Structure of income inequality and household leverage : Cross-country causal evidence

International audienceHow does income inequality and its structure affect credit? Based on various strands of the literature, we hypothesize that rising income inequality should lead to higher house- hold credit at the aggregate level, and that a substantial part of this effect should be driven by the impoverishment of the middle class relative to top-income households. These intu- itions are empirically confirmed by a study based on a country-level dataset over the pe- riod 1970–2017. To identify exogenous variations in inequality, we develop an instrumental variable approach based on two types of country-level instruments: the total number of ratified ILO conventions and factor endowments. Our results show exogenous variations in inequality have a positive impact on household credit: a one-standard-deviation increase in the Gini index generates a 5- to 8- percentage-point expansion in the ratio of house- hold credit to GDP. In addition, the impact is 1.5–1.8 times stronger when the increase in inequality is driven by the income of top earners relative to the middle class rather than by the increase in top earners’ incomes at the expense of the lowest percentiles of the dis- tribution. Those results are robust to various sets of instruments, databases, controls, and variable definitions. They also consistently disappear in countries where financial markets are insufficiently developed