Abstract
Poverty remains a persistent challenge in Indonesia, particularly in provinces with structural socioeconomic disadvantages, limited access to education, healthcare, infrastructure, and formal employment. This study examines the determinants of poverty using panel data from the ten provinces with the highest poverty rates during 2019 –2023. The independent variables include Gross Regional Domestic Product (GRDP), Unemployment, Human Development Index (HDI), School Participation Rate (SPR), Domestic Investment (DI), and Foreign Direct Investment (FDI). A quantitative approach employing panel data regression with the Random Effects Model (REM) was applied, based on Hausman test results indicating that unobserved heterogeneity across provinces is best treated as random. This method allows for capturing both inter - provincial and temporal variations, providing robust estimates of the magnitude and direction of each variable's effect on poverty. The empirical findings reveal that, simultaneously, all independent variables significantly influence poverty levels. Partial ly, unemployment and HDI negatively and significantly affect poverty, indicating that higher human development and lower unemployment reduce poverty. In contrast, school participation rate and domestic investment show significant positive effects, possibly reflecting the limited effectiveness or uneven distribution of education and investment benefits in these regions. Meanwhile, GRDP and FDI are not statistically significant, suggesting that general economic growth and foreign investment alone are insufficient for alleviating poverty. These results provide critical insights for policymakers, emphasizing the nee d for targeted strategies that integrate human development, employment generation, and effective domestic investment to achieve inclusive poverty reduction in Indonesia’s poorest provinces.