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\nthe Equilibrium Rate of Unemployment (ERU). By introducing observable variables\ninto the TV-NAIRU approach and unobservable variables into the structural\napproach, we show how these two methods can converge even though their\ndiagnoses differ appreciably in the French case. We considerably improve the\neconometric and explanatory properties of the French TV-NAIRU model by\nidentifying some of its determinants (interest rates, labour productivity). Moreover,\nby distinguishing between the concepts of long-term and medium-term ERU, we\nseparate the medium- from the long-term and the observable from the unobservable\ncomponents of the ERU.