This research evaluates the robustness of the regional Kuznets curve using Bayesian averaging methods for panel data. Simulation results indicate the method successfully identifies variables in the underlying data generation process. Key findings show that beyond real GDP per capita (linear and quadratic terms), the most robust determinants of regional inequality are natural resource rents, arable land and ethnic inequality. An inverted-U relationship appears between regional inequality and development spanning USD 189 to USD 71,682, with inequality stabilizing beyond this threshold rather than increasing.