Abstract
Despite their ostensible differences, both randomized controlled trials (RCTs) and neoclassical economics, which undergirds neoliberalism, are characterized by unrealistic assumptions. This is not accidental but stems from a common desire for scientism, despite a substitution of “economic man” by what we call “behavioral man” in the former. However, the interactions between human behavior and context produce much greater diversity than allowed for in such approaches. As a result of these failings, RCTs do not challenge neoliberalism, but rather can be inserted into and help legitimize it. Consequently they do not alter the structural conditions producing poverty, but represent a form of “virtualism” that attempts to make the real conform to the ideal of neoclassical economics through “nudging.” This is often associated with conditionality, representing a continuation of the historically Northern-centered coloniality of power, which sometimes leads to ethical breaches and perverse development outcomes. Some examples from RCTs conducted or published by the World Bank—which has become a lead agency in promoting RCTs to assess the effectiveness of aid—in Africa are presented to illustrate our argument.