Is the resource curse persistent? Dynamic causality evidence from resource-dependent economies

Abstract

This study examines the dynamic relationship between natural resource rents and economic growth in ten resource-dependent economies—Algeria, Democratic Republic of Congo, Egypt, Ghana, Indonesia, Kuwait, Malaysia, Oman, Qatar, and Saudi Arabia—over the period 1970–2021. Using a bootstrap panel causality test combined with a rolling-window framework, the analysis captures both the direction and time variation of predictive linkages within the Resource Curse Hypothesis (RCH). The results indicate that natural resource rents exhibit episodic and heterogeneous predictive relevance for growth across countries and periods, rather than a stable longrun effect. Positive predictive episodes are more frequently observed in Qatar, Indonesia, and Egypt, while Algeria, Malaysia, and Saudi Arabia display more recurrent adverse episodes. Kuwait and Oman show largely neutral patterns, consistent with limited sensitivity to rent fluctuations. Overall, the findings support a conditional and time-dependent interpretation of the resource curse, emphasizing that the relevance of resource rents for growth varies across historical phases and macroeconomic contexts. The study highlights the importance of accounting for temporal heterogeneity when assessing the resource–growth nexus, rather than relying solely on full-sample estimates.