Estimation and assessment of measures of the natural rate of interest: Evidence from Latin American economies with inflation targeting

We estimate and assess two measures of the natural rate of interest (NRI) for two Latin American economies, Chile and Peru, where monetary policy is conducted under inflation targeting. The first NRI measure is obtained through the estimation of a time varying parameter vector autoregression model with stochastic volatility (TVP-VAR-SV) as in Lubik y Matthes (2015), which we denote TVP-NRI. The second NRI measure is based on a recent methodology proposed by Benati (2020) and Benati (2023), which exploits the relationship between the interest rate and money velocity (Benati-NRI). In order to assess these two measures of NRI, we propose a new and simple criterion based on the idea that NRI is not expected to react to shocks that have no long-run effect on real interest rate (i.e. transitory shocks). The results for Chile and Peru indicate that TVP-NRI measures are relatively superior.