Every valuation model ever built falls into one of two categories: intrinsic valuation (what is this business worth on its own terms, independent of market prices?) or relative valuation (what are similar businesses trading for?). Damodaran spends the first chapter of The Little Book of Valuation establishing this taxonomy because the choice of approach determines everything about how you build your model, what inputs matter, and what errors you are most likely to make.
Valuation Approaches — Complete Taxonomy
Three fundamental approaches; every valuation method belongs to one branch
Business Valuation
What is this business worth?
Intrinsic Value
Cash flow–based
DCF (Discounted Cash Flow)
DDM (Dividend Discount Model)
FCFF / FCFE Models
Residual Income
Use When
Long-term fundamental analysis
✓ Grounded in fundamentals
✗ Sensitive to assumptions
Relative Value
Comparables-based
P/E, P/Book, P/Sales
EV/EBITDA, EV/FCF
Precedent transactions
Sum-of-the-parts
Use When
Quick screening, M&A, IPO pricing
✓ Market-anchored, fast
✗ Assumes comparables are fairly priced
Contingent Claims
Option-based
Black-Scholes (equity as call)
Real options analysis
Patent / resource valuation
Distressed debt options
Use When
Optionality embedded in assets
✓ Captures upside asymmetry
✗ Complex; hard to estimate inputs
Damodaran's Synthesis Rule
Use intrinsic valuation as the primary tool to establish independent value, then use relative valuation to sanity-check against what the market implies. When the two disagree, investigate why — the answer usually reveals either a mispricing or a flaw in your DCF assumptions.
Figure 2.1 — The three-branch taxonomy of business valuation. Most practitioners use all three approaches in combination; no single approach is complete.
Damodaran's organizing framework: there are exactly three approaches to valuing any asset. Intrinsic valuation anchors on the cash flows the asset will generate over its life, discounted at a rate that reflects their riskiness. Relative valuation anchors on what comparable assets trade for in the market today. Asset-based valuation anchors on what the assets could be sold for individually. Most practitioners use all three, triangulate between them, and investigate when they disagree rather than mechanically averaging the outputs.
| Approach | Core Logic | Primary Inputs | Best For | Main Limitation |
|---|---|---|---|---|
| Intrinsic (DCF) | Value = PV of all future cash flows at appropriate discount rate | Revenue forecasts, margins, reinvestment, WACC, terminal growth rate | Patient investors with long time horizons; companies with predictable cash flows; any asset where you want market-independent value | Input-sensitive: small changes in growth rate or WACC produce large value swings; requires explicit assumptions about every variable |
| Relative (Comps) | Value = what similar assets trade for × a relevant scaling variable (earnings, EBITDA, revenue) | Current market prices of comparable companies; choice of multiple; selection of peer group | Quick valuation sanity check; IPO pricing; M&A context; when markets are believed to be broadly efficient | Inherits market mispricing; 'fair value' per comps only if the entire peer group is fairly priced; peer selection is highly subjective |
| Asset-Based (NAV) | Value = market or replacement value of individual assets minus liabilities | Individual asset values (real estate, investment portfolio, patents, brands) | Holding companies, real estate investment trusts, asset-heavy businesses, liquidation scenarios | Misses going-concern value (synergies, franchise value, management quality); understates value for businesses whose assets generate returns above cost |
Intrinsic valuation is built on a single principle: an asset is worth the present value of all the cash flows it will generate over its life, discounted at a rate that reflects those cash flows' riskiness. This principle is mathematically rigorous and economically sound. It treats every business as a bond with uncertain cash flows — the challenge is estimating those cash flows honestly.
The DCF framework forces you to make your assumptions explicit and testable. When you say a stock is worth $80 per a DCF, you are simultaneously saying: revenue will grow at X%, margins will stabilize at Y%, the discount rate is Z%, and the terminal growth rate is W%. If any of those assumptions is wrong, the $80 is wrong. This transparency is a feature, not a bug — it allows you to identify which assumptions drive the value and stress-test them. Comps hide these assumptions inside the peer group's pricing.
Relative valuation does not ask what a business is intrinsically worth — it asks what the market is currently paying for comparable businesses. If Apple trades at 28× earnings and similar technology companies trade at 25×, one can argue Apple is slightly expensive relative to peers. Relative valuation is faster, simpler, and far more commonly used in practice than DCF — which is precisely why its limitations matter so much.
| Multiple | Formula | Best Use Case | Critical Error to Avoid |
|---|---|---|---|
| P/E (Price-to-Earnings) | Market Price per Share ÷ EPS (trailing or forward) | Mature, profitable companies; broad market comparisons | Comparing P/E across different growth rates without adjusting (PEG ratio fixes this); GAAP earnings distorted by non-cash items |
| EV/EBITDA | Enterprise Value ÷ EBITDA | Cross-capital-structure comparisons; LBO context; capital-intensive industries | EBITDA overstates cash flow by ignoring capex and working capital; companies with very different capex intensities are not comparable |
| EV/Sales | Enterprise Value ÷ Revenue | High-growth or unprofitable companies; early-stage businesses | Meaningless without margin context — a 10× EV/Sales on a 40% margin business is cheap; the same multiple on a 5% margin business is expensive |
| P/Book (Price-to-Book) | Market Cap ÷ Book Equity | Banks and financial institutions; asset-heavy businesses; distressed situations | Book value is accounting-driven and often diverges dramatically from economic value; particularly useless for asset-light businesses |
| EV/FCF | Enterprise Value ÷ Free Cash Flow to Firm | Capital-efficient businesses; steady-state cash generators; Buffett's preferred lens | FCF is lumpy year-to-year; one-time capex can distort the denominator significantly |
Relative valuation can only tell you whether an asset is cheap or expensive relative to comparable assets trading in the current market. It cannot tell you whether the current market price of those comparable assets is itself rational. During the dot-com bubble, every internet stock was 'fairly valued' by relative metrics because the entire peer group was trading at 100× revenue. During the 2009 financial crisis, every bank stock appeared expensive on P/E because earnings had collapsed — but in fact they were deeply undervalued relative to normalized earnings power. Comps inherit the market's errors.
Damodaran's practical guidance: use both approaches, triangulate between them, and investigate the gap when they disagree. A DCF saying $80 and comps saying $120 is not a reason to average to $100 — it is a prompt to understand what assumption in the DCF differs from the market's implied belief, and to decide which view you think is more correct.
In 2001, Amazon had no earnings and massive cash burn. On a P/E or EV/EBITDA basis: impossible to value (negative denominators). On an EV/Sales basis: even at depressed prices, Amazon traded at 2–3× revenue while bookstores traded at 0.3× revenue — comps suggested it was expensive. But a DCF using Amazon's eventual business model (scaling e-commerce + AWS at high margins) implied massive undervaluation even in 2001. The lesson: relative valuation was useless here because there were no truly comparable businesses. Only intrinsic valuation could capture the option value of the emerging business model.
Key Takeaways
In 2007, all major US banks traded at approximately 1.5–2.0× book value. An analyst using relative valuation concludes that Citigroup at 1.7× book is 'fairly valued' relative to peers. What is the critical flaw in this analysis?