Weighted average cost of capital: one rate for all the money

A company's money comes from lenders and from owners, and each wants a different return. When the company judges a new factory, values its own business or sets a hurdle for a project, it needs a single rate that stands for the cost of all its money together. That blended rate is the weighted average cost of capital, or WACC. Each source is weighted by its share of the funding, so a company funded mostly by owners has a WACC close to its cost of equity, and one funded mostly by lenders has one close to its after-tax cost of debt.

The last two lessons built the parts. This lesson puts them together, shows the choice of weights that causes most mistakes, and applies the method to Apple, where the answer is very different depending on whether the weights are taken from the accounts or from the market.

Debt enters after tax, for the reason given in the lesson on the tax shield. Equity enters at its full cost, because dividends are not deductible. Each weight is the share of the funding that source supplies. Suppose a company is 70% equity costing 10.5% and 30% debt costing 5% before tax with tax at 21%. The WACC is 0.70 × 10.5% + 0.30 × 5% × 0.79 = 7.35% + 1.185% = 8.54%.

The weights should reflect what each source is worth now, so market values are the standard. For equity that is the market value of the shares: the share price times the number of shares. For debt, market value is close to the amount owed unless rates have moved a great deal, and book debt is a usual stand-in. Using the balance sheet figure for equity is the common error. Book equity records the money put in and the profit kept, less buybacks and losses. It can be a small fraction of what the owners' stake is worth, or even negative. Using it overstates the weight on debt and understates the WACC.

Some analysts use a company's target structure instead of today's, on the argument that the mix changes in the long run. Others use net debt, debt less cash, because cash could repay the debt. Whichever route is taken, it should match the cash flows being discounted and be stated.

Set the share of capital that is debt, the cost of equity when the company has no debt, the extra return owners ask for each ten points of debt, the interest rate before tax and the tax rate. The left page shows the weights and each cost at this mix. The right page shows the blended cost, and how it compares with the cost of a company with no debt. The extra return for owners matters: as debt rises, owners' returns become riskier, so they ask for more. All settings are illustrative.

The first experiments might suggest that a company can lower its cost of capital by borrowing more, since debt is cheaper. That reasoning leaves out the effect on owners. When the company borrows, the same operating profit is shared among fewer owners with a fixed claim ahead of them, so their returns swing more and they ask for more. Economists Franco Modigliani and Merton Miller showed that, in a world without tax or costs of failure, the rise in the cost of equity exactly offsets the cheaper debt, and the cost of capital does not change with the mix.

Real companies are not in that world, and two things break the result. Tax makes debt cheaper by the shield, which pulls the WACC down. Costs of financial distress, such as legal fees, lost customers and forced sales of assets, grow as debt rises, which pulls it back up. The trade-off between the two gives each company a range of sensible borrowing. The slider for owners' extra return stands in for how steeply the second effect rises.

WACC is a discount rate for cash flows that belong to all providers of capital, lenders and owners together. It is the rate at which free cash flow is discounted to value a whole business, and it is the usual hurdle rate for projects that carry the same risk as the existing business. In the lesson on net present value, a project's cash flows were discounted at the required return. WACC is that required return for an average project of the company.

The default blend of 8.54% shows the effect. A project that costs $100,000 and returns $30,000 a year for five years has a net present value of $18,111 at that rate, and of $13,724 at 10%. Because WACC is a single rate, it is right only for projects whose risk matches the company's. A project that is much riskier than the business needs a higher rate, and one that is much safer a lower one.

Apple shows why the choice of weights matters. At 27 September 2025 its debt, commercial paper plus term debt, was $98,657 million, and its book equity $73,733 million. The market value of its shares held outside the company was $3,253,431 million, at 28 March 2025. For the costs, the lesson on the cost of debt gave a range of 4.07% to 4.83% for notes issued in 2025. Take the middle, 4.45%, and the effective tax rate of 15.6% from Apple's income statement, which gives a cost of debt after tax of 3.76%. For the cost of equity, this lesson uses an illustrative 9%, not an estimate of Apple's actual cost.

The weights in the cost of capital should be the market values of debt and equity, because those are what investors would have to be paid for today. A common shortcut takes the weights from the balance sheet, and for some companies this gives a seriously wrong result. Apple shows it. Using the same illustrative costs for both calculations, a cost of equity of 9% and a pre-tax cost of debt of 5%, and Apple's effective tax rate of 15.6%, the two approaches give very different answers.

• Book weights: equity of $73,733 million and borrowings of $98,657 million make debt 57.2% of the total, and the cost of capital 6.26%.
• Market weights: equity valued at $3,253,431 million, the market value of non-affiliates' shares stated on the cover of its annual report at 28 March 2025, makes debt 2.9% of the total, and the cost of capital 8.86%.

The weights alone move the answer by 2.6 percentage points. Book equity is small because of years of buybacks, and the cheaper debt gets far too much weight in the book calculation. The market-weight figure is the one that matches how investors price the business. The figures use the market value at a date that does not match the balance sheet date, so they are an illustration of the method, not a measured result.

A multinational has cash flows in many currencies, and the cost of capital belongs to the currency of the cash flows, as the earlier lessons showed for equity and debt. In a hypothetical case where the dollar cost of capital is 7.425% (equity at 9% with 70% weight, debt at 5% before tax with 30% weight and a 25% tax rate), the equivalent in yen, when the yen interest rate is 1% and the dollar rate is 4%, is 4.33%. A project with yen cash flows must clear the yen figure, not the dollar one.

A company with a finance arm adds another layer. A manufacturer with a captive lender, like Toyota, has two businesses with very different funding. A single blended cost of capital understates the manufacturing business's risk, and overstates the lender's. A careful analysis values each part with its own cost of capital, or at least tests how far the answer moves with the blend.