Bayesian Evidence on the Structure of Unemployment

This paper presents a Bayesian assessment of the likelihood of unit roots in the unemployment rates of 16 OECD countries. Bayesian techniques for detecting multiple structural breaks in time series have recently been developed by Wang and Zivot (2000). I apply these tests to a data set recently analyzed by Papell et al (2000). I also treat the number of structural breaks as an additional parameter to be estimated. I fin virtually no support for unit root hysteresis in OECD unemployment rates; this result is very robust to the choice of prior.