Does Social Assistance Expenditure Reduce Poverty? Panel Evidence from Indonesian Provinces

Abstrak

This study investigates the determinants of poverty across Indonesian provinces with a particular focus on the role of social assistance expenditure. Using provincial panel data from 2015 to 2024, the analysis combines information on poverty headcount ratios with fiscal, economic, and political variables, namely social assistance per capita, gross regional domestic product (GRDP) per capita, unemployment rate, and election years. The study employs a fixed effects panel regression model, selected on the basis of specification tests, and incorporates one-year lagged values of social assistance and GRDP per capita to capture the delayed effects of fiscal and economic policies. The results reveal that social assistance has a negative but modest effect on poverty, indicating its limited yet relevant role in supporting vulnerable households. GRDP per capita emerges as the strongest determinant, confirming the importance of inclusive growth in driving poverty reduction, while unemployment significantly worsens poverty outcomes. In addition, poverty rates tend to fall in election years, reflecting the influence of political cycles on welfare spending. Overall, the findings underscore that poverty reduction in Indonesia requires multidimensional strategies that combine sustained economic growth, labor market improvements, and well-targeted social assistance, supported by institutional safeguards that ensure consistency beyond short-term political incentives.