In this paper we study the long term effects of inequality on development. We construct a novel dataset of land and social inequality in Argentina, with data from multiple time periods and at the highest level of geographic disaggregation. This dataset allows us to exploit the great subnational variation in the levels of human and economic development, as well as fine-grained data on socioeconomic, structural, political, and geographic variables. We seek to answer the questions of whether, how and when inequality affects development, a debate that has motivated a big body of literature. We argue, along the lines of Coatsworth (2008), that inequality forged early in the process of state formation is especially consequential for long term trends of development, and we test our theory by an- alyzing several different dimensions of socioeconomic and land inequality. This paper makes multiple contributions. First, it puts together an original dataset on landholding and ethnic inequality at the municipal level from 1869. Second, it empirically unpacks the concept of land inequality, by considering landhold- ing inequality but also inequality in access to land by different ethnic groups, following Baldwin and Huber (2010). Third, by collecting data from a moment early in the process of state formation, it makes a claim about long term effects of inequality on development.