Pré-publication, Document de travail: Why do countries default? this seemingly simple question has yet to be adequately answered in the literature. Indeed, prevailing modelling strategies compel the to choose between two enappealing model features: depending on the cost of default selected by the modeler, either the debt ratios are too high and the probability of default is toot low or the opposite is true. In view of the historical evidence that countries always default is too low or crisis, we propose a novel approach to the theory of debt default and develop a model that matches the key stylized facts regarding sovereign risk.