This article reviews the proximate factors of human welfare since 1870 by discussing two strands of the economic history literature and identifying various key areas for further research. The first strand focuses on level accounting studies that attribute between-country economic inequality to differences in capital and productivity. I argue that most income gaps in the late 19th century were due to variation in physical and human capital endowments, while widening productivity differentials account for most of rising cross-country inequality during the 20th century. These patterns are likely explained by waves of skill- and capital-biased technological innovation, but additional research is needed to underpin these findings in, at least, three ways: capital and income series should be deflated by appropriate price indices, samples should include many more lower-income countries and methodologies could explore more realistic production functions. The second strand of the literature I review considers the measurement of long-run human development. Three approaches are popular among practitioners (capability, data-driven and utility frameworks), although there is still no consensus on which one to use. This makes it challenging to interpret broad trends in human welfare, as different well-being indices show contrasting patterns of growth and inequality. I argue that the field needs a more solid theoretical foundation to guide our choice of measurement frameworks. In this respect, utility-based indicators may be especially useful, as they address relevant issues raised in the literature, such as how to weight different dimensions, how individuals trade off between them, and how to interpret the results.