For decades, the practice and theory of development aid have lurched from one policy fashion to the next (Rist 2006). Although a plurality of aid paradigms has competed at any one point in time (Schuurman 2000), a single paradigm has usually dominated. From the launch of development assistance in the late 1940s (Frey et al. 2014) to the early 1980s, the notion held sway that ‘development’ could be measured by the annual growth of per capita national income. Moreover, it was believed that such growth could only ‘take off’ in developing countries if significant amounts of capital were transferred from the coffers of rich states to those of poor ones. This would allow the governments of those poor countries to increase spending on roads, canals, schools, electricity generation and distribution, irrigation schemes and other infrastructural projects, thus facilitating self-sustaining economic growth (Solow 1956; Rostow 1960; Chenery and Strout 1966).