How to Choose a Discount Rate That Actually Works

Picture two analysts covering the same quality business — identical projected cash flows, identical terminal assumptions, identical growth rates for the next decade. Analyst A runs a WACC of 8.4%, derived from a careful assessment of beta and the equity risk premium. Analyst B uses 10.2%, drawing on a slightly different beta estimate and a more conservative view of the market premium. Analyst A's fair value: $215 per share. Analyst B's: $167 per share. The stock trades at $185. Analyst A calls it undervalued. Analyst B says it's expensive. Neither has made a calculation error.

This isn't an edge case. It's a structural feature of discounted cash flow analysis — the discount rate sits in the denominator of every present value calculation, and small shifts in the denominator produce large shifts in the result. A 2-percentage-point difference in the assumed discount rate can move a fair value estimate by 25–35%, enough to flip a buy thesis to a hold, or a hold to a sell, on the same business with the same cash flows. Most investors understand this abstractly. The practical implication is less often absorbed: a meaningful portion of any DCF output isn't analysis. It's an assertion about a number that can't actually be known.

The question worth asking isn't which discount rate is correct. It's this: given how much uncertainty surrounds the discount rate, what range of rates makes sense for quality businesses — and how should a valuation framework handle that uncertainty rather than pretending it can be solved with more decimal places?

Why the Discount Rate Has Such Leverage

Start with a simplified form of the math. The Gordon Growth Model — a perpetuity formula — makes the leverage intuitive before bringing in the full machinery of a 10-year DCF analysis.

Fair Value = Owner Earnings Per Share / (Discount Rate − Long-Term Growth Rate)

Consider Moody's Corporation (MCO). From Moody's FY2024 annual report, filed with the SEC in February 2025: net income of approximately $2.08 billion, depreciation and amortization of $248 million, capital expenditures of $94 million. Owner earnings: $2.08B + $0.25B − $0.09B = roughly $2.24B. With approximately 193 million diluted shares outstanding at fiscal year-end, that's about $11.60 per share. Assume a 4% long-term growth rate — in line with nominal GDP — and watch what a single percentage point does to the output.

At 8%: $11.60 / (0.08 − 0.04) = $11.60 / 0.04 = $290 per share At 9%: $11.60 / (0.09 − 0.04) = $11.60 / 0.05 = $232 per share At 10%: $11.60 / (0.10 − 0.04) = $11.60 / 0.06 = $193 per share

From 9% to 8%, fair value rises $58 per share — a 25% jump. From 8% to 10%, it falls $97 per share — a 33% drop. One assumption. No change in the underlying business. A spread in fair value that would put one investor confidently buying and another confidently avoiding the same stock on the same day.

Moody's traded near $475 in late 2024, well above these perpetuity estimates — the market prices in growth meaningfully above a 4% steady-state assumption, which is appropriate for a business with Moody's dominant position in credit ratings. The perpetuity formula is a blunt instrument; MoatScope's three-scenario approach applies a multiplier to owner earnings rather than a perpetuity model and produces different numbers. But the mathematical relationship is the same regardless of which model you use: the discount rate input carries enormous leverage over the output.

What WACC Gets Right — And Where It Breaks Down

The academically standard answer to the discount rate question is the weighted average cost of capital. WACC weights the cost of equity and the after-tax cost of debt by their proportions in the capital structure, producing the blended rate at which a company's capital providers collectively need to be compensated.

WACC = (E/V × Re) + (D/V × Rd × (1 − T)) Where E = market value of equity, D = market value of debt, V = E + D, Re = cost of equity, Rd = pre-tax cost of debt, T = corporate tax rate

The cost of debt is observable: it's the yield on the company's bonds or the rate on its credit facility, adjusted for the tax deductibility of interest. The cost of equity is not. It requires an estimate, typically via the Capital Asset Pricing Model:

Re = Rf + β × ERP Where Rf = risk-free rate (10-year Treasury yield), β = beta, ERP = equity risk premium

The risk-free rate is observable. Beta and the equity risk premium are not — and here WACC's theoretical elegance collides with practical reality. Beta is estimated from historical price data: it measures how much a stock has moved relative to the market in the past, on the assumption that past volatility is a reasonable proxy for future risk. But beta is unstable over time. Moody's five-year beta has ranged from roughly 0.6 to 1.1 depending on which historical window you measure, whether you apply a Blume adjustment toward 1.0, and which index you use as the market proxy. That range alone produces a spread of more than 2 percentage points in the cost of equity before touching anything else.

The equity risk premium compounds the problem. Academic estimates run from 4% to 7% depending on methodology — historical realized premium, survey-based forward premium, or an implied premium derived from current market valuations. A 3-point spread in the ERP, layered on top of beta uncertainty, means two analysts can produce WACC estimates for the same company that legitimately differ by 3–4 percentage points. The formula will carry the output to one decimal place. The inputs don't support that precision. Illusory precision, ultimately.

The honest confession — and this is genuinely an area where my practice has drifted from textbook toward pragmatic over time, and I'm less certain about the textbook answer than I once was — is that for most individual investors, calculating company-specific WACCs adds a veneer of rigor without commensurately improving valuation decisions. The effort spent arguing over beta and ERP is often better invested in the judgment calls that actually move the needle: whether the competitive position is durable, whether earnings quality is real, whether management's capital allocation has been consistent over full market cycles.

The Case for a Simple Range: 8–10%

There's a reason Buffett and Munger used the long-term US Treasury rate — or, in some periods, a flat 10% as a conservative floor — rather than calculating company-specific WACCs. They understood that the precision is illusory. A range, honestly stated, is more truthful than a point estimate precisely calculated.

A workable starting framework for quality-focused investors:

• Wide-moat, high-quality businesses with predictable, recurring cash flows — credit rating agencies, payments networks, dominant enterprise software platforms: 8–9%
• Average-quality businesses with moderate competitive advantages and some earnings cyclicality: 9–10%
• Lower-quality, more cyclical, or capital-intensive businesses where earnings swing materially through cycles: 10–12% or higher

But there's a discipline many investors miss: if you embed quality into the discount rate — using a lower rate for better businesses — don't also embed it in the terminal multiple or long-term growth assumption. Both levers are doing the same work. Double-counting produces systematically optimistic fair values for quality stocks, which is perhaps the most pleasant-feeling way to make a valuation error. Use one mechanism, consistently.

The 8–10% range has rough empirical grounding. The US equity market has returned approximately 9–10% nominally over long historical periods. A discount rate below that implies you're assigning a specific business a higher present value than a diversified market portfolio would warrant — defensible for an exceptional, wide-moat business, but requiring genuine conviction about moat durability. A rate above 10% implies you're demanding a premium over the market's long-run return, appropriate for cyclical or capital-intensive businesses where the earnings path is inherently less predictable.

And resist the temptation to adjust your discount rate with every move in short-term interest rates. Your discount rate reflects the long-run opportunity cost of capital — not the Fed funds rate this quarter. Tying the discount rate too closely to current yields causes fair value estimates to swing with monetary policy cycles rather than reflecting the underlying economics of the businesses you're evaluating. The intrinsic value of a durable business doesn't change 20% because the 10-year Treasury moved 100 basis points.

How MoatScope's Scenarios Handle This

MoatScope's fair value methodology addresses discount rate uncertainty in a more honest way than a single-rate DCF: rather than optimizing for one correct rate, it presents three scenarios — Conservative (14× owner earnings), Base (27×), and Optimistic (40×) — that implicitly bracket a range of reasonable discount rate and growth combinations. The spread between Conservative and Optimistic is the framework's built-in acknowledgment that you cannot know the correct rate. The same business can be worth $140 under cautious assumptions or $400 under generous ones — and acknowledging that range is more useful than false confidence in a single figure.

This connects to how the reverse DCF fits into the picture. Instead of debating which discount rate to assume, you can ask what growth rate and discount rate combination is already embedded in the current stock price. The margin of safety you require is then calibrated to how far the current price sits from the Conservative scenario — a judgment about downside protection rather than a debate about unknowable point estimates.

Key Takeaways

• The discount rate is the single most consequential input in any present value calculation — and also one of the least knowable. A 1-percentage-point shift moves a perpetuity fair value by 17–25%; a 2-point shift can exceed 33%.
• WACC is theoretically correct but practically unstable. Beta ranges and equity risk premium uncertainty allow two careful analysts to land 3–4 percentage points apart on the same company's discount rate, both with defensible inputs.
• An 8–10% range, calibrated by business quality — lower for wide-moat, highly predictable businesses; higher for cyclical or capital-intensive ones — captures the honest range of defensible rates for most quality investments. Stop trying to optimize within that range; the precision you're recovering doesn't exist in the inputs.
• Never double-count quality: if you lower the discount rate for a better business, don't also raise the terminal multiple or growth rate. Both adjustments do the same work, and stacking them produces systematically optimistic fair values.
• MoatScope's three-scenario approach reframes the question: rather than asking which discount rate is correct, focus on how much downside protection the Conservative scenario provides at the current price.