Objectives The metacognitive model of rumination and depression (Papageorgiou & Wells, 2003, Cognitive Therapy and Research, 27, 261) postulates that beliefs that perseverative negative thinking, i.e. rumination, will help solve problems contributing to rumination. However, this activates negative beliefs about the uncontrollability and social consequences of ruminations, which exacerbate depression. The metacognitive model has been well‐supported but with some inconsistencies in specific pathways. It has also not yet been tested for postpartum depression (PPD). Therefore, this study aimed to examine the relations between the metacognitive model of rumination and depression when applied to PPD symptoms and to compare it with the cognitive model of depression. Design This is a cross‐sectional study. Method Postpartum mothers (N = 603) participated in an online study in their first postpartum year. They completed the Edinburgh Postnatal Depression Scale (EPDS), Postnatal Negative Thoughts Questionnaire (PNTQ), Ruminative Responses Scale (RRS), Positive Beliefs about Rumination Scale (PBRS) and Negative Beliefs about Rumination Scale (NBRS). Results A path analysis revealed that the model had an excellent fit to the data. Specifically, positive beliefs about rumination predicted engagement in rumination that, in turn, predicted PPD, both directly and indirectly, through negative beliefs about uncontrollability and the social consequences of rumination. A cognitive model with ruminations as a partial mediator between negative postpartum thoughts and PPD symptoms also had a good fit. Conclusion The findings of this study contribute to the understanding of the cognitive and metacognitive mechanisms underlying postpartum depression, which might be similar to depression in general and have important implications for treatment strategies.