Mahbuba Aktar / Makram El-Shagi
CFDS Discussion Paper 2026/5
Abstract
This paper studies how digital credit – specifically BigTech and fintech lending – relates to income inequality in 42 countries from 2013 to 2019. We develop a dynamic multi-equation panel framework based on isometric log-ratio transformations of income shares, which allows us to model shifts in the entire Lorenz curve while respecting the compositional constraints that standard approaches typically ignore. BigTech credit consistently reduces the top income share, and its inequality-reducing effect is most pronounced in developed economies, in countries with stronger institutions, and in financially open environments. Fintech credit exhibits the opposite pattern: it raises the top income share while lowering the bottom 50 percent’s share, with the magnitude of this effect similarly amplified by higher development, stronger institutional quality, and greater financial openness.