Economic growth and income inequality: Non-linearities and income level threshold effect

Despite the pioneering works of Kuznets and Kaldor in the mid-fifties, it was not until the nineties that economists began to thoroughly investigate whether income inequality is detrimental or conducive to growth. A brief survey of the literature does not yield a definitive conclusion. Some authors argue that inequality is beneficial, as higher-income populations are more likely to save and invest in innovative activities. Conversely, other authors contend that concentrating income in the hands of a few limits the number of potential investors, hinders the development of markets and industries with increasing returns, and may dis-incentivize investment in human capital. Since the mid-nineties, most empirical studies attempting to uncover a link between income inequality and growth have concluded that the relationship is predominantly negative. In this study, we reveal a more complex link, characterized by an inverted-U shape that is mostly negative and only applicable to countries with intermediate income per capita. For poor and rich countries, this link does not exist. We employ a panel threshold model approach, introducing income inequality in linear, quadratic, and cubic forms to allow for potential linear, U-shaped, inverted-U-shaped, and even more complex relationships between inequality and growth.