Does Default Pecking Order Impact Systemic Risk? Evidence from Brazilian data

In network models of systemic risk, the loss distribution of a distressed debtor among its creditors follows a pro-rata fashion. It is proportional to the loan granted to the debtor. Despite its simplicity, this assumption is unrealistic. In this study, we create a framework for the computation of the systemic risk assuming a heterogeneous pattern of loss distribution, the default pecking order. Distressed debtors employ some criterion (equity, out-degree, or loan extended) to rank the creditors they are willing to default on first. Applying this framework to an extensive Brazilian data set, we found out the adoption of the default pecking order increases significantly the systemic risk. The rise in the systemic risk brought by the heterogeneous distribution over the homogeneous case decreases with the level of the initial shock and is higher for small-sized agents. This result can be interpreted in the light of the dual role of the financial network, which can be a channel for both risk-sharing and shock propagation. We test this hypothesis by assessing the role of interconnectedness (as measured by the network density) in driving the systemic risk. The results corroborate this hypothesis. When the homogeneous loss distribution (which maximizes risk-sharing) is abandoned, the density has a positive impact on the systemic risk. It suggests in this case the financial network acts mainly as a channel for shock propagation rather than for risk-sharing.