The riddle of the natural rate of interest

We provide a general equilibrium model with optimizing agents to compute the natural rate of interest for the G7 countries over the period 2000 to 2017. The model is solved for the equilibrium natural rate of interest, which is determined by a parsimonious equation that is easily computed from raw observable data. The model predicts that the natural rate depends positively on the consumption - leisure growth rates gap, and negatively on the capital -labor growth rates gap. Given our computed natural rate, the short-term nominal interest rates in the G7 have been higher than the natural rate since 2000, except for Germany and the U.S. during the period 2009-2017. In addition, the data do not support the prediction of the Wicksellian theory that prices tend to increase when the short-term nominal rate is lower than the natural rate. Projections of the natural rate over the period 2018 to 2024 are positive in Germany, Italy, Japan, and the U.K. and negative in Canada, France, and the U.S. The model predicts that fiscal expansion is an expensive policy to achieve a 2 percent inflation target when the Zero Lower Bound (ZLB) constraint is binding.