On the heterogeneous link between public debt and economic growth

We use panel data for 115 countries over the period 1995-2016 to model the heterogeneity of the debt-growth nexus along with the underlying factors that might explain it. The grouped fixed effect (GFE) estimator is used to endogenously classify countries into groups and a multinomial logit model is employed to explore the drivers of the detected heterogeneity. The GFE estimator classifies countries into five groups for which debt has different impacts on growth. According to our results, the likelihood of a strong impact is moderated by the quality of the institutions and the proportion of productive expenditure but intensified by the level of indebtedness and the maturity of the debt