Financial cycles, credit networks and macroeconomic fluctuations: multi-scale stochastic models and wavelet analysis

This project focuses on the macroeconomics of financial cycles. Usually defined in terms of self-reinforcing interactions between perceptions of value and risk, attitudes towards risk and financing constraints, which translate into booms followed by bust, the recent empirical literature has recurred to two approaches – turning point analysis and frequency-based filters - applied to measures of credit and asset prices to pose a number of stylized facts. First, financial cycles tend to display a greater amplitude and a lower frequency in comparison to business cycles, with peaks associated with systemic crises. Second, financial cycles depend on policy regimes and on the pace of financial innovations, leading to a wide cross-country heterogeneity and a time-varying degree of global synchronization. The latter point is clearly related to the structural transformations occurred in financial systems over the last three decades, like the cumulative integration of traditional banking with capital market developments and the increasing degree of interconnections among financial institutions. However, to date very little is known about determinants and mechanisms behind financial cycles, and on how they interact with business cycles and medium-to-long-run macroeconomic performance. In this project we plan to research along three dimensions: i) measurement issues, in order to provide a comprehensive assessment of the evolution of co-movements between financial and real variables across a sample of financial developed countries, both over time and at different frequencies; ii) theoretical issues, aimed at exploring under what circumstances the network of interconnections among financial intermediaries and between intermediaries and non-financial borrowers might evolve cyclically, contributing this way to regulate the incentives agents have in taking risks, and to set the importance of credit and financial frictions in accounting for time-varying misallocations of resources; iii) policy issues, given the role assigned by international supervisory bodies to a proper characterization and knowledge of the financial cycle as a prerequisite for the macro-prudential regulation of banks, and the scope of monetary policy in promoting financial stability in addition to the typical mandate of price stability. Our task requires the employment of a new approach to macroeconomic analysis, diverse analytical tools and one unifying economic principle. As regards the latter, our focal point is the notion of risk externalities, across financial institutions and between the financial sector and the real economy. The set of tools we plan to employ spans from wavelets methods to multi-scale models in continuous time, and from strategic network formation to agent-based computational techniques. All these tools are instrumental in building and estimating macroeconomic models characterized by interrelated markets operating at different time scales.