This PhD thesis consists of three essays on income distribution, from the point of view of statistical production (methods) and economic development (history and institutions). The first chapter deals with the growing problem of household surveys to accurately portray the top tail if the income distribution. It proposes a new method to reconcile survey data with information from more trusted sources, such as tax data. The method produces a micro-dataset that preserves the consistency of other socio-demographic variables at both the individual and aggregate levels, to allow future research to be carried out under a more representative distributive framework. The procedure is illustrated by empirical applications to five countries, covering both developed and less-developed contexts over numerous years. The second and third chapters both make use of the method described in the first chapter to measure and analyse income inequality for different time periods and motives in Brazil -- a perennial late-developing economy, where household surveys are an increasingly problematic source from which to obtain credible information on the relative income growth of different parts of the population. The second chapter combines data from previously un-reconciled sources to uncover new evidence and a new understanding of income inequality in Brazil, focusing on the much debated period of the 2000s. It finds that inequality within the Bottom 90% of the distribution declined, but concentration at the top persisted at very high levels. This dichotomy was given by the strong average income growth in both tails of the distribution mainly between 2002 and 2013, while the middle of the distribution was squeezed. The fall in inequality among a large part of the population was due to the fall in labour earnings inequality, which was nonetheless insufficient to prevent the growing concentration of national income among economic elites. The chapter contextualises the findings to understand what may be driving the dynamics, from the progressive role of social policy, to the regressive role of the tax system and monetary policy. The third chapter extends the Brazilian inequality analysis over a longer historical time-frame to examine where it has come from. The overall objective is to shed new light on long-run distributional dynamics and their connection with economic growth in a late-developing country. Based on the construction of a rich inequality dataset covering the whole population since 1976 and a top income group since 1926, and its combination with other distributional information and macro statistics, the chapter shows the unprecedented levels and persistence of income concentration in Brazil, despite tumultuous economic and political change. It explains the absence of any sustained egalitarian levelling in the country through an endogenous theory of institutional shifts, which originate from structural-economic changes, but get ultimately appropriated by elites to avoid the redistribution of fundamental factors (land, capital, income, education) that the economic changes and related social actors seem to demand. It identifies the military coup of 1964 and its aftermath as a crucial moment in the history of Brazilian inequality, whose ideas and policies largely suspended inclusive growth, consigned distributional bottlenecks to future governments, and whose legacy can still be perceived today.