Capital Structure and the Cost of Capital

Scope. This article shows how a blended cost of capital is constructed and how much the answer moves with the inputs. It is not a claim about what any company's cost of capital is, because a cost of capital is not a fact about a company — it is an estimate assembled from choices. No optimal capital structure is identified, no leverage level is recommended, and no threshold is given for any measure. Discounting a cash-flow stream to a value belongs to Pillar 25; this article builds the rate and values nothing.

That combination is the weighted average cost of capital, and it is the single most-used and least-examined number in corporate finance.

It is worth being blunt about its status at the outset. Revenue is a fact. Interest paid is a fact. A cost of capital is a construction — it contains an expected return nobody observes, a weighting scheme somebody chose, and a tax rate that may not apply at the margin. Treating the output as a measurement rather than an estimate is the error this article is built to prevent.

The construction, step by step

Using Wexford Instruments, the Pillar 22 parameter set, and the bases fixed on this pillar's hub.

Two conventions inside that table deserve naming rather than assuming. Equity is weighted at market value and debt at book, which is standard and slightly inconsistent — debt has a market value too, and it is used less often because it is harder to observe. And the tax shield is applied at the effective rate, which assumes the company can use the deduction; a company with losses cannot, and its after-tax cost of debt is simply its pre-tax cost.

How much the answer moves

The same company, the same year, the same statements — and three defensible constructions.

And the input that moves it furthest is not a construction choice at all — it is the beta.

Worked example — the range is wider than most of the things the number is used to decide. Across the three canonical beta reference points the cost of capital runs from 6.32% to 9.43% — a span of 3.12 percentage points — and the choice between them is a judgement about how a company's returns move with the market, estimated from a history that may not describe the future. Across construction choices alone, holding beta fixed, the range is still 6.99% to 8.15%. Set that against what the number is used for. The Pillar 22 risk-free-rate article records that a one-percentage-point change in a discount rate moves a valuation by 12.5% — so the spread of defensible costs of capital for a single company is enough to move a valuation by tens of percent, before anyone has disagreed about the business. This is not an argument against computing it. It is an argument for stating the inputs beside the output every time, which is what this portal does and what a single quoted percentage never does.

The question capital structure actually asks

If debt is cheaper than equity, does adding debt lower the cost of capital? The arithmetic appears to say yes, emphatically.

That table is wrong, and it is wrong in the most instructive way available. It holds the cost of equity constant at 9.00% while loading fixed obligations ahead of the equity claim — which cannot be right, because the residual becomes more volatile as more of the company's cash flow is committed before shareholders are paid. The tell is the shape: the table declines without limit, implying that a company financed entirely by debt would have the lowest cost of capital of all. Anything that recommends infinite leverage has an input that is not responding. What the correct treatment does is offset the two effects. Substituting cheap debt for expensive equity lowers the average; the rising cost of equity pushes it back up; and how far the tax shield tips the balance is the subject of a large theoretical literature that this portal does not adjudicate. What can be said without adjudicating anything: the naive table overstates the benefit of leverage, and any argument built on holding the cost of equity fixed while leverage changes is making the same error.

What else changes with leverage, and does not appear in the rate

Fixed obligations reduce the margin for error, and that cost does not show up as a higher percentage. A company with more debt has less room to survive a bad year, fewer options when opportunities appear, and covenants that can constrain it at exactly the moment flexibility is worth most. None of that enters a weighted average.

Which is why capital structure is not solved by minimising a number. If it were, the answer would be at the maximum leverage a lender would accept, and no company behaves that way — including companies whose managements are perfectly capable of the arithmetic. The rate captures price; it does not capture fragility.