The purpose of this study is to test the Financial Kuznets Hypothesis for country groups with different levels of development for the years 2000-2021 and to determine whether the obtained findings differ according to the development level of the countries. For this purpose, four sample groups were formed in the study: “least developed”, “developing”, and “developed” country sample groups, and an “all countries” group consisting of all of these groups. To analyze the long-run and short-run dynamic relationships between the variables, the Panel Autoregressive Distributed Lag (Panel ARDL) model and the Pooled Mean Group (PMG) estimator were used. In order to check the robustness of the results, contemporary econometric methods such as the U-Test, the Two-Step System Generalized Method of Moments (Two-Step System GMM), and the Fixed Effects (FE) and Random Effects (RE) models were used. The findings obtained indicate that the Financial Kuznets Hypothesis is not valid for all of the sample groups. However, evidence has been found supporting the Inequality-Narrowing and Inequality-Widening Hypotheses for the sample groups. This evidence, obtained with respect to the Inequality-Widening and Inequality-Narrowing Hypotheses, differs significantly depending on the development level of the countries. For example, while financial development significantly reduces income inequality in least developed and developing countries, it increases it in developed countries. These findings indicate that the effect of financial development on income inequality is not homogenous and that, therefore, it may be important that financial policies to be implemented to reduce income inequality are designed based on the structure of countries' financial systems, their levels of development, and the type of policies they primarily wish to use to reduce income inequality.