Accounting for Post-Crisis Inflation: A Retro Analysis

Why was there no deflation and what accounts for inflation after 2008? Is the missing deflation puzzle an indictment of Phillips-curve-type analysis, or is all well and good if one fixes the model with the appropriate features? To shed light on this issue and in order to avoid confounding the answer with post-crisis model adaptations, we provide a "retro analysis" and employ the original benchmark of Smets and Wouters (2007) model. We show that this model implies that shocks to price and wage markups alone are nearly enough to account for inflation before 2008, and that they do so substantially post 2007 as well. While markup shocks account for most of the inflation movements, they do not play a significant role in the accounting of employment: the Phillips curve tradeoff is weak after 2007 according to the model, but also before 2007. We thus argue that the asserted post-crisis model failures were visible pre-crisis already. Extending the retro analysis with features introduced by the most recent part of the literature does little to alter our key insights. (Copyright: Elsevier)