Processes of public engagement in decision-making and research are increasingly discussed as ways of addressing democratic deficits in high-income countries. In this paper, we explore why these processes of engagement and involvement in the UK have been less successfully incorporated into social security policymaking aimed at the out-of-work by drawing a comparison with health policy, a sphere in which these processes have now become orthodox (albeit imperfect). There is, for example, no formal or institutionalised imperative to involve people with lived experience of out-of-work social security benefits in processes of policy development. Government departments might focus group new policies with members of the public or hold periodic discussions with beneficiaries but in recent years there have been a number of major reforms to out-of-work social security which have been developed almost entirely without involving those affected. This would have been unacceptable in the health policy arena. We argue that this difference is rooted in structural differences in how the field of power for this form of social policy is organised, in the different social imaginaries which construct patients and out-of-work beneficiaries, and in the limited scope for solidarity and collective action around resisting the stigmatisation of out-of-work beneficiaries.