Prevailing wisdom holds that the yield curve embodies all the data needed to predict bond yields, as it reflects market expectations of Federal Reserve interest rate moves.
But what if the yield curve provides only part of the picture? “The conventional framework for forecasting yields is in isolation of a larger economic story,” contends Stanford Graduate School of Business professor Kenneth Singleton.
That is why he and a few other economists contend that yield forecasters need to move beyond a narrow focus on the bond market. “We should also look at macro factors if we want to know where bond yields are headed,” Singleton asserts.
That sounds obvious, almost simplistic, but it is easier said than done. The mathematical modeling and the calculations involved are complex. Yet Singleton, working with Scott Joslin of MIT and Marcel Priebsch of Stanford, has created an economic model that purports to incorporate macro data to better forecast yields. It’s available on Singleton’s web page (www.stanford.edu/~kenneths/) under the off-putting title “Risk Premiums in Dynamic Term Structure Models with Unspanned Macro Risks.”
Singleton believes his approach will be useful to policy analysts, risk managers, medium-term investors and yield prognosticators, though less so to high-frequency traders because the macro data isn’t updated rapidly enough.
