Gender staff in the World Bank -- the world's largest and most influential development institution -- have a policy problem. Having prioritised efforts to get women into paid employment as the ȁ8cure-allȁ9 for gender inequality they must deal with the work that women already do -- the unpaid labour of caring, socialisation, and human needs fulfilment. This article explores the most prominent policy solution enacted by the Bank to this tension between paid and unpaid work: the restructuring of normative heterosexuality (...) to encourage a two-partner model of love and labour wherein women work more and men care better. Through a case study of Bank gender lending in Ecuador I argue that staff are trying to (re)forge normative arrangements of intimacy, a policy preference that remains invisible unless sexuality is taken seriously as a category of analysis in development studies. Specifically, I focus on four themes that emerge from the attempt to restructure heteronormativity in the loan: (1) the definition of good gender analysis as requiring complementary sharing and dichotomous sex; (2) the Bank's attempt to inculcate limited rationality in women such that they operate as better workers while retaining altruistic attachments to loved ones; (3) the Bank's attempt to inculcate better loving in men, such that they pick up the slack of caring labour when their (partially) rational wives move into productive work, and; (4) the invocation of a racialised hierarchy resting on the extent to which communities approximate ideals of sharing monogamous partnership. Aside from providing clear evidence that the world's largest development institution is involved in micro-processes of sexuality adjustment alongside macro-processes of economic restructuring, I also critique the Bank's sexualised policy interventions and suggest that they warrant contestation. (shrink)