Towards data-congruent models of the term structure of interest rates

Bond yields can be decomposed into two unobservable components: the expected sequence of short-term rates and term premia. The identification of these two components is crucial to understand bond pricing and the effect of monetary policy on the term structure of interest rates. This paper illustrates how M.H. Pesaran's prescription of congruency between the data and the reduced form, explicitly derived from stochastic dynamic optimization, effectively facilitates the relevant decomposition. By examining the historical evolution of term structure models, we demonstrate that the chosen specifications have not consistently aligned with the data, presenting a missed opportunity. In fact, a data-congruent specification helps in improving forecasts of the dynamics of US short-term rates and generates stationary dynamics for the term premia.