Fiscal Decentralization and Preference Heterogeneity

18 Pages Posted: 1 Nov 2019

See all articles by Takeshi Miyazaki

Takeshi Miyazaki

Kyushu University, School of Economics

Date Written: October 23, 2019


Oates' Decentralization Theorem is the basis of the theory of fiscal federalism. This study reconsiders the theorem with a focus on the role of preference heterogeneity. The Decentralization Theorem predicts that the larger (smaller) the heterogeneity in preferences for local public goods across regions, the larger (smaller) the benefits of decentralization. Contrary to this conventional view, the present study shows that larger degrees of preference heterogeneity may increase the benefits of centralization. First, if the difference in preferences is small and/or spillovers in local public goods are large, then a rise in the preference of a region with higher taste, among two regions with different tastes, is more likely to increase the welfare differential between centralization and decentralization. Thus, larger preference difference may make a centralized regime preferable to a decentralized regime in terms of welfare. In contrast, a rise in the preference of a region with lower taste increases the welfare differential. Second, the preference of a region with higher preference can have a positive effect on the welfare differential if social welfare under a centralized regime is superior to that under a decentralized regime. This means that the positive relationship between preference heterogeneity and the welfare differential does not bring about a regime switch from decentralization to centralization.

Keywords: Decentralization Theorem, fi scal federalism, preference heterogeneity

JEL Classification: H10, H70

Suggested Citation

Miyazaki, Takeshi, Fiscal Decentralization and Preference Heterogeneity (October 23, 2019). Available at SSRN: or

Takeshi Miyazaki (Contact Author)

Kyushu University, School of Economics ( email )

744 Motokoka Nishi-ku
Fukuoka, 8190395

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

Paper statistics

Abstract Views
PlumX Metrics