Intergovernmental grants, urban congestion and the provision of local public goods.

The basic motivation for the development of this research was to understand why we observe a degradation of local services in urban and suburban areas in Portugal. There is, I believe, an excess of demand for local services in relation to those supplied by local governments. The sprawl of illegal housing construction and the fiscal pressures associated with rapid and unplanned urbanization suggested that a "solution" to the problem should be investigated on the demand side of the problem. Famous court cases in the U.S.A. have dealt precisely with this issue. However, the enforcement of these planning rules is still a problematic issue. Zoning would be expected to be ineffective in countries like Portugal because people do not obey the Law in the same way as in Britain or the United States. Considering also that zoning raises difficult ethical issues, I realized I should start looking at the supply side of the problem. Assuming that urban growth is what it is, why do local governments not increase supply accordingly. The analysis of the particular case of Portugal clarified that part of the answer to this question relies on the centralized nature of government, where the ability of local governments to realize discretionary changes in their revenues (and therefore expenditures) is severely constrained. Therefore, this thesis can be understood as an inquiry into the implications on the quality of local services of a centralized system of government. However, most of the economic literature on local governments' decision making assumes a decentralized government and therefore the issue that naturally arises is whether this research has only a parochial scope (the Portuguese case) or a more broad range of interest. There are two main reasons to justify a broader scope for this thesis. Firstly, we might use Tullock's argument that people usually write about democracy although the majority of political regimes in the world are still autocracies. The same applies to decentralized and centralized countries if the conjecture that autocracies have usually politically centralized systems of government is accepted. Economists usually use models assuming decentralized governments (e.g. the median voter model) when most of the countries in the world are autocracies and therefore most likely centralized. Moreover, even within democratic countries there are centralized systems of government and on the other hand those which are decentralized are always subject to centralization trends. Secondly, some issues addressed in this thesis are not confined to centralized governments. This is the case of the analysis on the economies. This analysis is extremely important when considering intergovernmental grants with equalization purposes. The institutional and fiscal rigidities associated with centralized governments undermine the idea that local governments provide services according to the preferences of a representative voter within each jurisdiction (the median voter). On the other hand, it was an unexpected conclusion that there are affinities between the centralized governments' approach developed in this thesis and the approach developed by Tiebout (1956). Citizens unable to influence local decision-making through the vote will, ceteris paribus, migrate to jurisdictions where the fiscal "package" (local services and taxes) is more in line with their preferences. Alternatively, citizens who have more ability to pay may simply "exit" from public towards private provision. The analysis that follows is essentially the diagnosis of the problem. (Abstract shortened by UMI.).