GMM estimation of the new Phillips curve

This paper applies test statistics developed by Stock and Wright [Econometrica 68 (2000) 1055] to examine the two specifications of the New Keynesian Phillips curve in Galí and Gertler [Journal of Monetary Economics 44 (1999) 195]. The new test statistics demonstrate the inadequacy of conventional GMM methodology by indicating two features that are neglected by Galí and Gertler: observational equivalence in the baseline model and weak identification in the hybrid model.