For evaluating risky investments, understanding the term structure of discount rates is critical. Discount rates account for the riskiness of an investment, the time value of the money invested, and the opportunity cost of not investing in something else with those funds.
Smith finance professor Serhiy Kozak, with co-authors Stefano Giglio and Bryan Kelly from Yale School of Management, have created a new asset pricing model for investors. It can price claims to individual stock’s dividends (so-called “dividend strips”), laying out realistic term structures for discount rates for different equity portfolios. The model — built from 45 years of data, including several recessions — takes into account equity prices, dividends, returns and their dynamics without relying on data from dividend strips.
Kozak and his co-authors account for the dynamics of the economy and include recent research findings on return predictability in their model, published in the Journal of Finance.
“When you look at the returns of dividend claims with a specific maturity date, that explains why investors charge premiums to hold risky funds of different maturities,” says Kozak.
They use the model to generate a term structure of discount rates not only for aggregate cash flows in the S&P 500, but also for many other equity portfolios. They validate their model by looking at 102 portfolios, starting from the 1970s, to confirm their predictions. They evaluate their model for the discount rates of risky cash flows by comparing the implied dividend strips for their model to the prices of actually traded dividend strips.
“We show that our model—estimated using no dividend strip data at all—well matches the prices of traded dividend strips on the S&P 500 of maturities of one, two, five, and seven years, observed since 2004, along a variety of dimensions,” write the researchers.
“Most importantly, the model generates interesting differences in the average term structures across portfolios,” the researchers write. “These results give us new moments that can be used to evaluate structural asset pricing models that have direct implications about the risk premia of these portfolios (as well as any of the other 102 portfolios that we include in our analysis).”
The research, “Equity Term Structures without Dividend Strips Data,” is published in the Journal of Finance.
Media Contact
Greg Muraski
Media Relations Manager
301-405-5283
301-892-0973 Mobile
gmuraski@umd.edu
Get Smith Brain Trust Delivered To Your Inbox Every Week
Business moves fast in the 21st century. Stay one step ahead with bite-sized business insights from the Smith School's world-class faculty.