Evaluating difference-in-differences models under different treatment assignment mechanism and in the presence of spillover effects

Guilherme Lima, Igor Souza and Mauro Ferreira
Additional contact information
Guilherme Lima: UFMG
Igor Souza: UFMG
Mauro Ferreira: UFMG

No 662, Textos para Discussão Cedeplar-UFMG from Cedeplar, Universidade Federal de Minas Gerais

Abstract: We conduct Monte Carlo experiments to evaluate the performance of different Difference-in-Differences estimators under treatment assignment mechanisms affected by shocks suffered by treated units and also in contexts where the treatment effect spills over to units in the control group. In particular, we compare the estimators proposed by Callaway and Sant'Anna (2021), Borusyak et al. (2021), and Sun and Abraham (2021), as well as the two-way fixed effects (TWFE) estimator. The results demonstrate that the treatment assignment mechanisms we design, and the presence of spillover effects can severely compromise the performance of the considered estimators, leading to bias and, even more importantly, inconsistency. Therefore, cautious for interpreting the results should be taken in applications where the environment studied resembles those we consider. The development of more robust estimators is a necessity and a prosperous research venue.

Keywords: Difference-in-Differences; Causal Inference; Treatment assignment mechanisms; Spillover effects. (search for similar items in EconPapers)
Pages: 35 pages
Date: 2023-10
New Economics Papers: this item is included in nep-ecm and nep-exp
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.cedeplar.ufmg.br/pesquisas/td/TD%20662.pdf (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:cdp:texdis:td662

Ordering information: This working paper can be ordered from
Cedeplar-FACE-UFMG Av. Antonio Carlos, 6627 Belo Horizonte, MG 31270-901 Brazil

Access Statistics for this paper

More papers in Textos para Discussão Cedeplar-UFMG from Cedeplar, Universidade Federal de Minas Gerais Cedeplar-FACE-UFMG Av. Antonio Carlos, 6627 Belo Horizonte, MG 31270-901 Brazil. Contact information at EDIRC.
Bibliographic data for series maintained by Gustavo Britto ().