Threshold effects of governance on economic growth: evidence from Sub-Saharan Africa
摘要
This paper investigates the impact of the quality of governance on economic growth across 20 Sub-Saharan African countries over the period 2000–2023. A composite governance index is constructed using Principal Component Analysis on six indicators from the Worldwide Governance Indicators database. Employing a range of econometric techniques—Group Mean Fully Modified Ordinary Least Squares (FMOLS-GM) panel cointegration, Granger causality tests, and a Dynamic Panel Threshold Model—the study explores both the long-run relationship and non-linear effects between institutional quality and per capita GDP. The study accounts for cross-sectional dependence—a factor often neglected in panel studies—by augmenting the FMOLS-GM regression accordingly. The findings reveal strong evidence of cointegration between governance and growth, as well as bidirectional Granger causality. Importantly, the influence of governance on growth is found to be non-linear, with threshold effects observed to be statistically significant across all governance indicators. The growth-stimulating effects of improved governance indicators materialize only when the associated institutional quality crosses the threshold representing the minimum critical institutional quality. The threshold levels vary across indicators, suggesting that some aspects of governance require a significantly stronger foundation before their growth-enhancing effects can be fully realized in the economy. Country-specific governance performance insights are also presented to help identify the governance dimensions that require policy attention. The analysis underscores the importance of targeted institutional improvements. It offers critical implications for reform strategies in the region, highlighting that stronger institutions are not only growth-enhancing but also responsive to improved economic performance.