A Note on the Stability of the Swedish Philips Curve

We use Bayesian techniques to estimate bivariate VAR models for Swedish unemployment rate and inflation. Employing quarterly data from 1995Q1 to 2017Q3 and new tools for model selection, we compare a model with time-varying parameters and stochastic volatility to a specification with constant parameters and covar-iance matrix. We find strong evidence in favour of the specification with time-varying parameters and sto-chastic volatility. Our results indicate that the Swedish Phillips curve has not been stable over time. However, our findings do not suggest that the Phillips curve has been flatter in more recent years.