This PhD thesis falls within the fields of public finance and public choice. It studies the following issues: ideological polarization, decentralization and conflict. It consists of three chapters can be read independently.
In the first chapter the focus is on the importance of ideological polarization between but also within political parties on the level and composition of cantonal public spending. The analysis is on Swiss cantonal parliaments and is based on the use and econometric treatment of a very rich database. We made two contributions to the literature on the impact of political fragmentation on public spending. First, we showed that ideological dispersion within political parties is associated with higher public spending while the dispersion between political parties is associated with less public spending. This finding implies that the mode of organization and discipline within political parties are both important. This result is particularly interesting given that previous literature has completely abstracted from ideological dispersion within political parties. Lastly, we showed that ideological dispersion between and within political parties is mainly relevant for current expenditures.
The second chapter focuses on the conflict of objectives that may exist between economic agents in a two-stage game with asymmetric payoffs. The literature showed that coordination failures are frequent in one-stage games and that communication can help individuals coordinate with the use of take turn strategies that allow for a reduction in the payoffs’ inequality in the long run. Our contribution was to study whether communication has the same capacity in a two-stage game where the subgame perfect Nash equilibrium requires both players to make opposite choices at both stages of the game and accept unequal payoffs. Our results show that coordination failures occur half of the time and that 18% of the pairs use the take turn strategy. Communication significantly increases coordination on the subgame perfect Nash equilibrium because it increases the ability of subjects to initiate a take turn strategy and to maintain it. Thus, communication allows subjects to establish a long-term strategy that increases efficiency and reduces payments differences, induced by the asymmetry in payoffs, by exchanging their relative positions in a fair way. Our results show that communication can solve coordination conflicts, even in more complex situations than in one-stage games usually studied until now.
The last chapter focuses on a particular type of conflict: political secessions. Specifically, it addresses the link between secessions, decentralization and group identity. We aim at determining, by means of a laboratory experiment, if decentralizing the provision of public goods reduces the likelihood of secession. We emphasize the importance of local identity. Our experimental design, obviously, do not reproduce all the trade-offs that separatist movements face in the real world. It allows, however, capturing at least two key forces that drive these movements: the proximity to public goods that reduces the phenomena of free-riding and brings public goods closer to citizens’ preferences and the importance of local identities. Our results indicate that the Decentralization treatment increases contributions to local public goods and decreases the likelihood of secessions also when identity is made more salient. If fact, reinforcing local group identity increases the likelihood of secession only when local group members are cooperators.
This paper uses a contract theory framework to analyze the mechanisms of eurozone financial governance, with a focus on centralization vs. decentralization and incentive problems. By constructing a Stackelberg game model with n Ministries of Finance as the first movers and the European Central Bank as the second mover, we show that each government can create growth in its own country (self-benefit) by increasing government spending, but that this will increase inflation, resulting in a decrease in the value of the euro. As these effects are shared equally by eurozone countries (cost sharing), an incentive to free-ride at the expense of other countries is present. We then analyze a penalty-based solution to the free-rider problem and derive a second-best solution where a commitment not to renegotiate penalties ex-post is impossible. The optimal solution shows that ¡°limited sovereignty, ¡± that is, substantially constrained fiscal sovereignty, should be imposed as a high marginal cost for the issuance of public debt. Finally, we close the paper by discussing the possibility of Fiscal Integration (Fiscal Union).
Ali Kakhbod, Joseph C. Koo, Demosthenis Teneketzis
We present a decentralized message exchange process (tatonnement process) for determining the level at which a certain public good will be provided to a set of individuals who finance the cost of attaining that level. The message exchange process we propose requires minimal coordination overhead and converges to the optimal solution of the corresponding centralized problem.
A family of core extensions for cooperative TU-games is introduced. These solution concepts are non-empty when applied to non-balanced games yet coincide with the core whenever the core is non-empty. The extensions suggest how an exogenous regulator can sustain a stable and efficient outcome, financing a subsidy via individual taxes. Economic and geometric properties of the solution concepts are studied. When taxes are proportional, the proportional prenucleolus is proposed as a single-valued selection device. An application of these concepts to the decentralization of a public goods economy is discussed.
River floods can be very damaging. Since numerous human responses to floods are possible, the question arises: What is the organization of flood management that leads to the most efficient results? The dissertation first considers flooding and flood management in Germany. Based on that analysis, a theoretical, political-economy model of public flood defense is developed. The basic model is then extended to allow migration between upstream and downstream regions as well as between flood-prone areas and zones without risk. As different flood protection measures have quite different spatial characteristics, the distribution of responsibilities between the different actors is decisive for the overall performance of flood management. The distribution of public responsibilities among the different federal actors is crucial for two reasons. First, there are unidirectional upstream-downstream spillovers that raise the question whether decentralized or centralized provision of public goods is more efficient. Second, benefits from flood protection are concentrated to flood-prone areas near rivers. This concentration creates a natural heterogeneity of preferences for flood defense. These two aspects are also relevant for other issues in water management. Unidirectional spillovers and spatially heterogeneous preferences for public goods challenge the federal organization of flood defense. Following the lead of recent political-economy contributions to fiscal federalism, both aspects are investigated in a two region model with majority voting. Four different decision-making structures are compared: classical decentralization or centralization (based on jurisdictions containing voters both with and without preferences for the public good) and decentralized or centralized single issue authorities (whose jurisdictions, by definition, contain only high preference voters). Decentralized jurisdictions separate upstream and downstream voters, whereas a centralized jurisdiction comprises both groups. It turns out that centralized jurisdictions lead to a very low public good surplus and that either classical decentralization or decentralized single issue authorities achieve the best results under most conditions. The centralized provision of public goods is flawed because there is either an extreme over- or an extreme under-provision of the downstream public good. This can, however, be mitigated if there is a common standard for flood protection or if only the upstream provision of the public good is centralized. Whether or not these two solutions are beneficial depends on the magnitude of spillovers and also on the voting majorities in the two regions. Expanding the model to consider migration led to more complex results. First, citizens can migrate to flood-prone areas. Such migration flows can be induced by too large public flood defense, which is favored if citizens outside of flood-prone areas also finance the public good. Second, migration flow is also influenced by spillovers. The socially efficient population distribution is U-shaped with respect to spillovers. With symmetry assumptions, negative as well as positive unidirectional spillovers favour larger downstream populations. With myopic voters that neglect migration flow, this pattern is not achieved and the downstream population is large for positive spillovers and small for negative spillovers. Centralized jurisdictions are preferable to decentralized ones for smaller spillover effects if migration is possible. This argument is even stronger if voters take migration responses to the provision of public goods into account. The results shed light not only on the federal organization of flood management, but also on the institutional difficulties that arise from the current policy paradigm of the river basin approach to water management.
Abstract. This paper examines the process of mass democracy as the fundamental cause of transfer seeking and the centralization of governance, using Austrian-school theory and methodology such as decentralized knowledge, disaggregated phenomena, and the structure of capital goods. The alternative of decentralized, small-group gov-ernance reduces the demand for campaign financing and makes more effective use of decentralized knowledge. In addition, when public revenues originate in the local districts and are passed on to higher levels of governance, it provides incentives for revenue sources which do not have an excess burden on production. The governance struc-ture of cellular, bottom-up, multi-level voting, with public revenue flowing up from the lower to the upper levels, provides a contrast for a comparative systems analysis that can yield insight into the transfer seeking endemic in mass democracy.
Some basic facts of public goods theory are presented in the primitive set-up of a collection of projects devoid of any linear structure. There is a single private good. Characterizations of Pareto optimal and core states in terms of valuation functions (i.e. supporting “prices”) are obtained. Voluntary financing schemes are discussed.