Spread-Driven Dividend Discount Models

A key determinant of shareholder value is the franchise spread—the company's incremental return on new investments over the cost of capital. Explicitly incorporating this spread into the valuation process paves the way for a more compact, two-parameter formulation of the standard three-parameter dividend discount model. This transformation leads to a number of interesting implications. In particular, the spread-driven representation of the DDM (1) clarifies the role of growth-driven ROEs versus the role of spread-driven ROEs, (2) facilitates the development of two-phase models that reflect a typical company's earnings pattern, (3) shows how earnings growth and franchise spreads can underpin a wide range of P/E levels, (4) addresses the problem of artificially high P/Es being forced by low estimates for the risk premium and/or the inflation rate, (5) provides a useful expression for the growth rate of shareholder value, and (6) under certain stability conditions, leads to a pro forma equity duration that is—surprisingly—equal to the P/E itself. Three parameters—the discount rate, earnings growth, and the payout fraction—are explicit in the standard one-phase dividend discount model (DDM). A fourth parameter, however, plays a crucial role in any projection of shareholder value. This fourth parameter, “the franchise spread,” is the company's incremental return on new investments above and beyond its cost of capital. Without a positive franchise spread, growth adds nothing to shareholder value. Consequently, this fourth parameter may be the most significant factor in company valuation. The franchise spread never appears explicitly in standard DDM formulations, but it is always embedded somewhere in any DDM structure. Many valuation effects can be better understood by exploring the implied movements in this key parameter.The franchise spread can be readily incorporated into the standard dividend discount model to create a spread-driven DDM. This form may be used as a single-phase, stand-alone model or as the terminal stage of a multiphase model. Moreover, by making a clear distinction between growth-driven returns and spread-driven returns, this framework sheds new light on a number of current topics in valuation theory.For example, various studies have argued that P/Es far above historical norms are possible today because of the lower market discount rates that result from more controlled inflation and from reduced equity risk premiums. The standard one-phase DDM is often invoked as the analytical basis for such arguments. Outsized P/Es are generally computed by reducing the discount rate while simultaneously maintaining constant values for earnings growth and for the payout fraction.With the growth rate fixed in the standard DDM, however, a falling discount rate will automatically create a point-for-point increase in the embedded franchise spread, and this increased spread is what accounts for a large part of the P/E escalation. Because this effect is never made explicit, the DDM analyst may not be aware that the lower discount rates are automatically raising the franchise spread. Not only is this inverse “spread-discount rate” relationship questionable in itself, but it can also easily take the franchise spread to unreasonably high levels, especially within the long-term context of the single-phase DDM.When a more stable franchise spread is postulated, the DDM's sensitivity to discount rate changes (i.e., the equity duration) is greatly reduced. With spread stability, lower market rates lead to more moderate (and, some would say, much more reasonable) P/Es. Thus, even in the face of assumed reductions in the risk premium, the inflation rate, or both, franchise-spread considerations tend to undermine DDM-based arguments justifying stratospheric P/Es for the market as a whole.The concept of the franchise spread also has a number of useful byproducts in terms of the DDM model itself. First, it leads to a more comprehensive gauge of the company's growth in shareholder value. Moreover, with a stable franchise spread, this “value-growth rate” can be made independent of market discount rates. Second, using the value-growth rate, one can transform the standard three-parameter DDM into a compact, two-parameter spread-driven DDM. Third, the revised two-parameter DDM leads to a pro forma mathematical equity duration that—surprisingly—exactly coincides with the P/E itself!Finally, the most far-reaching outcome of the franchise-spread concept may be the cleaner distinction it makes between “spread-driven returns” that can be reasonably sustained over a long terminal phase and the supercharged “growth-driven returns” that high-P/E companies can enjoy over a limited front-end phase. This distinction readily leads, in turn, to multiphase models consisting of a growth-driven first phase segueing into a longer-term spread-driven phase. In particular, using a two-phase model of this “dual-driver” format, one can easily account for the high P/Es of growth stocks while obtaining a more palatable P/E response to lower discount rates. Thus, by explicitly incorporating the concept of the franchise spread and its associated value-growth rate, one can construct valuation models that are actually more intuitive in terms of their basic structure than the standard DDM, more readily specified in terms of input estimation, and more reasonable in terms of projected P/E patterns over a range of market scenarios.