What a discount rate is doing

A discount rate is not a fudge factor and it is not a safety margin. It is the return available to you on the next best use of the same money at the same risk. Discounting a future cash flow by it answers one question: how much would you have to commit today, elsewhere, to end up with that cash flow at that date.

Everything that is awkward about picking one follows from that sentence, because "the next best use at the same risk" is a fact about you and your alternatives rather than a fact about the company.

Two things it is pricing, and they are not the same thing

The first is time. A cash flow arriving in five years is worth less than the same cash flow today, even if it is certain, because the money could have been earning something in the meantime. That part is the risk-free rate, and it is observable: it is what a government bond of the same maturity yields.

The second is risk. The cash flow is not certain, and you require compensation for that. This part is not observable anywhere. It is the excess return you personally require over the risk-free rate to hold this business rather than the bond, and no document states it.

A discount rate is those two added together. When people argue about discount rates they are almost always arguing about the second part while quoting the sum.

Why it is the company's cost of capital and not your hurdle rate

The standard rate for valuing a whole business is its weighted average cost of capital: what the company pays for the money it uses, weighted by how much of each kind it has. That is a fact about the company's financing, not about your alternatives, which looks like a contradiction of the paragraph above.

It is not, and the reason is worth holding onto. A discounted cash flow of the whole firm values the cash available to everyone who financed it, debt and equity together. So the rate has to be the return everyone who financed it requires, blended. The company's cost of capital is that blend. Use your own hurdle rate instead and you have valued something, but not the thing the cash flows describe.

The practical consequence: if you discount cash flows to the firm, use the firm's cost of capital and subtract net debt at the end. If you discount cash flows to equity, use the cost of equity alone and subtract nothing. Mixing the two is the most common error in the whole exercise, and it does not announce itself, because both versions produce a number that looks like a valuation.

The tax term, in one paragraph

Interest is deductible against corporate tax and dividends are not. A company paying 5 percent on its debt at a 21 percent tax rate is really paying 3.95 percent, because the interest reduces its tax bill. That is why the debt side of the formula is multiplied by one minus the tax rate and the equity side is not.

This is also why the formula quietly says that debt is cheap, and why a mechanical reading of it says the cheapest possible capital structure is almost all debt. The formula does not model the part where a heavily indebted company's cost of equity rises, its cost of debt rises, and eventually its lenders stop lending. Treat the tax term as an accounting fact and not as advice about financing.

Why nobody can give you the number

Two people can compute a correct cost of capital for the same company on the same day and get 7 percent and 11 percent. Neither has made an arithmetic error. They chose different equity risk premiums, measured beta over different periods, or used a different maturity for the risk-free rate. All of those are defensible choices and the formula multiplies the difference between them straight into the answer.

That is the actual state of the art, and knowing it is more useful than any particular rate. It is why the right output of a valuation is a range with the assumptions written next to it, and why a single number carried forward with no note of where it came from is the point at which a valuation stops being evidence.

What this leaves out

This lesson is about the reasoning, not about a number. It does not tell you which discount rate to use and it deliberately does not suggest a range, because a suggested range is a recommendation with a hedge on it.

It describes the standard textbook treatment of a discount rate for a going concern with ordinary financing. Financial companies, businesses funded in several currencies, companies with material minority interests or convertible capital, and anything close to distress all need a different treatment, and the weighted average cost of capital is often the wrong tool for them.

Nothing here is advice about any security.