The New Keynesian Phillips Curve Model
Recall the definition in Chapter 7: the NPCM states that inflation is explained by expected inflation one period ahead E(Apt+i | It), and excess demand or marginal costs xt (e. g. the output gap, the unemployment rate, or the wage share in logs):
Apt = bpi E(Apt+i | It) + bp2’xt. (8.12)
The ‘hybrid’ NPCM, which heuristically assumes the existence of both forward – and backward-looking agents and obtains if a subset of firms has a backward-looking rule to set prices, nests (8.12) as a special case. This amounts to the specification
Apt = bp1E(Apt+i | It) + bpiApt-i + bp^xt. (8.13)
Our analysis in Chapter 7 leads to a rejection of the NPCM as an empirical model of inflation for the Euro area and we conclude that the profession should not accept the NPCM too readily. Still, the model maintains a dominant position in modern monetary economics and it is widely used in analyses of Euro-area data.
With reference to the original contributions by Gall and Gertler (1999) and Gall et al. (2001), Smets and Wouters (2003) estimate a New Keynesian Phillips curve as part of a stochastic dynamic general equilibrium model for the Euro area. The inflation equation is estimated as part of a simultaneous system with nine endogenous variables in a Bayesian framework using Markov-chain Monte Carlo methods, and the authors find parameter estimates which are in line with Gall et al. (2001) for a hybrid version of the New Keynesian Phillips curve (with weights 0.72 and 0.28 on forward and lagged inflation, respectively).
Also, Coenen and Wieland (2002) investigate whether the observed inflation dynamics in the Euro area (as well as in the United States and Japan) are consistent with microfoundations in the form of staggered nominal contracts and rational expectations. On Euro-area data, they find that the fixed period staggered contract model of Taylor outperforms the New Keynesian Phillips curve specification based on Calvo-style random duration contracts and they claim support for the hypothesis of rational expectations.