Time- or state-dependent price setting rules? Evidence from micro data

We use two rich micro-datasets on Portuguese firms to analyse the ability of time- and state-dependent price setting rules to explain durations of price spells, or the probability of price changes. Using a duration model with time-varying regressors, we find some evidence of state-dependent price setting behaviour, which suggests that time-dependent models are unable to fully describe the features of the data. Specifically, we find statistically significant impacts on the probability of a price change of inflation, the level of economic activity and the magnitude of the last price change. Besides being statistically significant, in some cases these effects are also economically important. Finally, it is found that negative and positive values of the covariates have different impacts on the expected duration of prices.