Monetary policy and structural change in the United States.

Summary This thesis first reviews the econometric theory of unit root tests allowing or not the possibility of one or more breaks. These tests are then applied to a set of US macroeconomic series. We then extend the analysis to the multivariate VAR case, in order to examine the stability or otherwise of the propagation mechanisms of monetary impulses. We will then consider the possibility of multiple breaks, at dates unknown a priori. The relevance of this extension will be examined in the light of the analysis of American monetary policy since the beginning of the 1960s. First, we consider two structural models, in which we identify a Taylor rule. In the first model we use the output gap, the federal funds rate and the current inflation rate as endogenous variables. In the second model we use the output gap, the federal funds rate and the expected inflation rate as endogenous variables. This should allow us, on the one hand, to contribute to the evaluation of the effects of a change in monetary policy on the US output gap and the two inflation rates, and on the other hand, to compare the effectiveness of US monetary policy between different periods. In a second step, we consider the same models but this time we assume three shocks are estimated simultaneously, a demand shock, a supply shock and a monetary shock. Our aim is to identify the sources of fluctuations in the variables in question.