اسپرد نرخ بهره بعنوان پیش بینی کننده تورم در آلمان و چرخه کسب و کار

We have studied the comparative performance of a number of interest rate spreads as predictors of the German inflation and business cycle in the post-Bretton Woods era. The two-regime Markov-switch model that we used as a nonlinear filter allows the dynamic behavior of the economy to vary between expansions and recessions in terms of duration and volatility. We found that the bank term structure, the public term structure, and the spread based on the call rate predicted all recessions with a comfortable lead, although they lagged some of the recoveries by a few months. The bank–public spread generates a series of false signals, and missed completely the upturn in the mid-1970s, but detected the last two recoveries with an average lead of nearly 12 months. The source of the predictive power of interest rate spreads lies in the information they contain not only about monetary policy, but also about an assortment of general macroeconomic shocks. The filter probabilities from three of the interest rate differentials also foreshadowed the long swings in the German inflation rate remarkably well, with a lead time of 2–4 years without any false signals.