Bayesian Estimation of the Unemployment Gap in the CzechRepublic

Unemployment gap is important indicator that help monetary authority to pursue good economic policy. This indicator is based on differences between observable unemployment rate and and its equilibrium unobservable counterpart. The equilibrium unemployment rate is usually connected to the non-accelerating inflation rate of unemployment (NAIRU) because the issue of monetary stability is incorporated in this theoretical concept. Alternative estimates of the unemployment gap for a small open economy (Czech Republic) are presented and analyzed in this paper. Estimates are made within the framework of the hysteretic approach and non-hysteretic approach. We use techniques of Bayesian analysis (Gibbs sampler and Metropolis-Hastings algorithm) to identify the models. Unobserved states are estimated using Kalman filter. Non-hysteretic model is solved as Dynamic Stochastic General Equilibrium (DSGE) model with rational expectations using the Dynare toolbox. Model estimates indicate hysteretic patterns of the unemployment and imply that actual lowering unemployment rates may be thus sustainable in the long run.