Two models of collective firm bankruptcies will be presented. The first one uses the Potts spin glass approach for firms rating evolution where two sources of defaults are taken into account: individual dynamics of economic development and ordering interactions between firms. We show that such a defined model leads to a phase transition, which results in collective defaults. In the case when the individual firm dynamics favors dumping of rating changes, there is an optimal strength of firms interactions from the risk point of view. For small interaction strength parameters there are many independent bankruptcies of individual companies. For large parameters there are giant collective defaults of firm clusters. The second model represents defaults of companies in multi-stage supply chain networks. We introduced a modified approach that represents in more details the real economic environment in which firms are operating. We focused on the influence of local processes on the global economic behaviour of the system and studied how the proposed modifications change the general properties of the model. To realistically simulate the economic environment of companies, we introduced the following features to the model: evolution of a supply chain network with the reconfiguration of links, price dispersion and the dynamics of prices and costs of production. At the certain point of the system’s evolution, the meta-stable structures of the network occur. As a result of the dynamics of prices and costs of production, we observed both the emergence of highly profitable supply chains with the high market share and the avalanches of bankruptcies.