تمرکز مالی در مقابل عدم تمرکز: اثرات رشد و رفاه سرریز، مالیات لویاتان، و تحرک سرمایه
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|27806||2012||12 صفحه PDF||سفارش دهید||12181 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Urban Economics, Volume 71, Issue 2, March 2012, Pages 177–188
This paper develops an endogenous growth model with spillovers of public goods, Leviathan taxation, and mobile capital to examine the relative merits of centralized and decentralized fiscal systems for economic growth and social welfare. We show that a decentralized system dominates a centralized system in terms of economic growth; however, the difference in social welfare between a decentralized and a centralized system is non-monotonic and displays a hump-shaped relationship with respect to capital mobility. Since higher capital mobility induces stronger tax competition, this finding implies that there is an optimal degree of tax competition; some tax competition is desirable, but fierce tax competition may be harmful. We also show that there is a critical level of spillovers of public goods above which centralization dominates decentralization in terms of social welfare, as in previous studies; however, if spillovers are below this critical level, capital mobility also matters in the welfare comparison between centralized and decentralized systems.
This paper develops an endogenous growth model with spillovers of public goods, Leviathan taxation, and mobile capital to examine the relative merits of centralized and decentralized fiscal systems for economic growth and social welfare. Fiscal centralization (FC) internalizes spillovers across jurisdictions and eliminates tax competition for mobile capital via the coordination of fiscal policy, while fiscal decentralization (FD) features tax competition and spillovers due to coordination failures. Put differently, whether or not fiscal policy is coordinated between jurisdictions is the defining characteristic that distinguishes between FC and FD in our study.1 We consider a plausible scenario in which politicians are partly self-interested (i.e., Leviathan or rent seeking) and partly benevolent (i.e., citizen-welfare maximizing). The “weight” that politicians attach to their self-interest could be viewed as reflecting the varying extent of government accountability across different political institutions (Lockwood, 2006). With such political preferences, we develop an endogenous growth model characterized by three exogenous parameters capturing (i) the degree to which politicians are rent seeking, (ii) the degree of spillovers of public goods across jurisdictions, and (iii) the degree of capital mobility. Different economies, due to their histories or for other reasons, are likely to vary in terms of these three parameters. The central purpose of this paper is to provide a framework to analyze the advantages and disadvantages of FD versus FC under different combinations of these three parameters. We examine how economic growth and social welfare are impacted by the choice between FC and FD. Under FC, the central government internalizes spillovers and provides a relatively high level of public goods, but the economy is vulnerable to excessive Leviathan taxation due to the lack of tax competition. Under FD, local governments are constrained in Leviathan taxation due to the presence of tax competition, but they fail to internalize spillovers and may provide an insufficient level of public goods. We show that FD dominates FC in terms of economic growth; however, the social welfare difference between FD and FC is non-monotonic and displays a hump-shaped relationship with respect to capital mobility. Since higher capital mobility induces stronger tax competition, this finding implies that there is an optimal degree of tax competition; some tax competition is desirable, but fierce tax competition may be harmful. We also show that there is a critical level of spillovers of public goods above which centralization dominates decentralization in terms of social welfare, as in previous studies; however, if spillovers are below this critical level, capital mobility also matters in the welfare comparison between centralized and decentralized systems. Edwards and Keen (1996) introduce Brennan and Buchanan’s (1980) idea of taming the Leviathan via tax competition to the standard tax competition literature.2 We follow their paper closely, but with three departures. First, while Edwards and Keen analyze the welfare effects of tax competition versus tax coordination in a static framework, we analyze both the growth and welfare effects in a dynamic framework. As will be shown, pursuing growth and welfare may be in conflict with each other. Second, while Edwards and Keen consider the “local” case where capital is perfectly mobile, we allow for all degrees of capital mobility. We show that higher capital mobility under FD tends to tame the Leviathan government in taxing capital, which promotes economic growth, but at the same time, higher capital mobility could worsen social welfare by reducing the provision of public goods relative to FC. The tradeoff leads to an optimal degree of capital mobility preferred by citizens and a hump-shaped welfare difference between FD and FC with respect to the degree of capital mobility. This non-monotonic hump-shaped relationship is absent in Edwards and Keen (1996) since they only consider the “local” case where capital is perfectly mobile. Third, interactions between governments could come from the expenditure side as well as the revenue side. We address both interactions in our model, whereas Edwards and Keen only address the interaction on the revenue side. It will be shown that leaving spillovers out of the picture in an analysis of tax competition may overweigh the beneficial power of tax competition in taming the Leviathan. In the absence of tax competition and in a static framework, Besley and Coate (2003) adopt a political economy approach to the choice between centralization and decentralization.3 They basically show that there is a critical level of spillovers of public goods above (below) which FC dominates (is dominated by) FD in terms of welfare. This result resembles Oates’s (1972) classic finding regarding the tradeoff between FC and FD, but its derivation is immune to the artificial assumption of policy uniformity across jurisdictions as in Oates. We also follow Besley and Coate (2003) closely and, indeed, a main part of our model is borrowed directly from their paper. However, unlike Besley and Coate (2003) in which an explicit political process is modeled, we summarize the inefficiency of the political process with a parameter capturing the degree to which politicians are rent seeking. Epple and Nechyba (2004, p. 2455) sum up the gist of the recent political approach to FC versus FD as popularized by Besley and Coate (2003): “decentralization becomes less attractive as inter-jurisdictional spillovers increase, and inefficiencies in political systems provide a decentralizing force.” Within our model, the political inefficiency of FC relative to FD arises due to the lack of a discipline device in the form of tax competition in FC. As will be shown, Besley and Coate’s results will be qualified to a significant extent when tax competition is present. A number of studies, such as Devereux and Mansoorian, 1992, Lejour and Verbon, 1997, Razin and Yuen, 1999 and Koethenbuerger and Lockwood, 2010, analyze the growth effect of tax competition versus tax coordination. Brueckner, 1999 and Brueckner, 2006 considers a growth model with overlapping generations of households and shows that economic growth is higher under fiscal federalism.4 However, all of these studies follow the Pigouvian tradition by assuming that governments are benevolent. This approach assumes away the potential role of tax competition in constraining the extravagant public sector and, consequently, FC always dominates FD in terms of welfare since FC internalizes externalities while FD does not.5 By contrast, FD may dominate, or be dominated by, FC in terms of welfare in our framework. Our study follows Rauscher (2005) in exploring the possibility of taming Leviathan governments via tax competition in an endogenous growth model, but we differ from Rauscher (2005) in terms of the modeling details and derived results. A key difference is that we compare the performance of FD versus FC in terms of both economic growth and social welfare while Rauscher’s study is confined to the growth performance within FD. The remainder of the paper is organized as follows. Section 2 introduces our model. Section 3 performs the analysis. Sections 4 and 5 make comparisons between FD and FC in terms of economic growth and social welfare, respectively. Section 6 concludes. Proofs are relegated to the Appendix.