Interregional competition, spillovers and attachment in a federation
Introduction
According to conventional wisdom in fiscal federalism, regional public goods that generate interregional spillovers – that is, goods whose economic jurisdiction exceeds the political jurisdiction of a regional or local government – should be subjected to a Pigouvian subsidy determined by a higher-level government, namely, the central government (see, e.g., Oates, 1972). Local governments should provide local public goods as far as economic and political jurisdictions coincide, but local public goods that generate interregional spillovers should be provided by a supra-local level of government.
Recent studies, however, demonstrate that there are circumstances under which competing regional governments may not only provide public goods that generate transboundary spillovers efficiently, but also implement interregional transfers to each other so that a socially efficient population distribution is obtained.1 In Wellisch (1994), competing regional governments fully internalize externalities associated with provision of public goods in their regions whenever households are not attached to regions. Caplan et al. (2000) demonstrate that decentralized provision of a pure public good – a good whose economic benefit is available for an entire federation – may be efficient for a game where regional governments are policy leaders and a central government is a policy follower. Each regional government decides how much of the pure public good to provide in anticipation of the redistributive income policy implemented by the center. The efficiency result does not depend on the degree of household attachment.
In this paper, we extend the framework advanced by Caplan et al. (2000) in three significant ways. First, we consider commodities that generate regional-specific benefits as well as a federal benefit (good) or cost (bad). Consider, for example, provision of energy. Energy generates private consumption benefits as well as federal damages, since its consumption or production typically yields emissions of pollutants in the atmosphere (e.g., carbon dioxide). Regional governments are concerned with controlling emissions of carbon dioxide in their own regions. We postulate that such a control is done through setting up quotas of pollution permits that can be sold in a federal market for pollution permits. We also assume that the energy commodity is traded within and across regions. Thus, unlike Caplan et al. (2000), we consider an economy featuring tradable commodities and an impure public bad.
The second extension is that we examine the actions of large government authorities which fully recognize that their policies affect prices of all commodities traded in the economy. In our framework, there are three commodities which are traded globally: numeraire and energy goods and pollution permits. The third extension comes from the fact that we depart from the unrealistic assumption used in Caplan et al. (2000) that each household owns an equal share of the resources available in the region in which it resides. We consider instead a more general and realistic setting where two types of households, mobile and immobile, reside in each region. Each region’s immobile population owns the regionally fixed productive resource, which we shall assume to be land. The mobile population in this economy is its population of workers. Landowners earn land rents and workers earn wages. In addition, both types of households possess positive shares of profits produced in both regions, earn income from the sale of pollution permits and receive or pay centrally determined income transfers.
As in Caplan et al. (2000), we are interested in analyzing the allocation of resources for a federal economy characterized by imperfect labor mobility and decentralized leadership. Our example of such a federal economy coincides with theirs, namely, the European Union (EU). The governments of the member nations are endowed with considerable economic and political powers vis-à-vis the center concerning most types of policies, including environmental policies. Accordingly, in our setting, the regional governments select the regional quantities of pollution permits to be supplied in the pollution permit market prior to the center’s decisions of how to redistribute income in the federation.
This paper is organized as follows. Section 2 presents the model and determines the competitive equilibrium given the set of policy instruments. The problems faced by the decentralized agents (consumers and producers) yield the relevant indirect utility and profit functions that are taken into account by the governmental authorities when they make their decisions. Section 3 examines a setting in which the center controls all policy instruments and demonstrates that the implied allocation is socially optimal. Section 4 describes and solves the two-stage federal policy game played by regional and central authorities. We demonstrate that the subgame perfect equilibrium for this game is socially optimal by showing that it is isomorphic to the center’s most preferable allocation. Section 5 concludes the paper.
Section snippets
The model
Consider a federation consisting of two regions, two regional governments and a central government. There are two types of households, workers and owners of a fixed regional resource, which we shall consider to be land. Landowners are immobile. Workers are imperfectly mobile due to regional attachment benefits. Let and denote respectively the populations of immobile and mobile households who reside in region j, j = 1, 2. We assume that , j = 1, 2. Let be the population of
Centralized policy making and the social optimum
We first determine the allocation that the center implements if it controls all policy instruments. This is the center’s most preferred allocation. We later show that this allocation corresponds to the social optimum for the economy.
The center chooses nonnegative to maximize subject to Eq. (1), where and E(·) ≡ E(ρ(Q)). Let μ be the Lagrangian multiplier associated with constraint (1). Assuming an
Decentralized leadership: the federal policy game
Regional and central authorities play a two-stage game as follows. In the first stage, region j chooses to maximize , where , E(·) ≡ E(ρ(Q)) and Q = Q1 + Q2, taking Qk, k ≠ j, as given. In the second stage, the center observes {Q1, Q2} and then chooses to maximize subject to (1), the federation’s redistributive constraint. The equilibrium concept used is subgame
Conclusion
This paper analyzes decentralized regulation of energy supply, an impure public bad. Energy generates positive consumption benefits and pollution damages. Our federal structure is characterized by decentralized leadership and imperfect labor mobility. We show that decentralized environmental policy is efficient if it is combined with an optimal redistribution policy carried out by the center provided that energy and pollution permits are traded within and across regions and the center
Acknowledgments
We greatly benefited from valuable comments made by the editor, Jan Brueckner, and two anonymous referees. We also thank Jun-ichi Itaya, Se-il Mun and the participants of a seminar at Hokkaido University and the Japanese Economic Association Meetings held at Oita University for helpful comments on earlier versions.
References (7)
- et al.
Pure public goods and income redistribution in a federation with decentralized leadership and imperfect labor mobility
Journal of Public Economics
(2000) - et al.
Attachment to home and efficient purchases of population in a fiscal externality economy
Journal of Public Economics
(1993) Optimality, free mobility, and the regional authority in a federation
Journal of Public Economics
(1990)
Cited by (15)
Strategic effects of future environmental policy commitments: Climate change, solar radiation management and correlated air pollutants
2015, Journal of Environmental ManagementSelective decentralized leadership
2014, Journal of Urban EconomicsNon-cooperative pollution control in an inter-jurisdictional setting
2013, Regional Science and Urban EconomicsCitation Excerpt :This timing, referred to as decentralized leadership, is comparable to that initiated by Silva and Caplan (1997) and pursued by Caplan et al. (2000). Its application to decentralized pollution policies is studied in Naoto and Silva (2008) in a model similar to ours, and in Silva and Yamaguchi (2010). We then extend this base case in a number of directions.
Tax competition in the presence of environmental spillovers
2022, International Tax and Public FinanceInterregional negotiations and strategic delegation under government subsidy schemes
2021, Journal of Public Economic Theory