Abstract Background Whilst evidence suggests cognitive behaviour therapy (CBT) may be effective for depressed older people in a primary care setting, few studies have examined its cost-effectiveness. The aim of this study was to compare the cost-effectiveness of cognitive behaviour therapy (CBT), a talking control (TC) and treatment as usual (TAU), delivered in a primary care setting, for older people with depression. Methods Cost data generated from a single blind randomised controlled trial of 204 people aged 65 years or more were offered only Treatment as Usual, or TAU plus up to twelve sessions of CBT or a talking control is presented. The Beck Depression Inventory II (BDI-II) was the main outcome measure for depression. Direct treatment costs were compared with reductions in depression scores. Cost-effectiveness analysis was conducted using non-parametric bootstrapping. The primary analysis focussed on the cost-effectiveness of CBT compared with TAU at 10 months follow up. Results Complete cost data were available for 198 patients at 4 and 10 month follow up. There were no significant differences between groups in baseline costs. The majority of health service contacts at follow up were made with general practitioners. Fewer contacts with mental health services were recorded in patients allocated to CBT, though these differences were not significant. Overall total per patient costs (including intervention costs) were significantly higher in the CBT group compared with the TAU group at 10 month follow up (difference £427, 95% CI: £56 - £787, p < 0.001). Reductions in BDI-II scores were significantly greater in the CBT group (difference 3.6 points, 95% CI: 0.7-6.5 points, p = 0.018). CBT is associated with an incremental cost of £120 per additional point reduction in BDI score and a 90% probability of being considered cost-effective if purchasers are willing to pay up to £270 per point reduction in the BDI-II score. Conclusions CBT is significantly more costly than TAU alone or TAU plus TC, but more clinically effective. Based on current estimates, CBT is likely to be recommended as a cost-effective treatment option for this patient group if the value placed on a unit reduction in BDI-II is greater than £115. Trial Registration isrctn.org Identifier: ISRCTN18271323