Testing New Versions of the Wage Phillips Curve in the MeMo-It Model Used by Istat

• Activity Number: 439 - Topics in Marketing
• Type: Contributed
• Date/Time: Thursday, August 12, 2021 : 4:00 PM to 5:50 PM
• Sponsor: Social Statistics Section
• Author(s): Deborah Scaccabarozzi* and Daniele Toninelli and Davide Zurlo and Fabio Bacchini and Roberto Iannaccone
• Companies: University of Bergamo and University of Bergamo and Istat and Istat and Istat
• Keywords: Phillips curve; error correction model; macroeconometric model

Abstract

The Phillips curve is a research object of interest from several years. The literature proposed a series of alternative formulations: the original Phillips curve, the expectations-augmented Phillips curve, the New Classical Phillips curve, the New Keynesian Phillips curve, the Hybrid Phillips curve. In this work we test a new version of the wage Phillips curve evaluating its goodness of fit as well as its forecasting performance for the Italian context. This version includes an error correction term representing the theoretical cointegrating relationship between wage, prices and productivity. This new specification is also applied to the Italian data ranging from 1980 to 2019. The proposed specification of the Phillips curve has been tested both in terms of its goodness of fit and in terms of forecasting ability inside the macro-econometric model for Italian economy (MeMo-It) developed by the Italian National Institute of Statistics. Our findings confirm a good fit of the model, highlighting a similar ability of the new specification of the Phillips curve equation in explaining wage growth. Nevertheless, the results in term of forecasting performance show a relevant improvement for all the considered measures, in comparison to what obtained with the original curve specification included in MeMo-It model.