부산대학교 도서관

In this paper we propose a multi-objective evolutionary algorithm for supporting the definition of a personal income tax reform. As a case study, we apply this methodology to the Italian income tax, and consider a recently implemented tax cut. Our optimization algorithm is able to determine a set of tax structures that maximize the redistributive effect of the tax while minimizing its inefficiency - considering for the former the Reynolds-Smolensky index and for the latter the weighted average of taxpayers' effective marginal tax rates. The approach also takes into account two additional factors: the tax has to guarantee a specific revenue and to minimize the share of losing taxpayers with respect to the pre-reform situation. Experimental results clearly demonstrate that the methodology we employ can support the policy-maker's decisions in complex, real- world situations.