In a meeting recently, a portfolio manager said that discounted cash flow (DCF) models are worthless. He argued that because future cash flows are discounted at the cost of capital, which is based on the Capital Asset Pricing Model, the value derived from the model is “crap.”
Like so many aspects of investment analysis, academic theory can only be taken you so far. At some point, a more pragmatic approach must be asserted. The calculation of WACC is no exception. As we all know, the weighted average cost of capital is simply the blended cost of equity and debt capital used by a company. The inherent flaw in the WACC calculation is how the cost of equity is derived. Specifically, the cost of equity capital is estimated using the Capital Asset Pricing Model (CAPM), where risk is measured by beta. By using beta, investors are defining risk as volatility. “Using past price movements (or volatility) as the basis for determining the riskiness of a particular stock can often lead to faulty conclusions,” notes Joel Greenblatt in You Can Be A Stock Market Genius. “For example, a stock that has fallen from $30 to $10 is considered riskier than a stock that has fallen from $12 to $10 during the same period. Although both stocks can now be purchased for $10, the stock that has fallen the farthest, and the one that is now priced at the biggest discount to its recent high price, is still considered the riskier of the two. It could be that most of the stock’s downside risk has been eliminated by the huge price drop.” [i] If so, this would argue for a better definition (and measure) of risk.
In my opinion, risk is best defined as a permanent loss of capital. Investors who share my view are more concerned about the downside risk in a stock than they are about its volatility; in fact, upside volatility should be welcomed by long investors. In this context, the riskiness of a stock would be better measured by a valuation multiple. After all, purchasing a high multiple stock, other things being equal, is riskier than purchasing a low multiple stock. With this in mind, Richard Tortoriello in Quantitative Strategies for Achieving Alpha suggests substituting enterprise value-to-sales for beta in the Capital Asset Pricing Model.[ii] Based on his research, the un-weighted enterprise value-to-sales multiple for the S&P 500 Index has averaged ~ 1.0x for the past 20 years and thus serves as a good proxy for risk in a CAPM framework. The only issue that I’ve encountered with this approach is that very high or very low multiples can dominate the cost of capital calculation. I contacted Richard to see if he’d identified a work-around for this issue. He suggested bounding the multiple, using 0.2 on the low end and 2.0 on the high end.
By substituting enterprise value-to-sales for beta, the cost of capital better reflects the risk borne by equity holders of the company. Therefore, those same equity holders can have more confidence in using WACC as the discount rate in a free cash flow model or in evaluating the spread between return on capital and the cost of capital in an EVA model. It isn’t perfect, but no model is. That’s why I always suggest “triangulating” on valuation using multiple methods as noted in a prior post.
[i] Greenblatt, Joel. You Can Be A Stock Market Genius: Uncover the Secret Hiding Places of Stock Market Profits. New York: Fireside, 1997.
[ii] Tortoriello, Richard. Quantitative Strategies for Achieving Alpha. New York: McGraw-Hill, 1998.