Creating an index of local autonomy – theoretical, conceptual, and empirical issues

Any attempt to create an overall measurement of local autonomy is confronted with difficult questions: Is local autonomy a one-dimensional concept? If it is multidimensional, what dimensions should be taken into account? Are legal, functional, financial, organizational, and vertical aspect of autonomy of equal importance, or must they be weighted? How can they be aggregated: can they simply be summed or are more complicated methods needed? On the basis of an international research project covering some 39 countries over 25 years, we will discuss the choices that have to be made and show what implications they have on results. The various solutions will be tested with respect to their internal consistency and their relations to other measurements of decentralization. An index always reduces complexity and tells only a part of the story, but it can be very useful for rankings, comparisons and further analyses.