Impact bonds and the ambiguous politics of market ethics

Recent years have seen the emergence of impact bonds, which are a striking case of both the post-crisis ‘social turn’ of finance and the financialisation of social and development policy. In an impact bond, investors pre-finance new social interventions for marginalised target groups. If such a programme can demonstrate positive effects on the lives of the target group, investors are compensated by a public or charitable institution. This thesis offers an in-depth examination of the wider rationalities through which actors ‘make sense’ of and authorise such vehicles—and what political consequences this has. Using a Foucault-inspired analytics of governmentality framework, I develop three case studies of impact bonds that address homelessness and long-term health conditions in the UK, and the gender gap in educational attainment in India, respectively. My investigation is based on document analysis, 32 semi-structured interviews with project stakeholders, and participant observation at project sites and conferences. It demonstrates that, rather than simply reproducing a financial ‘script’, these programmes configure scientific, ethical, managerial, financial, and socio-political rationalities into a programmatic form that employs a fact-based approach to addressing social ills. I unpack how the focus on evidenced results both reassures individuals that they are not mistaken about what it means to invest for good and immunises projects against political confrontation. These considerations lead me to make three interventions to emphasise the ambiguity of impact bonds: (i) outcome measurement is more provisional than commonly suggested, (ii) impact bonds not only finance experimental projects but become a normalised contracting model, and (iii) their focus on ‘victims’ obscures broader socio-economic causes of the social problems addressed. At the same time, however, impact bonds also rework exclusionary constellations ‘from within’ and enable the operation of new (if imperfect) projects and everyday face-to-face interactions. They thus constitute contingent ethical moments that require ongoing engagement