Skip to Main content Skip to Navigation

De la révolution lucasienne aux modèles DSGE : réflexions sur les développements récents de la modélisation macroéconomique

Abstract : This dissertation provides a history of macroeconomic modeling practices from RobertE. Lucas’s works in the 1970s up to today’s dynamic stochastic general equilibrium (DSGE) approach. Working from a historical perspective, I suggest that the recent rise of DSGE models should be characterized as a compromise between opposing views of modeling methodology—on the one hand, the real business cycle (RBC) view, on the other hand, the new Keynesian view. In order to justify this claim, my work provides an epistemological reconstruction of the recent history of macroeconomics, building from ananalysis of the criteria defining the validity and the pertinence of a model. My assumption is that recent macroeconomic modeling practices can be described by three distinctive methodological criteria : the internal validity criterion (which establishes the consistency between models’ assumptions and concepts and formalisms of a theory), the external validity criterion (which establishes the consistency between the assumptions and results of a model and the real world, as well as the quantitative methods needed to assess such a consistency) and the hierarchization criterion (which establishes the preference for internal over external validity, or vice versa). This epistemological reconstruction draws primarily from the literature about models in the philosophy of science. My work aims to make four contributions to the history of recent macroeconomics. (1) To understand the rise of DSGE models without referring to the explanation providedby the macroeconomists themselves, who tend to think that macroeconomics evolved through theoretical consensus and exogenous technical progress. By distancing itself fromthis perspective, my work draws attention to the disruptive character of methodological controversies and to the interdependence between theoretical activity and the developmentof statistical and econometric methods. (2) To overcome the existing divide betweenthe history of macroeconomic theories and the history of quantitative methods. Throughits epistemological perspective, my work reconciles these two historiographies and specifiesthe basis for a comprehensive understanding of recent developments in macroeconomics.(3) To put the accent on the external validity condition as the main controversial issue separating different views of macro-modeling methodology. Furthermore, I illustrate how the debate about external validity is closely related to the problem of casual explanation and, finally, to the conditions for providing economic policy evaluation. (4) To characterize the DSGE approach: although DSGE models are often presented as a“synthesis”, or as a “consensus”, they are better described as a shaky compromise between two opposing methodological visions.
Document type :
Complete list of metadata
Contributor : Abes Star :  Contact
Submitted on : Thursday, May 17, 2018 - 10:32:06 AM
Last modification on : Tuesday, January 19, 2021 - 11:08:37 AM
Long-term archiving on: : Monday, September 24, 2018 - 8:42:35 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01793960, version 1



Francesco Sergi. De la révolution lucasienne aux modèles DSGE : réflexions sur les développements récents de la modélisation macroéconomique. Economies et finances. Université Panthéon-Sorbonne - Paris I, 2017. Français. ⟨NNT : 2017PA01E059⟩. ⟨tel-01793960⟩



Record views


Files downloads