Discount Rate Asymmetry

In the AI Story

The mathematical consequence is large. Because the discount rate applies to every year of projected cash flows and to terminal value, even small asymmetries compound into substantial valuation differences. Two companies with identical projected cash flows of $500 million per year growing at 8% with a 3% terminal growth rate produce intrinsic values that differ by approximately 50-60% when discounted at 12% versus 9%. The discount rate is not a technical detail; it is the valuation.

The asymmetry is justified by the structure of the competitive shift. AI has not uniformly increased risk across technology companies. It has increased the risk of code-dependent companies (whose moats have been breached) and decreased the risk of ecosystem-dependent companies (whose moats have been relatively strengthened, because the proliferation of AI-built software increases demand for platforms that govern complexity). Using a uniform discount rate across the sector — whether based on historical betas, sector averages, or general technology risk — systematically misprices both categories.

The practical implementation requires judgment beyond formula. The CAPM produces a starting point; the analyst's assessment of the company's specific moat structure determines the adjustment. A single-product vertical SaaS company with no ecosystem warrants a larger disruption premium than a broad-platform company with moderate code dependency. A platform company with deep data moats and extensive marketplace network effects warrants a larger durability discount than a company whose ecosystem is shallower.

The connection to Damodaran's broader framework is direct: the discount rate is, in the final analysis, a measure of the analyst's confidence in the narrative. A low discount rate says the analyst is confident the story will play out as projected. A high discount rate says the analyst sees significant risks of deviation. The AI disruption has clarified which competitive advantages support confidence and which do not, and the discount rates must reflect this clarification.

Origin

The framework emerged in Damodaran's 2024-2026 work on AI's impact on equity valuation, building on his decades of writing about discount rate calibration and his recurring criticism of mechanical CAPM application without judgment overlay.

Key Ideas

Beta is backward-looking. Historical beta cannot capture forward exposure to a structural disruption that did not exist in the historical data.

Asymmetric impact requires asymmetric rates. Code-dependent companies warrant a disruption premium; ecosystem-dependent companies warrant a durability discount.

The math compounds. Small asymmetries in the discount rate produce large differences in intrinsic value because the rate applies to every year and to terminal value.

Judgment overlays formula. CAPM is a starting point; the analyst's moat assessment determines the adjustment magnitude.