Parametric representation of the top of income distributions: Options, historical evidence, and model selection

Empirical distributions of top incomes suffer from statistical problems affecting the measurement inequality and its trend. Researchers practitioners have been increasingly noting parametric regularities across income turning to functions approximate or supplement observed distributions, both for descriptive purposes correcting distributional statistics derived data. The proliferation distinct branches modeling literature has highlighted need compare alternative options, develop systematic tools discriminate between them. This paper reviews state methodological empirical knowledge regarding adoptable functions, lists references programs allowing apply these models microdata in household surveys administrative registers, grouped-records data national accounts statistics. Implications other economic outcomes including consumption wealth, below topmost tail, are drawn. For incomes, a handful distribution hold promise tails based on theoretical properties—namely extreme values generalized Pareto, Singh–Maddala beta type 2. Understanding relation commonly invoked alternatives is contribution this review.