = always expensive' bias. A company at P/E 40× growing 40% has PEG 1.0; a company at P/E 15× growing 5% has PEG 3.0. The PEG framework correctly identifies the first as potentially fair-valued and the second as potentially expensive for its growth.
• PEG limitations — Damodaran's critique: (1) Risk is ignored: a high-risk startup at PEG 0.8 is not more attractive than a low-risk consumer staple at PEG 1.5; the higher risk deserves a lower price per unit of growth. (2) Growth quality is ignored: growth at ROIC > cost of equity is valuable; growth at ROIC < cost of equity destroys value. A company growing 40% at 7% ROIC (below WACC) deserves a low P/E — a PEG of 0.5 is not a bargain. (3) EPS growth vs. FCF growth: a company that grows EPS 40% through accounting manipulation but only grows FCF 10% has a PEG that overstates attractiveness. Always confirm that EPS growth is backed by FCF growth.
• Modified PEG: Damodaran suggests dividing P/E by a growth variable that also captures ROIC or reinvestment efficiency. More sophisticated screen: P/FCFE per share ÷ FCF per share growth rate — this directly tests whether you are paying a fair price for growing economic value.

One of the most underappreciated aspects of valuation multiples is their sensitivity to interest rates. Equities, like bonds, have duration — and higher discount rates compress the present value of future earnings:

• The mechanism: P/E = Payout × (1+g) ÷ (Ke − g). Cost of equity (Ke) = Risk-free rate + ERP × Beta. When risk-free rates rise, Ke rises, the denominator (Ke − g) rises, and P/E falls — all else equal. This is the 'rates rising = multiples compressing' relationship observed in 2022 when the Fed raised rates from 0% to 5% and technology multiples (which had priced in decades of future earnings) compressed 60–80%.
• Long-duration equities are more rate-sensitive: a company whose value is concentrated in near-term cash flows (short duration) is less sensitive to rate changes than one whose value depends on cash flows 10–20 years in the future (long duration). High-growth companies with minimal current earnings but enormous future earnings potential are the longest-duration equities — their P/E and EV/EBITDA multiples compress the most in rising rate environments. Profitable, mature companies with high current FCF yields are shorter-duration and more rate-resistant.
• Quantifying multiple compression from rates: using the one-stage DDM, a rise in Ke from 9% to 11% for a company growing at 6% reduces justified P/E from (payout × 1.06)/(0.09−0.06) = 35.3× (assuming 100% payout for simplicity) to (payout × 1.06)/(0.11−0.06) = 21.2×. A 2pp rate rise compresses justified P/E by 40% for a 6%-growth company — entirely from the discount rate channel, with no change in business fundamentals.
• Damodaran's equity risk premium dynamics: ERP is not static. In crises (2008, 2020), ERP spikes as investors demand more risk compensation — compressing all equity multiples even as risk-free rates fall. Post-crisis normalization sees ERP compress back — expanding multiples even if risk-free rates don't change. Understanding both components of Ke (risk-free rate + ERP) is necessary to forecast multiple trajectories.
• Median vs. mean in comparable analysis: Damodaran recommends using median rather than mean P/E and EV/EBITDA in comparable company analysis. Why? A few outliers with extreme multiples (unprofitable companies with near-zero EPS produce enormous P/Es; acquisitions with premium prices inflate multiples) will distort the mean severely. The median is robust to these outliers and better represents the 'typical' company in the peer group.