A Project and Division Discount-Rate Case

Why a Discount-Rate Case Requires Judgment

A discount rate for a project or division is an estimate of the return investors require for bearing the risk of that activity. The company’s overall rate is not automatically appropriate: a project may have different operating risk, and a division may have different cash-flow sensitivity from the rest of the firm. The practical task is to use the best available comparable businesses, adjust for differences that matter, and show how uncertainty in those adjustments could affect the decision.

Comparable-company estimates are imperfect. A peer may operate in more than one business, carry a different amount of debt, hold excess cash, or face different operating conditions. A defensible estimate therefore requires more than copying a reported beta or averaging rates mechanically. The analyst should state the assumptions, use market-value financing weights where the WACC model calls for them, and examine whether reasonable alternative assumptions change the conclusion.

Definition: A project discount rate is the required return used to value cash flows with the project’s risk. A divisional discount rate is the corresponding required return for a division’s operating risk. A comparable is a business whose operating risk is informative about the project or division being assessed. A comparable is rarely identical, so its observed equity beta and financing mix must be interpreted before they are used.

A Consistent Estimation Framework

The CAPM estimates a risky asset’s required return as the risk-free rate plus its beta times the market risk premium. For equity, $r_f$ is the risk-free rate, $\beta_E$ is the equity beta, and $E[R_M]-r_f$ is the expected market risk premium. The CAPM assumes investors are diversified, care about systematic risk, and form expectations in a market represented by the chosen market portfolio. It does not compensate for diversifiable firm-specific risk, and its estimate is only as reliable as its beta and market-premium inputs.

k_E = r_f + \beta_E\bigl(E[R_M]-r_f\bigr)

For a company financed with debt and equity, a standard after-tax WACC is $E/V$ times the cost of equity plus $D/V$ times the after-tax cost of debt, where $V=D+E$. Here $D$ and $E$ are market values, $k_D$ is the pre-tax cost of debt, and $T$ is the corporate tax rate. This form assumes the interest tax shield is available at the stated rate and that the financing weights are suitable for the project’s financing policy. It is not appropriate to apply the tax adjustment when interest is not tax-deductible or the assumed tax benefit cannot be used.

\mathrm{WACC} = \frac{E}{V}k_E+\frac{D}{V}k_D(1-T), \qquad V=D+E

To compare operating businesses with different leverage, analysts commonly use the Hamada relation. Under its simplifying assumptions—debt beta is negligible, the tax shield is treated as having debt-like risk, and the debt-to-equity ratio is measured consistently—the relationship between equity beta and asset beta is $\beta_E=\beta_U[1+(1-T)D/E]$. Here $\beta_U$ is the unlevered, or asset, beta. This is a model, not a universal identity: material debt risk, changing leverage, or a different tax-shield risk can require a more tailored treatment.

Worked Example: Estimating a Project Rate From Two Comparables

A proposed software-services project has operating risk judged comparable to two public businesses. Comparable A has equity beta 1.20 and market $D/E=0.20$; comparable B has equity beta 1.35 and market $D/E=0.40$. Use a 25% corporate tax rate. Assume both peers are pure plays, their debt betas are negligible, and the tax-adjusted Hamada relation is suitable. The project is expected to use target $D/E=0.30$. Let $r_f=3.5\%$, the expected market risk premium be 5.0%, and the project’s pre-tax debt cost be 5.5%. Estimate its WACC.

Method. First remove each peer’s financial leverage to estimate operating beta. Then average those asset-beta estimates, apply the project’s target leverage, estimate the cost of equity by CAPM, and calculate WACC using the target market-value weights.

For comparable A, $(1-T)D/E=0.75(0.20)=0.15$, so its denominator is $1+0.15=1.15$. Thus $\beta_{U,A}=1.20/1.15=1.0434783$.

For comparable B, $(1-T)D/E=0.75(0.40)=0.30$, so its denominator is $1+0.30=1.30$. Thus $\beta_{U,B}=1.35/1.30=1.0384615$.

The equally weighted estimate is $(1.0434783+1.0384615)/2=1.0409699$. Equal weighting is a modeling choice here: it is reasonable only if the two peers are comparably informative. It does not imply that the businesses have equal market value.

At the project’s target $D/E=0.30$, the relevering factor is $1+0.75(0.30)=1.225$. Therefore, $\beta_E=1.0409699(1.225)=1.2751886$.

CAPM gives $k_E=3.5\%+1.2751886(5.0\%)=3.5\%+6.375943\%=9.875943\%$. The target $D/E=0.30$ implies $E/V=1/1.30=0.7692308$ and $D/V=0.30/1.30=0.2307692$. The after-tax debt cost is $5.5\%(1-0.25)=4.125\%$.

The equity contribution is $0.7692308(9.875943\%)=7.596879\%$, and the debt contribution is $0.2307692(4.125\%)=0.951923\%$. Thus WACC is $7.596879\%+0.951923\%=8.548802\%$, or 8.55% to two decimal places. As a check, the weights sum to $0.7692308+0.2307692=1.0000000$, and the WACC lies between the after-tax debt cost and the cost of equity.

Common mistake avoided: Using $D/E=0.30$ as though it were $D/V=0.30$ would give incorrect WACC weights. The stated ratio is debt relative to equity; converting gives $D/V=0.30/1.30$, not 0.30.

When Comparable Evidence Is Uneven

A division may have several imperfect comparables rather than two clean pure plays. Before combining their evidence, ask whether each peer’s beta reflects the operating activity of interest. A peer with substantial unrelated operations may be less informative than a smaller but more focused business. A peer’s cash and nonoperating assets can also affect its observed equity beta; the analyst should treat those assets consistently rather than attributing their risk to the operating division.

For a division, it is often useful to distinguish three stages: estimating the division’s operating beta from peers; relevering that beta to the division’s assumed financing mix; and translating the resulting beta into a cost of equity and WACC. The target leverage must reflect the financing assumption relevant to the valuation. A companywide financing mix is not automatically the right choice for every division.

Worked Example: A Divisional WACC With Rounded Inputs

A division’s market-based analysis has produced an estimated equity cost of capital of 11.0202%. For this calculation, take that estimate as the CAPM result; equivalently, it can be generated by $r_f=3.00\%$, market risk premium $5.50\%$, and $\beta_E=1.45821818$, since $3.00\%+1.45821818(5.50\%)$ is approximately $11.0202\%$. The division’s target market-value weights are $E/V=0.7143$ and $D/V=0.2857$. Its pre-tax debt cost is 6.20%, and its tax rate is 25%. Calculate WACC using the stated rounded weights.

Method. Apply the tax adjustment to the cost of debt, multiply each financing cost by its market-value weight, and add the contributions. The two supplied weights sum to $0.7143+0.2857=1.0000$.

The after-tax cost of debt is $6.20\%(1-0.25)=6.20\%(0.75)=4.65\%$.

Using the stated weights, the equity contribution is $0.7143(11.0202\%)=7.87172886\%$, which rounds to $7.8717\%$ at four decimal places. The debt contribution is $0.2857(4.65\%)=1.328505\%$, which rounds to $1.3285\%$ at four decimal places.

Therefore, WACC is approximately $7.8717\%+1.3285\%=9.2002\%$, or 9.20% to two decimal places. Using the unrounded products gives $9.20023386\%$, which also rounds to 9.20%. As a check, the answer is between the after-tax debt cost of 4.65% and the equity cost of 11.0202%, as positive weights summing to one require.

Common mistake avoided: $0.7143(11.0202\%)$ is $7.87172886\%$, not $7.8716\%$. Keep enough digits in the weighted products before rounding; rounding one contribution incorrectly can obscure whether the final sum is reliable.

Why the Beta Adjustment Works

The leverage adjustment can be justified by considering the risk borne by equity holders. Under the Hamada assumptions, the operating assets have value $V_A$, debt has value $D$, and equity has value $E$. The asset return is a value-weighted combination of the returns to debt and equity. If debt beta is negligible, the asset beta is approximately the equity beta multiplied by the equity share of the asset value. With the standard tax-shield adjustment, this relationship rearranges to the Hamada formula.

\beta_U \approx \frac{E}{E+(1-T)D}\beta_E \quad\Longleftrightarrow\quad \beta_E=\beta_U\left[1+(1-T)\frac{D}{E}\right]

The second expression follows by dividing $E+(1-T)D$ by $E$. It shows why comparable equity betas cannot be averaged directly when leverage differs: more leverage magnifies the equity beta even when operating risk is unchanged. Unlevering aims to compare operating risk; relevering then maps the estimated operating risk to the leverage assumed for the project or division. The approximation should not be treated as precise when the assumptions are materially violated.

Estimation Uncertainty and the Decision

A calculated rate is not an exact measurement. Uncertainty can arise from the peer set, beta estimates, tax assumptions, target leverage, the market risk premium, and the degree to which the business is truly comparable. Reporting a single point estimate without examining these choices can make a fragile decision look certain.

A practical response is to calculate a base estimate and a reasonable range using defensible alternative assumptions. The range is not a probability distribution unless one has been specified and justified. It is a sensitivity analysis: it shows how the valuation changes under alternative inputs. If the decision changes within a plausible range, the analyst should report that dependence rather than present the base case as conclusive.

Worked Example: A Project Decision That Changes Across the Rate Range

A hypothetical project requires an initial investment of $12.00 million and is expected to generate a level after-tax cash flow of $1.08 million at the end of each year forever, beginning one year from now. Assume no growth, no terminal asset value, and a constant discount rate. Comparable analysis supports a plausible project-rate range from 8.5% to 10.4%, with a base estimate of 9.2%. Determine whether the project’s NPV is positive throughout that range.

Method. Under the stated level-perpetuity assumption, the present value of the cash flows is $C/r$, where $C$ is the annual cash flow and $r$ is the annual discount rate. NPV equals that present value less the initial investment. This valuation is not appropriate if the cash flow is temporary, growing, risky in a way not represented by the discount rate, or paid at a different time.

At 8.5%, the present value is $$1.08\text{ million}/0.085=$12.705882\text{ million}$. NPV is $$12.705882-$12.00=$0.705882\text{ million}$, positive by approximately $705,882.

At the 9.2% base estimate, present value is $$1.08/0.092=$11.739130\text{ million}$. NPV is $$11.739130-$12.00=-$0.260870\text{ million}$, negative by approximately $260,870.

At 10.4%, present value is $$1.08/0.104=$10.384615\text{ million}$. NPV is $$10.384615-$12.00=-$1.615385\text{ million}$, negative by approximately $1.615 million.

As a check, the break-even rate solves $1.08/r=12.00$, so $r=1.08/12.00=0.09$, or 9.0%. The NPV is positive below 9.0% and negative above 9.0%, consistent with the three calculations. Therefore, it is not positive throughout the plausible range: the decision changes with the discount-rate estimate. The appropriate conclusion is to identify the sensitivity and assess the assumptions behind the range, not to select a favorable rate simply because it yields a positive NPV.

Common mistake avoided: Treating the 9.2% base estimate as certain would hide that the plausible low-rate case has positive NPV while the base and high-rate cases do not. A range does not prove which rate is correct, but it does reveal whether the decision is robust.

Common Traps in a Discount-Rate Case

Mixing up $D/E$ and $D/V$. Convert the ratio to the weights required by the WACC formula. If $D/E=x$, then $D/V=x/(1+x)$ and $E/V=1/(1+x)$.
Adjusting for tax twice or not at all. In the stated WACC formula, multiply pre-tax $k_D$ by $1-T$ once. Do not use an after-tax debt cost and then apply the tax factor again.
Using observed peer equity betas as operating betas. When leverage differs and the model’s assumptions are acceptable, unlever first and relever to the target financing mix.
Assuming imperfect peers are interchangeable. Explain why a peer is informative, and consider whether its business mix, nonoperating assets, or financing makes it less comparable.
Reporting false precision. Beta, market premiums, and target leverage are estimates. Carry sufficient digits for arithmetic, but report a sensible final precision and show how the conclusion responds to plausible alternatives.
Changing the discount rate to compensate for a separate risk adjustment. A risk already reflected in the cash-flow forecast or in a rate adjustment should not be counted again without a clear model.

Key takeaway: A project or divisional discount rate is a reasoned estimate, not a directly observed fact. Use comparable operating risk, make leverage and tax assumptions explicit, calculate market-weighted returns consistently, and test whether uncertainty changes the decision.

Check Your Understanding

For each problem, state the calculation clearly and check that the result is consistent with the assumptions.

A comparable has equity beta 1.25, market $D/E=0.20$, and a 25% tax rate. Under the Hamada assumptions, what is its unlevered beta? (Answer: $1.25/[1+0.75(0.20)]=1.0870$, approximately.)
A division has estimated unlevered beta 0.95. Its target $D/E=0.40$, and the tax rate is 30%. Under the Hamada assumptions, what is its relevered equity beta? (Answer: $0.95[1+0.70(0.40)]=1.216$.)
For a project, $r_f=4.0\%$, $\beta_E=1.10$, and the expected market risk premium is 5.5%. What cost of equity does CAPM imply? (Answer: $4.0\%+1.10(5.5\%)=10.05\%$.)
A project has $E/V=0.60$, $D/V=0.40$, $k_E=10.0\%$, pre-tax $k_D=6.0\%$, and tax rate 25%. What is its after-tax WACC? (Answer: $0.60(10.0\%)+0.40(6.0\%)(0.75)=7.80\%$.)
A hypothetical project costs $10.00 million and generates $0.90 million at each year-end forever, with no growth. What is its NPV at an 8% rate and at a 10% rate, and does the decision remain unchanged? (Answer: At 8%, NPV $=$0.90/0.08-$10.00=+$1.25$ million; at 10%, NPV $=$0.90/0.10-$10.00=-$1.00$ million. The decision changes.)

What You'll Learn