Weighted Average Cost of Capital (WACC)

The formula in one line

The weighted average cost of capital is the single discount rate that compensates every supplier of capital — debtholders and shareholders — at the rate each demands, weighted by how much capital each provides:

WACC = (E / V) · r_e  +  (D / V) · r_d · (1 − T_c)

where  V = E + D
       E  = market value of equity
       D  = market value of debt
       r_e = cost of equity
       r_d = cost of debt (pre-tax)
       T_c = marginal corporate tax rate

Three pieces, no more. The market-value weights (E/V and D/V) tell you the firm's capital mix. The cost of equity r_e is the return shareholders demand for bearing the firm's systematic risk. The cost of debt is multiplied by (1 − T_c) to reflect that interest payments are tax-deductible — the famous tax shield. The government subsidises debt; equity gets no comparable break.

Almost every published corporate valuation in the world uses WACC. It is the discount rate inside every public-company DCF model, the hurdle rate every capital-budgeting committee references, and the input that, if you nudge it by a percentage point, can revalue a company by a quarter.

Worked example: a mid-cap industrial

Take a company with the following snapshot:

Market cap (E)       = $8.0 B
Net debt (D)         = $2.0 B  (market value of bonds)
V = E + D            = $10.0 B
E / V                = 0.80
D / V                = 0.20

Risk-free rate r_f   = 4.0%   (10-yr Treasury)
Equity beta β        = 1.15
Equity risk premium  = 5.0%
Cost of equity r_e   = 4.0% + 1.15 × 5.0%  =  9.75%

Yield on bonds r_d   = 5.5%
Corporate tax T_c    = 21%
After-tax debt cost  = 5.5% × (1 − 0.21)   =  4.35%

WACC = 0.80 × 9.75%  +  0.20 × 4.35%
     = 7.80%  +  0.87%
     = 8.67%

So free cash flows for this firm should be discounted at 8.67% in a DCF. Any project that promises a higher unlevered return creates value; anything that returns less destroys it.

Getting the cost of equity right

The default approach is CAPM. The cost of equity is the risk-free rate plus the firm's beta times the equity risk premium:

r_e  =  r_f  +  β · (r_m − r_f)
     =  r_f  +  β · ERP

Each input has its own minefield.

• Risk-free rate. For long-duration cash flows, use a long-duration government bond — 10-year Treasury is the U.S. convention. Avoid T-bills; their reinvestment risk biases the discount rate too low for valuations with terminal value.
• Beta. Bloomberg's default is two-year weekly raw beta. Damodaran prefers five-year monthly. For thinly-traded names or recently-public firms, use the median beta of comparable listed peers, unlevered (Hamada formula) and re-levered to your firm's structure. Single-firm raw regressions are too noisy: Tesla's beta has ranged 1.4–2.4 across recent windows.
• Equity risk premium. Historical realised: 6–7% over Treasuries for U.S. equities back to 1928. Forward-looking implied (Damodaran's monthly figure): 4–6%. Most U.S. corporate-finance teams settle near 5%. International ERPs add country-risk premia; emerging markets often run 7–10%.

Alternatives exist. The dividend-discount inversion gives r_e = D₁ / P₀ + g, useful for stable dividend payers (regulated utilities). Fama-French three-factor adds size and book-to-market premia. The build-up method, common in private-company appraisal, sums risk-free + ERP + size premium + specific-company risk + industry risk — clumsy but auditable.

Getting the cost of debt right

The pre-tax cost of debt is the yield-to-maturity on the firm's outstanding bonds — not the coupon rate. Coupon reflects what bondholders agreed to receive when the bond was issued; yield reflects what they would demand today. For a firm with publicly traded bonds, pull the yield-to-worst on the longest-dated senior unsecured tranche.

Firms without traded bonds need a synthetic rating. Build an interest coverage ratio (EBIT / interest expense), map it to a Moody's/S&P rating bucket, and apply the spread that bucket trades at today over the Treasury curve. Damodaran publishes the table; for example, in a recent year, an interest-coverage ratio of 4× mapped to roughly BBB and a 1.5% spread.

The tax shield comes from multiplying by (1 − T_c). Use the marginal corporate tax rate, not the effective rate. Marginal is what an extra dollar of interest deduction saves; effective rates blend in foreign earnings, prior-year credits, and one-off items that don't apply forward. U.S. marginal: 21% federal plus ~5% state-weighted ≈ 26% blended.

How WACC turns into a valuation

A discounted-cash-flow valuation projects unlevered free cash flow (FCF) for an explicit forecast period and then assumes perpetuity growth afterwards:

Enterprise Value
   = Σ  FCF_t / (1 + WACC)^t        (years 1 to N, explicit forecast)
   + TV / (1 + WACC)^N              (terminal value, year N+ onwards)

with  TV  =  FCF_(N+1) / (WACC − g)         (Gordon growth)

To get equity value: subtract net debt from enterprise value, add non-operating assets (cash, equity stakes), and divide by shares outstanding. The discount rate enters every term — both the explicit-period present values and the terminal value — so its sensitivity is enormous.

The terrifying sensitivity

Differentiate the perpetuity value V = FCF / (WACC − g) with respect to WACC:

dV / V  =  − dWACC / (WACC − g)

The spread (WACC − g) sits in the denominator, and that spread is small in practice. Plug in real numbers:

WACC	g	WACC − g	1% WACC change → value change
8%	2%	6%	−17%
7%	3%	4%	−25%
10%	5%	5%	−20%
6%	3%	3%	−33%
12%	2%	10%	−10%

For high-growth names where the spread compresses, a half-percent WACC argument can swing a price target by a third. This is why two competent analysts can value the same firm 50% apart while agreeing on every cash-flow assumption — and why presentations of DCF results always include a WACC × growth-rate sensitivity grid.

Where real-world WACCs land

Company / sector	Typical β	r_e (CAPM, ERP=5%, r_f=4%)	D/V	WACC	Why
U.S. regulated utility	0.45	~6.3%	0.45	~5.5–6.0%	Stable regulated cash flows; heavy debt; rate-base returns
Consumer staples (P&G, KO)	0.6	~7.0%	0.30	~6.5%	Defensive demand, moderate leverage
Apple	1.2	~10.0%	0.05	~8.0%	Slight tilt to market, lightly levered, huge cash pile
S&P 500 average	1.0	~9.0%	0.25	~8.0%	Definition: market beta = 1
Software / SaaS growth	1.3	~10.5%	~0	~10.5%	All-equity, high beta, growth optionality
Pre-revenue biotech	1.7	~12.5%	~0	~12–15%	Binary trial outcomes, no debt service
Cyclical materials	1.4	~11.0%	0.35	~8.5%	High beta tempered by tax-shielded leverage

Utilities sit at one extreme — regulated returns, lots of debt, low beta — and biotech at the other — option-like equity payoffs, no debt, beta near 2. Most large-cap names land between 7% and 10%. A WACC outside 5–15% almost always signals a calculation error or an extreme balance sheet.

Modigliani-Miller and the capital-structure question

Modigliani and Miller's 1958 paper made a sharp claim. With no taxes, no bankruptcy costs, and no informational frictions, capital structure does not affect firm value — and therefore does not affect WACC. The logic is arbitrage: if a levered firm and an identical unlevered firm sold at different prices, an investor could replicate the levered firm's payoffs at home by borrowing personally. Whatever leverage premium existed would be arbitraged away.

In WACC terms: as you add debt and lower D/V's contribution from cheap debt, the cost of equity rises by exactly enough — because shareholders now bear more financial risk — to keep the weighted average flat. This is Modigliani-Miller Proposition III, the "WACC invariance" result. Visually, plot WACC against D/V and you get a horizontal line.

The 1963 correction inserted taxes. Interest is deductible; dividends are not. So the after-tax cost of debt is lower than the pre-tax cost by exactly T_c · r_d. As leverage rises, the tax shield grows, and WACC falls. Carried to its logical extreme, the optimal structure is 100% debt — which obviously isn't observed. What pulls WACC back up at high leverage is financial distress costs: bankruptcy administrative fees, lost customers, fire-sale asset values, key-employee departures. The expected present value of these costs grows nonlinearly with leverage. Combined with the linear tax shield, you get a U-shaped WACC curve, with a minimum at the firm's optimal capital structure.

V_levered = V_unlevered  +  PV(tax shield)  −  PV(financial distress costs)

This is the modern trade-off theory of capital structure. The optimal D/V varies by industry: regulated utilities sit at 0.5+ because cash flows are stable and predictable; biotech sits near zero because cash flows are nonexistent until trials succeed.

When WACC breaks: APV

WACC implicitly assumes constant target leverage — the firm rebalances to D/V every period as equity value drifts. This works for steady-state public companies. It breaks for transactions where capital structure is changing materially:

• Leveraged buyouts that pay down debt aggressively over five years.
• Project finance with a specific debt tranche tied to a specific asset.
• Cross-border deals where the tax shield depends on where debt is parked.
• Distressed restructurings, where leverage is anything but stable.

Stewart Myers introduced Adjusted Present Value (APV) in 1974 as the cleaner alternative. APV decomposes value into two pieces: an all-equity (unlevered) base case discounted at the unlevered cost of capital r_u, plus the present value of side effects — primarily the interest tax shield, discounted at the cost of debt (or r_u, depending on convention).

V_APV  =  Σ FCF_t / (1 + r_u)^t   +   Σ (T_c · interest_t) / (1 + r_d)^t   −   PV(distress costs)

APV makes the tax shield value explicit, which is exactly what you want when leverage is in flux. Most LBO models use APV implicitly, even if labelled "WACC valuation": they unlever the cash flows, project debt paydown, and add the tax-shield value directly.

WACC as the hurdle rate

For project evaluation, WACC is the hurdle rate. A project's NPV at the firm's WACC is positive if and only if the project's unlevered IRR exceeds WACC. Conceptually:

NPV  =  −Investment₀  +  Σ FCF_t / (1 + WACC)^t

Accept if  NPV > 0    ⇔   IRR > WACC

This is the standard capital-budgeting rule. Two important refinements:

• Divisional WACCs. A conglomerate evaluating a project should use the WACC appropriate to the project's risk, not the firm's blended WACC. ExxonMobil evaluating a software acquisition should not discount at oil-major WACC; the cash flows have different beta exposure.
• Subsidised cost of capital. Government-backed loans, R&D tax credits, and strategic-partner subsidies change the effective WACC for specific projects. Use the project-specific WACC, not the firm-wide one.

Common pitfalls

• Book weights. Using book equity for E/V drastically understates the equity weight for mature profitable firms. Always use market values.
• Historical realised premium as forward-looking. The 6–7% historical U.S. premium reflects what was earned; forward-looking implied premia from S&P 500 prices and forecast earnings have run 4–5% in recent decades. Defaulting to the historical figure inflates WACC.
• Coupon as cost of debt. The coupon is yesterday's required return. Today's cost of debt is the yield bondholders demand today — pull yield-to-maturity from a recent trade, not the prospectus.
• One beta forever. Five-year monthly betas move 20–30% between estimation windows. Re-estimate periodically; consider industry-average levered betas for stability.
• Mismatched tax rates. Use the marginal corporate tax rate (what one extra dollar of interest saves), not the effective rate (which mixes in international and one-off items). For multinationals, the relevant marginal rate is the jurisdiction-weighted blend of where the interest gets deducted.
• Static WACC for changing leverage. Don't use a single WACC for an LBO valuation or a firm in transition. Either iterate WACC year-by-year for the projected D/V path, or switch to APV.
• Discount-rate inconsistency. Cash flows are forecast in nominal currency; the discount rate must match. Don't apply a nominal WACC to real cash flows or vice versa, and don't apply a USD WACC to EUR cash flows without an FX adjustment.