Abstract
Edson Bastos
University of São Paulo (USP)
Systemic Risk is frequently studied under the perspective of the financial sector. Nevertheless, during the financial crisis of 2008, the real sector also showed to be an important driver of systemic risk. Good examples are the bailout of General Motors and FSOC (Financial Stability Oversight Council) designating non-bank SIFI (Systemically Important Financial Institutions). However, having a suitable mapping of both sectors under the same risk measure involves dealing with multilayered networks, requiring data that is not usually available to researchers. This study proposes a framework linking contagion in the real sector with contagion in the financial sector. We rely on a unique data set which accounts for not only networks in the Brazilian system of payments and the Brazilian interbank exposures but also data from the Federal Revenue of Brazil, Ministry of Labor and Employment, Brazilian derivatives clearings, and banks fillings comprising accounting and credit risk exposures. This very comprehensive data allows a better understanding of the dimensionality problem of systemic risk and highlights how risk flows from the real sector to the financial sector. We provide two empirical examples, describing some properties of the networks considered and measuring the impact of systemically important corporations on the Brazilian financial system. We conclude with the remaining data gaps and open challenges.