Pré-publication, Document de travail: The Swedish banking crisis in the early 90s counts as one of the five most severe financial crises in history. We examine how firms more exposed to this event adjusted employment in the longrun and the mechanisms involved. Our analysis draws on matched employer-employee data containing the financial statements for a large sample of firms. Our difference-indifferences estimates show that firms with a greater pre-crisis debt burden experienced more difficulties in accessing external capital during the crisis compared to firms with lower baseline debts. This is consistent with the most exposed firms becoming financially constrained. More exposed firms exhibit stronger downward employment adjustments than less exposed firms, and the reductions are mainly concentrated among low-skilled workers. Employment in more exposed firms started to recover four years after the crisis and had fully recuperated about a decade later. These firms also temporarily saw a larger drop in both productivity and investment. We do not find a significant effect on the wage bill, and the estimates are precise enough to rule out even moderate effect sizes.