This paper uses historic data from Cundinamarca, Colombia to empirically assess the impact of land inequality persistence, inherited from the colonial rule, on economic development in the long run. Based on the Engerman & Sokoloff hypothesis and the use of GIS, I use plausible exogenous variation in land endowments to design an instrumental variable strategy. In contrast to recent studies, I find that more unequal municipalities in the XIX and XX century are associated with better growth, human capital and public goods provision measures today. Political economy channels instead of agricultural productivity gains can explain these results. In municipalities where land was historically more concentrated, powerful landowners were more successful in solving their collective action problem of accessing political power to influence the allocation economic resources in their interests.