Capital asset pricing model and alternative valuation approaches

CAPM

There are several different models that can be used to help determine the cost of capital for a company. Each is based on a model, and can be understood not only in terms of its formula but also in terms of its underlying assumptions. These assumptions will provide the foundation for the model, and will inform the financial manager about the strengths and weaknesses of each model. This report will outline in detail three such major models for determining the cost of capital. The first is the capital asset pricing model, known as CAPM. The second is the dividend discount model, and the third is arbitrage pricing theory.

The capital asset pricing model is the first of the three major models for determining the cost of capital. CAPM is widely used to determine the cost of equity in particular. The underlying theory of CAPM is that stock returns relative to risk can be used to determine what cost the market needs for that equity. In short, "investors need to be compensated in two ways: time value of money and risk" (Investopedia, 2013). Thus, the return on a security equates to the risk-free rate of money in the economy, plus a risk premium. The risk premium in the capital asset pricing model is comprised of two elements. The first is the general market risk premium, which is the risk of the market in general above and beyond the risk-free rate. The second component of risk in the capital asset pricing model is the firm-specific risk. For equities, the firm-specific risk is reflected in the beta. The beta is the correlation of the firm's stock price movements in relation to the movements of the market as a whole. Thus, the formula for the capital asset pricing model is as follows:

Source: Investopedia (2013)

The capital asset pricing model is easy to use, because all of the information inputs into the model are easy to come by. The risk free rate is the rate of return on Treasury securities, and the market risk premium is often considered to be around 7% above the risk-free rate. The beta is easy to calculate on a spreadsheet and indeed for all publicly-traded securities is available from any number of financial websites like MSN Moneycentral or Yahoo Finance. The capital asset pricing model is relatively accurate for stocks, but less accurate for determining the cost of bonds, since they do not trade as frequently, and there is no base bond index like there is for stocks and thus no beta. The underlying assumptions of the capital asset pricing model are actually quite reasonable. The cost of equity is the return that investors demand, and the capital asset pricing model directly equates the cost of capital to the returns on the stock.

The dividend discount model argues that while stock returns are important, investors only hold a stock for the known cash flows, that is to say the value of a stock is comprised solely of the present value of expected future cash flows. A key assumption here is that investors do not hold stocks for capital gains, as that would be irrational since all future growth should be priced into the stock. The dividend discount model therefore holds that the cost of capital can be determined by estimating the present value of future cash flows and working backwards from the current stock price. The formula for the dividend discount model is therefore:

source: Investopedia (2013)

While this model is consistent with efficient market hypothesis and therefore explicitly rules out the prospect of speculative investment for capital gains, such a view is at odds with market reality. In truth, many investors hold stocks for the prospect of capital gains. While the dividend discount model takes the view that such investors are really holding the stock for future potential dividends, that is at odds with reality. Some companies expressly state that they do not expect to pay dividends in the foreseeable future yet retain high multiples -- investors are looking for capital gains. While the model can be applied to dividend-heavy, slow-growth stocks, it is entirely inappropriate for companies that do not pay dividends or that are more growth-oriented.

The arbitrage pricing theory is similar in structure to the capital asset pricing model, but seeks to explain the cost of capital in terms of correlations with other variables besides stock price. Thus, the beta is replaced by one or more other variables, often key macroeconomic indicators. This makes APT more precise, and allows the user to tailor it to specific conditions relating to the company -- linking bank stock prices to interest rates for example -- but comes with its flaws. For one, APT is much tougher to calculate because unlike the beta, its correlations are not publicly-available and therefore must be calculated. Further, the financial manager must determine what variables, and in what weights, are to be used in the APT formula. This customization and arbitrariness makes arbitrage pricing theory a powerful tool, but it also makes it much tougher to use. The underlying assumptions of arbitrage pricing theory are strong, in that there is more to a stock's value than the beta, and that more accurate information can be gained from having a deeper understanding of stock performance. However, in choosing the appropriate variables, there are going to be assumptions that the financial manager makes. While those assumptions are critical, they may or may not be particularly accurate or viable -- bad assumptions will ultimately lead to a worthless calculation.

The best of the three models is the capital asset pricing model. There are two reasons for reaching this conclusion. The first is that the company needs a model that is easy to use, and CAPM is much easier to use than APT. The second is that the company needs a model that has more realistic assumptions. Unfortunately, the dividend discount model is built on the assumption that all investors are perfectly rational. While this assumption makes for a good theory, it is not particularly reflective of actual market conditions. CAPM strikes the best balance between ease of use and realistic assumptions.

Part II.

The beta is the correlation between stock performance and market performance. Market performance is influenced by any number of different factors. Profits -- and the expectations for future profits -- are the biggest driver of stock market performance. Other factors come into play as well. An example is total leverage -- the ratio of debt to equity in the firm's capital structure. Companies that have too much debt are likely to have more volatile stock market performance because the value of their equity will fluctuate more strongly with performance. The degree to which firm performance is dependent on overall economic performance is an important factor as well. Some companies -- like McDonalds -- perform well in both good and bad markets. A company like Sony is more dependent on things like technology cycles, and that can create gaps between firm and market performance that reflect in a much higher beta. It is important to understand the drivers of stock price changes for each firm to understand where beta comes from.