Decentralization and Progressive Taxation

31 Pages Posted: 28 Jan 2021

See all articles by Simon Berset

Simon Berset

University of Fribourg - Department of Economics

Mark Schelker

University of Fribourg; CREMA; CESifo (Center for Economic Studies and Ifo Institute); SIAW, University of St. Gallen

Date Written: 2021


The traditional normative literature on fiscal federalism argues that redistributive policies should be centralized in order to avoid welfare- or tax-induced migration. However, recent evidence shows that even in a setup where the progressivity of the income tax schedule is centralized to an upper-layer government and local governments are involved in tax competition with only a tax shifter, local mobility induces income sorting. Hence, despite centralized redistributive taxation, the resulting effective tax schedule is less progressive than what is set in the tax code. We argue that upper-layer governments anticipate the impact of local income sorting and strategically adjust their statutory tax schedules. We analyze Swiss panel data and apply causal machine learning methods to identify the effects of decentralization on the statutory tax structure. We provide evidence that more decentralized cantons reduce the tax burden for lower and intermediate income classes and hence implement more redistributive statutory tax schedules. This strategic adjustment is limited by the mobility of the tax base.

JEL Classification: H730, H770, H710

Suggested Citation

Berset, Simon and Schelker, Mark, Decentralization and Progressive Taxation (2021). CESifo Working Paper No. 8862, Available at SSRN:

Simon Berset (Contact Author)

University of Fribourg - Department of Economics ( email )


Mark Schelker

University of Fribourg ( email )


CREMA ( email )

Gellertstrasse 18
Zurich, CH 8006

CESifo (Center for Economic Studies and Ifo Institute) ( email )

Poschinger Str. 5
Munich, DE-81679

SIAW, University of St. Gallen ( email )

Bodanstrasse 8
Saint Gallen, St. Gallen CH-9000

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics