Direct payments – cash for people eligible for adult social care and spent by them on care and support – are claimed to enable care to better reflect user preferences and goals which improve outcomes. This paper compares outcomes of older direct payment users and those receiving care via a managed personal budget (where the budget is spent on the recipients behalf by a third party). The study adopted a retrospective, comparative design using a postal questionnaire in three English councils with adult social care responsibilities in 2012–13. Included in the study were 1,341 budget users aged 75+, living in ordinary community settings. The overall response rate was 27.1 per cent (339 respondents). Three validated scales measured outcomes: EQ-5D-3L (health status), the Sheldon–Cohen Perceived Stress Scale and the Adult Social Care Outcomes Toolkit (social care-related quality of life). The study found that direct payment users appreciated the control conferred by budget ownership, but in practice, for many it did not ‘translate’ into improved living arrangements. It also found no statistically significant difference in outcomes between direct payment and managed personal budget users. The paper argues that despite policy and other guidance and research evidence about effective implementation of direct payments for older people, the absence of evidence for better outcomes may at least in part be attributable to values underpinning policies relating to personalisation and personal budgets.