Psychotherapy and medication treatments are both effective in reducing depressive symptoms. However, only psychotherapy provides an enduring effect by reducing depressive vulnerability following treatment termination. This differential efficacy may reflect mode-specific effects on the longitudinal relationship between depression and stress. The current study examined post-treatment data from 153 outpatients enrolled in the Treatment of Depression Collaborative Research Program. Longitudinal analyses using the latent difference score (LDS) framework (a structural modeling technique that combines features of latent growth curve and cross-lagged regression models) evaluated the temporal relationship between severity of depression and frequency of stressful life events, assessed by interviewers at treatment termination and at 6, 12, and 18 months following treatment. Results supported a stress reactivity model in that stressful events led to elevations in the rate of depression change. Furthermore, multi-group LDS analysis indicated that this longitudinal stress reactivity occurred only for outpatients in the medication conditions. Results demonstrate that the enduring impact of psychotherapy involves the development of enhanced resiliency to stressful life events.