We use micro data on product prices linked to information on the firms that set them to test for selection effects (state dependence) in micro-level producer pricing. In contrast to using synthetic data from a canonical menu-cost model, we find very weak, if any, micro-level selection effects when running price change probability regressions on actual data. Moreover, when fitting a model that nests both time- and state-dependent elements (the CalvoPlus model of Nakamura and Steinsson, 2010) to the data, the resulting parameters mimic the standard Calvo (1983) model. Thus, upstream in the supply chain, price-setting is best characterized as time-dependent.
Second fulltext is an updated version of the first fulltext.