Recently, I have been thinking about what the “correct” value of the equity risk premium should be. The equity risk premium is the extra return over the risk-free return (typically set as the yield of a long-term Treasury bond) that investors demand for investing in a stock. Surveys of investment managers indicate that most use a premium of 4-7%. Historically, returns from the stock market have been 3-5% above bond yields.
Of course, stocks differ in volatility and risk, therefore risk premiums should be stock-specific. One way to evaluate the risk of a stock is through its beta. Beta is a measure of a stock’s volatility as compared to the entire stock market. If a stock moves exactly in line with the market, it has a beta of 1. A stock with a beta of less than 1 moves less than the market does (e.g. utility stocks), and a stock with a beta greater than 1 moves more than the market (e.g. small caps). Theoretically then, stocks with similar earnings per share and earnings growth will have their PEs determined primarily by their beta. In practice, this is rarely the case. (In fact, some hedge funds use a strategy known as beta arbitrage, relying on the fact that some stocks are not priced as their beta suggests they should.)
For my own portfolio, I divide stocks into 3 rough categories. Low risk stocks are defensive stocks that are robust in the face of business cycles, including stocks in the retail, utility, healthcare, and food industries. I require that low risk stocks have a return rate of riskless rate + 4%. Moderate risk stocks include stocks such as consumer goods, business goods, and luxuries (middle income people stop their purchases in a depressed economy, but rich people stay rich and tend not to cut back on purchases). Moderate risk stocks are discounted with a rate of riskless + 6%. High risk stocks include companies dealing in commodities (companies with zero pricing power), consumer durables (people delay major purchases in a down economy), travel, and property. High risk stocks are discounted with a rate of riskless + 8%.
In the end, I decided not to spend too much time on this subject. The use of the various equity risk premiums has only a moderate impact on valuations, and in many cases, growth assumptions are a bigger factor. Perhaps that is why many investors use only a single value as a generic equity risk premium.
