In case anyone is interested, I put together a lot of stuff I have been doing with inflation expectations into a soft of manuscript here (warning 19 poorly written pages)
Inflation expectations are not anchored. A simple regression model fits both the median Livingston Survey respondent’s expected CPI inflation and five year TIPS break evens quite well, and fits 20 year breakevens fairly well. There is no sign of some widely alleged credibility effects in the Livingston forecasts. The Livingston forecasts add some useful information not in lagged inflation, but this useful role is strongly concentrated in the period of the oil shocks. They are very poor forecasts. A simple regression model estimated using data from June 1973 and before (so before the oil shocks) fits inflation post December 1989 better than the median Livingston forecast.