In this chapter, I introduce some concepts and a formal notation that will be useful for examining arguments about the consequences of decentralization. Of course, there are many ways one might think about the political process, and the conceptualization I offer here is chosen for convenience. I begin by defining various types of decentralization.

Defining Decentralization

“‘Centralization’ is now a word constantly repeated but is one that, generally speaking, no one tries to define accurately,” Tocqueville wrote in 1835 (Tocqueville 1969 [1835], p. 87). These days, the problem is more that people define centralization and decentralization in many different ways, and any two scholars or policy makers who sit down to debate the subject will usually have different things in mind. Some also try to squeeze several concepts under a single label. I find it helpful to distinguish different types or aspects of decentralization. The definitions I suggest here are useful primarily for making and analyzing theoretical arguments. To adapt these to cross-national empirical comparison would take a great deal of additional thought.

Before defining decentralization, I need to define a few other terms. Each state governs a particular territory, or set of points in space. A jurisdiction is a subset of the territory consisting of contiguous points. While some jurisdictions contain others, I assume the borders of jurisdictions do not cross. The largest jurisdiction comprises the state's entire territory, and I call this the first tier jurisdiction.