Estimate of hidden activity for the 2020 benchmark revision of national accounts

C. Welter-Medee and Simon Quantin
Additional contact information
C. Welter-Medee: Insee CREST

Documents de Travail de l'Insee - INSEE Working Papers from Institut National de la Statistique et des Etudes Economiques

Abstract: This study proposes an estimate of the unobserved activity of registered firms. The purpose of this estimate is to update the estimates currently used in national accounts for estimating gross domestic product, which covers all activities carried out on the territory, including those that are hidden. The study is based on tax audit management data from the DGFiP. By assimilating the sample of audited companies to survey respondents, the undeclared turnover and omissions or undue declarations of intermediate consumption observed during tax audits can be extrapolated to all non-financial companies and sole proprietorships in the field of national accounting. This extrapolation requires taking into account the process that led to the selection of firms subject to these audits, since it may generate a significant selection bias. For this purpose, we adopt the two-step methodology already implemented in Quantin et Welter-M d e (2022). The first step aims to estimate a weight for each controlled firm and is inspired by methods of nonresponse adjustment by reweighting. In a second step, based on these weights, we extrapolate the undeclared turnover and the omissions or undue declarations of intermediate consumption by ratio estimators applied to groups of firms with homogeneous declarative behavior. We estimate the total amount of undeclared turnover to be 33.1 billion euros and the underestimation of intermediate consumption expenditures to be 1.5 billion euros for 2014. This second amount corresponds to 8.0 billion euros of undeclared intermediate consumption expenses, less 6.5 billion euros of undue declarations of intermediate consumption. In total, this results in an unobserved amount of value added of 31.6 billion euros, slightly less than the undeclared turnover.

Keywords: Machine Learning; Selection Bias; Fiscal Audits; National Accounts (search for similar items in EconPapers)
JEL-codes: C55 C83 E01 H26 (search for similar items in EconPapers)
Date: 2023
New Economics Papers: this item is included in nep-acc
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.bnsp.insee.fr/ark:/12148/bc6p08f1wr6?rk=386268;0 Document de travail de la DESE numero 2023/18 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nse:doctra:2023-18

Access Statistics for this paper

More papers in Documents de Travail de l'Insee - INSEE Working Papers from Institut National de la Statistique et des Etudes Economiques Contact information at EDIRC.
Bibliographic data for series maintained by INSEE ().