In this paper, I will describe how two strong connections between, on the one hand, income inequality and welfare generosity, and, on the other, punitiveness, have been built in both theoretical and empirical explorations in the contemporary comparative literature on the sociology of punishment. Then, I will point out the strong concentration of these explorations on national cases from the Global North as a potential limitation. From there, I will try to ‘southernize’ this debate, through three empirical exercises related to a region of the Global South, Latin America. First, I will include this region in a global comparison of clusters of countries to define whether there is an association between the levels of income inequality and welfare generosity and the levels of punitiveness, both now and in the recent past. Second, I will analyse if the same relationships exist within Latin America countries, both now and in the recent past. Finally, I will examine whether these same relationships are relevant for understanding the evolution of the levels of punitiveness in Latin America over the last three decades. Based on the results of these three exercises, I will examine the shortcomings stemming from assuming these strong statements as universal, placeless and timeless, warning that the styles of comparison that have generated them have to be taken as starting points rather than as arrival points of the analysis and stressing that our analyses about contemporary penal differences, while taking macroscopic dimensions into account, should give a strong centrality to the ‘proximate’ processes that mould penal actions and results.