Lesson 4 of 5
The equity risk premium
The one input in a discounted cash flow that nobody files. Three ways to estimate it, and what two points of it do to a valuation.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- The equity risk premium is the extra return investors as a group require for holding stocks over a government bond, and no filing states it.
- It is estimated three ways, historical, survey and implied, and the three routes disagree with each other by design.
- Two points of premium can move a valuation by more than a fifth, almost entirely through the terminal value rather than the near-term forecast.
Every rate built with the capital asset pricing model needs this premium. It is not a company fact and no filing contains it, because it is not a property of a company. It is a view about the price of equity risk in general, on the day the valuation is made.
Two points of premium, and where the fall lands
- Value at a 5% premium1,171
- Forecast years, years 1 to 5−22
- Terminal value−245
- Value at a 7% premium= 904
Three routes, and they disagree
Damodaran, in Investment Valuation, sets out three routes and is explicit that they do not agree, because each measures a different thing.
| Route | What it measures |
|---|---|
| Historical | The realised excess of stock returns over a government bond, averaged over some past period |
| Survey | What investors say they require, collected by asking them |
| Implied | Backed out of today's prices and a forecast of the cash they will distribute |
A historical premium assumes the next decade resembles the sample it was measured over. A survey premium measures opinion, and opinion moves with recent returns. An implied premium is forward-looking by construction, which is Damodaran's own case for preferring it.
Why the terminal value takes the hit
A perpetuity is a cash flow divided by the rate less its growth rate, so a change in the rate strikes it far harder than it strikes a handful of discounted years. In the figure above, the terminal value supplied 792 of the 1,171 at a 5% premium and 547 of the 904 at a 7% premium. So 245 of the 267 point fall landed there, and only 22 on the five forecast years. The least verifiable input lands almost entirely on the least verifiable part of the answer.
Write the route down beside the number
Whatever premium you use, write down which route produced it and the date, next to the valuation. Historical, survey or implied; which period or which index level. A spreadsheet still opens a year later; the reasoning behind one number inside it does not survive unless somebody wrote it down at the time.
Check yourself
What is the equity risk premium?
- The current dividend yield on the stock market
- The extra return investors as a group require for holding stocks over a government bond
- The interest rate a company pays on its own bonds
Show the answer
Answer: B.
It is a market-wide view about the price of equity risk, not a fact about any one company.In the worked example, a rise from a 5 percent to a 7 percent premium moved the total from 1,171 to 904. Where did most of that 267 point fall come from?
- The five forecast years
- The terminal value
- Split evenly between the two
Show the answer
Answer: B.
Terminal value is a perpetuity divided by rate less growth, so it absorbs almost all of a rate change: 245 of the 267 points here.What is the right way to size an unestimated adjustment to a discount rate, for something like small size or illiquidity?
- Add a fraction of a point that feels reasonable
- There is not one: an adjustment needs its own estimation procedure or it should not be added
- Always round the discount rate up to the next whole point
Show the answer
Answer: B.
An addition with no procedure behind its size is a preference wearing a decimal point, and it cannot be checked by anyone reading the valuation later.Sources
- Williams, The Theory of Investment Value, 1938: value as the present value of distributable cash.
- Sharpe's capital asset pricing model: expected return as the risk-free rate plus beta times the premium.
- Damodaran, Investment Valuation and The Dark Side of Valuation: the three estimation routes, and the warning against unestimated adjustments.
What this leaves out
- No premium is suggested here.
- The valuation above uses round invented inputs so every figure can be checked by hand, and is a single-company sketch rather than a method for valuing anything.
- The WACC calculator asks you for your own cost of equity. The return-on-capital-less-cost-of-capital spread on research pages uses the product's own estimate built from the company's beta, not a figure you chose.
- Nothing here is advice about any security.