On the Decomposition of Regional Stabilization and Redistribution

This study proposes decomposition and estimation methods that can be applied to analyze both regional stabilization and redistribution. The method proposed herein follows the approach taken by Shorrocks (1982), and applies it to per-capita level quantities of the relevant variables rather than the log-linear quantities used by Asdrubali et al. (1996) for regional stabilization and the normalized per-capita quantities used by Bayoumi and Masson (1995) for regional redistribution. I directly calculate the proportional contributions to the decomposition and bootstrap their confidence intervals rather than indirectly obtain them as OLS estimates from the artificial regressions by Asdrubali et al. (1996). I then apply the proposed method to Japanese prefectural accounts data so that we can compare the presented analysis with those in previous studies. Furthermore, I also apply the method to municipal budgetary data in Japan in order to demonstrate its usefulness.