L'histoire (faussement) naïve des modèles DSGE

Abstract : The purpose of the article is to analyze and criticize the way how DSGE macroeconomists working in policy-making institutions think about the history of their own modeling practice. Our contribution is, first of all, historiographical: it investigates an original literature, emphasizing in the history of DSGE as it is told by its own practitioners. The results of this analysis is what we will call a “naïve history” of DSGE modeling. Modellers working from this perspective present their models as the achievement of a “scientific progress”, which is linear and cumulative both in macroeconomic theorizing and in the application of formalized methods and econometric techniques to the theory. This article also proposes a critical perspective about the naïve history of the DSGE models, which drawns, by contrast, the main lines of an alternative, “non-naïve” history. of the DSGE models is incomplete and imprecise. It mainly ignores controversies, failures and blind alleys in previous research; as a consequence, the major theoretical and empirical turning points are made invisible. The naïve history also provides an ahistorical account of assessment criteria for modeling (especially for evaluating empirical consistency), which hides the underlying methodological and epistemological debates. Finally, we will claim that the naïve history plays an active and rhetoric role in legitimizing the DSGE models as a dominant tool for policy expertise.
Document type :
Other publications
Complete list of metadatas

Cited literature [116 references]  Display  Hide  Download

https://halshs.archives-ouvertes.fr/halshs-01222798
Contributor : Lucie Label <>
Submitted on : Friday, October 30, 2015 - 3:53:52 PM
Last modification on : Tuesday, January 30, 2018 - 5:50:04 PM
Long-term archiving on : Friday, April 28, 2017 - 7:25:06 AM

File

15066.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : halshs-01222798, version 1

Collections

Citation

Francesco Sergi. L'histoire (faussement) naïve des modèles DSGE. 2015. ⟨halshs-01222798⟩

Share

Metrics

Record views

637

Files downloads

1998