Information theory and an entropic approach to an analysis of fiscal inequality

Abstract

In his influential study, Theil (1967) developed the notion of entropy on the basis of information theory. He then advocated the use of entropy-based measure for the analysis of income inequality. In this paper, the first of its kind, we apply Theil's notion of entropy to public finances in multi-tiered governments, in particular for a measurement of fiscal decentralisation, which is currently very crude in terms of the ratio between local government revenue and total revenue. It is the claim of this paper that such an approach of measuring fiscal decentralisation completely ignores important distributional aspects of fiscal arrangements. Findings from this paper indicate that studies on measuring various aspects of fiscal activities-such as fiscal decentralisation-should carefully take into account the dispersion of revenue (and expenditure) across regions. On that basis, the entropic approach developed in this paper is able to accommodate these dispersions across subnational governments. As an illustration for the case of Vietnam, the true degree of fiscal decentralization has effectively been decreased in comparison with estimates from other simple measurements due to the presence of substantial dispersions of revenue and expenditure from the subnational governments across 63 provinces in Vietnam.