Reassessing the Relationship between Billionaires, Billionaires’ Wealth and Per-Capita GDP: A Gini Regression Analysis

31 Pages Posted: 3 Feb 2018 Last revised: 27 Oct 2018

See all articles by Aloys Prinz

Aloys Prinz

University of Münster - Institute for Public Economics

Steffen Bollacke

affiliation not provided to SSRN

Date Written: January 26, 2018

Abstract

In this paper, Gini Regression Analysis (Olkin and Yitzhaki 1992) is applied to check the robustness of the positive correlation of the number of billionaires, billionaires’ wealth, and the GDP per-capita, as reported in Prinz (2016). Although Gini Regression Analysis (GRA) as a method exists since more than a decade, its value for robustness checks seems not well understood. In contrast to OLS estimations, GRA is based on Gini’s mean difference instead of the standard deviation. Furthermore, the method of Yitzhaki and Schechtman (2012) that is also based on GRA, is applied in this paper to test the monotonicity of the relationship between the independent billionaire variables and GDP per-capita. The result is that although the positive correlation of the number and wealth of billionaires with GDP per-capita is robust for the complete sample of countries, this cannot be sustained when countries are separated into poor and rich ones.

Keywords: Gini regression, NLMA-curves, normality test, billionaires, GDP per-capita

JEL Classification: C1, I3, O5

Suggested Citation

Prinz, Aloys and Bollacke, Steffen, Reassessing the Relationship between Billionaires, Billionaires’ Wealth and Per-Capita GDP: A Gini Regression Analysis (January 26, 2018). Available at SSRN: https://ssrn.com/abstract=3110532 or http://dx.doi.org/10.2139/ssrn.3110532

Aloys Prinz (Contact Author)

University of Münster - Institute for Public Economics ( email )

Wilmergasse 6 – 8
Münster, 48143
Germany

Steffen Bollacke

affiliation not provided to SSRN

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
87
Abstract Views
610
Rank
524,237
PlumX Metrics