Good governance and natural resource management in oil and gas resource-rich countries: A machine learning approach

Abstract

Good governance and performance in managing oil and gas resources is an area of growing interest in resource-rich countries in the global economy. Countries with higher good governance indexes are expected to be more successful in managing natural resources. This study uses machine learning to investigate the good governance role in oil and gas resource management in 55 countries in 2017. Although our primary focus is on the oil- and gas-rich countries, the findings can be applied to other resource-rich countries. Using the Principal Components Analysis and K-clustering approach, countries are clustered into six groups based on good governance data. The result shows that only the four developed countries (i.e., 7% of the sample countries including Norway, UK, Canada, and US) show a high level of good governance. However, the other countries, including 82% of the sample, mainly indicate a low level of good governance. Since this majority wholly includes developing countries, the results not only confirm the resource curse hypothesis but also reveal the association between good governance and natural resource management. In addition, the results indicate that the countries in the same cluster of good governance are located in the neighboring regions, implying the effects of various cultural, social, historical, traditional, and trade factors. These countries should manage the resource revenue in a transparent and adequately taxed framework while increasing economic diversity and complexity. Furthermore, they should enhance regional cooperation and knowledge sharing by considering the spatial and geographical flows and effects to improve good governance.