Price change frictions in production plants : learning from a simulation study
MetadataShow full item record
- Master Thesis 
Price stickiness is often taken for granted in modern macroeconomic models, without adequate knowledge of the underlying microeconomic foundations. We want to assess whether the assumptions of price stickiness are consistent with actual pricing patterns. There is a broad consensus in the literature that prices exhibit a pattern of inaction followed by large price changes, so called “zeros and lumps”. A key topic, however, is how to explain the observance of small price changes. This thesis proposes a model specification which sets out to explain small price adjustments, as well as inaction and large price changes. We search for evidence of thresholds and inertia in producer price data. Parameters are estimated using a Simulated Method of Moments (SMM) approach, based on yearly product specific price observations from the Norwegian manufacturing industry. In the simulation model, the adjustment towards the frictionless price is conditional on thresholds and partial adjustments. Price frictions seem to play a major role in explaining how producers change prices, as modeling with friction parameters gives a much better fit than frictionless modeling. Overall, the evidence in this thesis supports assumptions of nominal stickiness. We find evidence of both thresholds and inertia in the price setting, which indicates that prices are affected by different forms of rigidities. Even when we control for inflation, our findings suggest that there are more frictions downwards than upwards. Thus, we can cannot exclude the possibility that it is easier to increase than to decrease prices. An assessment of the literature shows that, in general, macroeconomic models fail to include all the evidence presented in this thesis. While some models assume that firms have pricing thresholds, others assume inertia in the price setting. However, none of the models considered incorporates the combination of both features. Our findings therefore suggest new ways in which macroeconomic models can be improved.