The Heterogeneous Effect of Economic Complexity on Growth and Human Development: New Empirical Evidence Using Buffered Panel Threshold Regression

ABSTRACT In this study, we apply the Buffered Threshold Panel Data model developed by Belarbi et al., to explore the effect of economic complexity (EC) on growth and human development. We use a panel data set from 92 developed and developing countries from 2002 to 2016. Our findings reveal a nonlinear relationship between EC growth and human development, contingent on the studied countries' institutional quality and resource dependence. Contrary to previous research, we demonstrate that the effects of EC on growth and human development are heterogeneous and could be negative both in the short, medium or long term in some growth regimes. We also provide evidence supporting the convergence hypothesis for growth while suggesting a divergence for human development. A noteworthy finding of our study is that the extent of the negative impact of EC on growth varies based on a country's resource dependency. Furthermore, we observe a dual effect of institutional quality on growth, as measured by the Rule of Law, when Oil Rents is used as a threshold variable. Specifically, improvements in institutional quality have a positive impact on growth in economies less reliant on natural resources, particularly oil. Conversely, the effect is negative in economies significantly reliant on oil exploitation.