This study explores the impact of fiscal decentralization on income convergence across Colombian municipalities. Using the theory of fiscal federalism and the endogenous growth model, I assess whether increased fiscal autonomy has contributed to reducing income disparities at the municipal level. To address the challenge of measuring local economic activity, I apply an Artificial Neural Network trained on Night-time Lights data to estimate municipal per capita Value-Added from 1993 to 2018. A two-stage least squares (2SLS) approach with instrumental variables is used to account for potential endogeneity in the explanatory variables. The results show that fiscal decentralization, particularly revenue-based measures, significantly enhances income convergence, with annual rates ranging from 2.7% to 4.2%, while a 1% increase in decentralization is associated with a 0.16% to 0.28% increase in per capita VA growth. On the other hand, the impact of expenditure-based decentralization, mainly higher autonomy on expenditure decisions, on growth and convergence is inconclusive. The effect is more pronounced in lower-income municipalities, where investments in public capital ought to be more productive. This research contributes to the literature by providing new insights into the role of fiscal decentralization in reducing regional inequality, particularly at the municipal level.