We analyse firm survival and focus on several levels of analysis (both firm level and macro-level). We employ a unique longitudinal data set, recorded at the firm-level and covering nine complete entry cohorts of Swedish companies. The companies were founded between 1899 and 1992, and each firm is followed over nearly a decade. We adopt the semi-parametric complementary log-log (cloglog) model. The main novelty of our approach is that, unlike extant studies so far, we are able to distinguish between the impact on the hazard rate of founding conditions and contemporaneous, post-entry conditions. Using our new approach we test several hypotheses derived from the Industrial Organization and Organizational Ecology literatures.