Emigration first increases and then decreases as a country experiences economic development. This inverted U-shaped, cross-sectional relationship between emigration and development was first hypothesized by Zelinsky's theory of the mobility transition. Although several mechanisms have been proposed to explain the upward segment of the curve (the most common being the existence of financial constraints), they have not been examined in a systematic way. In this paper, we propose two decomposition methods to disentangle the main drivers of the mobility transition curve to OECD destination countries. Our simple decompositions shed light on the role of both microeconomic drivers (i.e., financial incentives and constraints) and macroeconomic drivers, as well as the skill composition of the population. Our double decomposition further distinguishes between migration aspirations and realization rates by education level. Overall, we provide consistent evidence that the role of financial constraints, while relevant for the poorest countries, is limited. Rather, a large fraction of the increasing segment is explained by the skill composition and by macroeconomic drivers (i.e., by factors that do not change in the short-run).