Macroeconomic determinants of poverty in South Africa: the role of investments in artificial intelligence

Objectives: This study marks a departure from conventional research that predominantly focuses on AI's impact on economic growth. Instead, it uniquely centers on the interplay among fiscal policy, AI investment, monetary policy, and poverty alleviation in South Africa, a nation recognized for its extreme income inequality. In doing so, this research distinguishes itself from the prevailing macro-level studies by scrutinizing the intricate relationship between artificial intelligence, poverty, monetary policy, and fiscal policy, particularly within the South African context. Methods/Approach: It employs the Praise-Winsten model and other econometrics diagnostics for the empirical analysis. Results: The findings reveal that government expenditure on education and investments in artificial intelligence only stimulate household consumption and, therefore, reduce poverty when they are interacted. A one-unit rise in the interaction term increases household consumption by 0.031. Additionally, a unit increase in government national expenditure and broad money growth results in a 0.007 and 0.014 decrease in poverty, respectively. Similarly, gross capital formation positively affects household consumption, thereby reducing poverty in the country by 1.022 due to a one-point increase. In general, taxes on goods and services and the repo rates exert a non-statistically significant effect on the level of poverty in the country. Conclusions: To effectively address these findings, it is imperative to conduct a thorough evaluation of government spending on education to ensure prudent resource allocation and prevent the mismanagement of public funds.