Structural Inflation Models with Real Wage Rigidities: The Case of Canada

Bank of Canada working papers are theoretical or empirical works-in-progress on subjects in economics and finance. The views expressed in this paper are those of the authors. No responsibility for them should be attributed to the Bank of Canada. ii Acknowledgements We would like to thank Jordi Galí and Sharon Kozicki for helpful suggestions and comments. Abstract Real wage rigidities have recently been proposed as a way of building intrinsic persistence in inflation within the context of New Keynesian Phillips Curves. Using two recent illustrative structural models, we evaluate empirically the importance of real wage rigidities in the data and the extent to which such models provide useful information regarding price stickiness. Structural estimation and testing is carried out using Canadian data and identification-robust methods. Results based on one of the models are relatively uninformative. Our tests reveal important identification difficulties and considerable estimate uncertainty, as can be seen from the wide projections for the estimates. However, we obtain economically reasonable ranges for estimates of average frequency of price changes and some evidence for rigidity in real wages (as measured by a rigidity index) based on the other model we examine. In addition, our specification for the latter model yields significant [at usual levels] and correctly-signed reduced-form coefficient estimates, showing a trade-off between unemployment and inflation in the New Keynesian Phillips curve. From a methodological perspective, these results derive from our treatment of the productivity term as observable although with error, which seems to capture vital information and improve overall identification. From a substantive perspective, our findings suggest that wage-rigidity based New Keynesian Phillips Curves hold promise empirically and provide interesting research directions. Résumé Des études récentes proposent que l'on introduise des rigidités des salaires réels dans les modèles fondés sur la nouvelle courbe de Phillips keynésienne pour générer une persistance intrinsèque dans la dynamique de l'inflation. En prenant pour illustration deux modèles structurels récents, les auteurs évaluent empiriquement l'importance de la rigidité des salaires réels ainsi que la mesure dans laquelle ces modèles fournissent de l'information utile sur la rigidité des prix. Pour réaliser l'estimation structurelle des modèles et les tests, les auteurs ont recours à des données canadiennes et appliquent des méthodes qui permettent de surmonter, s'il y a lieu, les problèmes d'identification. Les résultats du premier modèle donnent relativement peu d'information. Les tests révèlent d'importantes difficultés d'identification et un degré d'incertitude très élevé des coefficients estimés, qui …