The empirical literature shows mixed results regarding the impact of public spending on the gross domestic product (GDP) of countries in general. This paper presents evidence on the impact of public spending on regional GDP in 24 departments of Colombia, during the period 1984-2012. In addition, we apply a causality test with panel data with the aim to determine whether this relationship is supported by the Keynesian approach or the Wagnerian approach. Using a model of cointegrated panel data, we obtain the long-run equilibrium relationship and estimate its coefficients. The results show a significant and positive effect of primary public spending on GDP, which is in line with the Keynesian approach and confirms the hypothesis that increases in public spending contribute to economic growth.