By Jason Hsu and Brad Cornell
The following is a summary of the above-named paper. Click here for the original paper.
Cornell and Hsu assert that the standard consumption-based model has failed to explain the cross-section of expected returns because its assumption that security prices are set by end investors, who wish to maximize their intertemporal consumption, is counterfactual.
They observe that practice differs from theory in that cognitive and information processing constraints cause most end investors to delegate their investment decision making to an agent. The typical end investor does not have the skills or access to needed information to make well-informed investment decisions. To compensate they hire an active money manager to do it for them.
The active fundamental money manager, however, does not use a consumption-based model in the asset selection process, but rather a discounted cash flow model. Therefore, the reality in the investment management ecosystem is that the discount rate used in the valuation model (in contrast to the marginal utility of end investors) drives prices and determines the cross-section of average returns. Once a model is accepted and adopted, a self-fulfilling asset-pricing loop is created: the model used in estimating the discount rate becomes the determinant of expected returns.
The authors’ approach is distinguished from the rationale of behavioral finance models in which prices differ from the standard consumption-based model as a result of end-investor trading mistakes. The premise of Cornell and Hsu is that investors are cognizant of their lack of knowledge and proactively delegate to professional money managers. In delegating, the end investor does not mandate that the manager must evaluate securities based on the investor’s utility maximization problem, and a natural alignment of the investor’s and manager’s interests is not present: managers are incentivized to outperform relative to a stated benchmark or to peers, rather than to operationalize the consumption model of the end investor.
In addition to imperfect contracting, a rather muddled principal–agent relationship results as multiple layers of delegation emerge. The example used to illustrate the complication is pension funds. In the case of corporate pension funds, investment is directed by a chief investment officer appointed by a board of trustees, typically composed of representatives whose interests are biased toward minimizing pension expense. The result often creates a misalignment with the interests of pension fund beneficiaries. In the case of public pension funds, investments are directed by municipal officials who often delegate the investment strategy and asset manager hiring recommendations to consultants, causing even greater divergence between the interests of beneficiaries and sponsor.
Active fundamental managers play the dominant role in setting prices for two reasons. First, passive managers are price takers whereas active managers are not; and second, relative-value managers anchor on price levels influenced by fundamental analysts. Broad published evidence suggests that the majority of fundamental valuation is based on a discounted cash flow analysis. In this context, the cross-section of expected returns is determined by the discount rate used. After all, if a stock’s price equals the discounted present value of its expected future cash flows, its expected return must equal the discount rate. The important question then becomes: How is the discount rate chosen?
The perhaps surprising answer to this question is that we do not know much about this decision. To date, very little research has been done on how active fundamental managers estimate discount rates. An attempt to find the answer gains urgency given the perspective of the authors’ argument that active managers, not end investors, play the dominant role in setting market prices, and in furtherance to that, the critical importance of the role the discount rate model plays in the self-fulfilling prophecy of asset pricing.
Summarized by Kay Jaitly, CFA of Research Affiliates