Observable and unobservable variables in the theory of the equilibrium rate of unemployment, a comparison between France and the United States

This paper confronts, theoretically and empirically, two estimation methods for the Equilibrium Rate of Unemployment (ERU). By introducing observable variables into the TV-NAIRU approach and unobservable variables into the structural approach, we show how these two methods can converge even though their diagnoses differ appreciably in the French case. We considerably improve the econometric and explanatory properties of the French TV-NAIRU model by identifying some of its determinants (interest rates, labour productivity). Moreover, by distinguishing between the concepts of long-term and medium-term ERU, we separate the medium- from the long-term and the observable from the unobservable components of the ERU.