Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/230675 
Autor:innen: 
Erscheinungsjahr: 
2020
Schriftenreihe/Nr.: 
arqus Discussion Paper No. 243
Versionsangabe: 
Revised and renamed August 2020
Verlag: 
Arbeitskreis Quantitative Steuerlehre (arqus), Berlin
Zusammenfassung: 
Despite the growing literature on the effectiveness of research and development (R&D) tax incentives, little is known about the differing design aspects of the underlying tax policies. In this paper, I apply meta-regression analysis (MRA) to separate the distinct provisions through which various tax schemes affect firms' R&D expenditures. Using 192 estimates from 19 studies exploiting the direct approach, the results indicate, on average, greater input additionality effects of hybrid regimes in comparison to volume-based and incremental ones. MetaForest, a novel machine learning algorithm, confirms these results: the moderator for hybrid schemes is the most important variable in explaining the heterogeneity among estimates. Unlike previous MRA, I find only weak evidence for publication bias in this stream of literature. Overall, the relation between tax incentives and R&D expenditures is positive, on average, but the strength varies with methodological variations across studies.
Schlagwörter: 
R&D
tax incentives
additionality effects
direct approach
meta-regression analysis
random forest
Dokumentart: 
Working Paper

Datei(en):
Datei
Größe
500.55 kB





Publikationen in EconStor sind urheberrechtlich geschützt.