III Course on Financial Stability

The Course was held on a digital format on October 25-29, 2021, and was attended by 92 representatives from 28 institutions and associates of CEMLA from the following countries: Bolivia, Chile, Costa Rica, Honduras, Argentina, Dominican Republic, El Salvador, Peru, Ecuador, Uruguay, Brazil, Spain, Guatemala, Colombia, Mexico, Morocco, Canada, Jamaica, Mauritius, République d'Haïti, Belize, Curaçao en Sint Maarten, Suriname, Ukraine, Turks and Caicos Islands. The event was focused on the following topics: introduction to financial stability analysis and systemic risk, foundations of macroprudential regulation, panel data analysis, financial safety nets, banking resolution and supervision, interconnectivity, stress testing and COVID-related challenges.

 

Session 1: Introduction to financial stability analysis

Financial stability analysis and systemic risk.
This session described key concepts of financial stability; risk and vulnerabilities, and the relevance from a macro-financial approach. It is important to organize and structure the large amount of information into certain indicators to use them effectively. Finally, empirical evidence shows a significant relation between economic and financial cycles. The main take-aways of the session included:

• - Financial stability can be defined as a state of the financial system with a limited probability of occurrence of distortions and perturbations in the financial intermediation process that can impact negatively to the real economic activity.
• - The systemic risk indicator (SRI) includes twelve individual stress indicators and takes into account cross-correlations among the four market segments (monetary market, public debt, stocks and financial intermediaries).
• - The use of macroprudential tools to moderate the financial cycle (ie. the activation of the Countercyclical Capital Buffer, CCyB, can avoid excessive credit growth and reduce the probability of crisis episodes).

Foundations of macroprudential regulation I.
This session discussed some introductory issues in macroprudential regulation and the evolution of the regulation pre-2008 and post-2008. Some of the drawbacks of micropru regulation were ignoring systemic risk and spillovers effects. Capital, liquidity and borrower-based measures are some of the instruments use in the macropru policies which its use can be heterogeneous across countries (advanced vs emerging economies). The main take-aways of the session included:

• - Regulation addresses market failures in the financial system (asymmetric information, adverse selection and market externalities).
• - Microprudential regulation is limited.
• - International coordination is important to ensure macroprudential policies’ effectiveness.

Foundations of macroprudential regulation II.
This session explained the development of macroprudential policy instruments and the importance of looking at a macro picture. There are different dimensions of macropru policy: cyclical (time dimension) and structural instruments (size, complexity and interconnectedness of banks). Some of the capital-based tools are the credit-to-GDP gap defined as a benchmark guiding indicator for setting-up the CCyB. The main take-aways of the session included:

• - Buffers address the adverse effects on financial stability of large and complex banks.
• - Borrower-based instruments include the lending standards and default probability.
• - Reciprocity mechanisms are fundamental to limit leakages of macroprudential measures.

 

Session 2: Panel data for financial stability analysis

Panel data analysis in financial stability using microdata.
This session discussed the microeconometric settings, common estimators and the different techniques such as OLS, Difference-in-difference, and instrumental variable approach. Some of the challenges and best practices for researchers on financial stability. Some of the main take-aways of the session included:

• - Fixed effects remove unobserved heterogeneity. Clustered standard errors account for situations where error terms are not i.i.d.
• - For a Dif-Dif design, we have to take into account an exogenous sizeable event from pre-existent market dynamics and to verify a large cross-sectional variation that justifies the approach.
• - IVs are used when an explanatory variable of interest is correlated with the error term.

Applications of panel data for policy evaluation.
This session described the applications using different micro-data and empirical methods such as Dif-Dif, IVs and regression-discontinuity design for panel-data settings for financial stability analysis. There are different data sources, in particular, it was described some of macroprudential policy datasets, banking-sector data with different degree of granularity. Some of the main take-aways of the session included:

• - Define the research question is key, data and method are auxiliary tools to researcher main objective.
• - Address a general interest question (not country-specific).
• - Define a research design to isolate a causal relationship.

Applications using credit-register data for policy evaluation.
This session explained the differences between randomized and natural experiments. A randomized experiment can estimate the average treatment effects among individuals (random sample), whereas the natural experiments use real-life situations to work out effects on the world. Finally, the session showed the use of granular credit-level data, in particular, using data from the Spanish Credit Register. Some of the main take-aways of the session included:

• - Causal inference is one of the important topics in economics.
• - Identification strategies are defined as the resolutions to the fundamental problem of causal inference.
• - There are areas where randomized experiments are either impossible or impractical to do at scale.

 

Session 3: Financial safety nets, banking resolution, and supervision

Identification and supervision of weak banks.
This session identified the regulation weaknesses during the financial crisis and how the regulatory answer to the crisis. Also, it was explained the supervisory mechanism and the monitoring of weak bank based on qualitative and quantitative elements and an Emergency Action Plan (EAP) in the ECB. Some of the main take-aways of the session included:

• - The need of an intensification of banking supervision.
• - The supervisory judgement has a key role. The quantitative and qualitative factors need to be considered.
• - Supervisory approach needs to be flexible to identify future risks.

Institutional foundations of banking resolution.
This session explained the definition of resolution, its objectives and triggers. Also, it describes the steps for the financial stability assessment which are the systemic relevance, direct contagion, indirect contagion, maintaining market discipline and the effect on the real economy. Some of the main take-aways of the session included:

• - Public interest assessment (PIA) is key in resolution and determines the Preferred Resolution Strategy.
• - Single Resolution Board uses a wide range of financial-econometric tools and models to perform the PIA.
• - In resolution, the financial stability assessment is a new field of research.

Banking resolution from the perspective of central banks.
This session discussed how the central banks participate in resolution regimes, there are two main models: integrated and separated model. The importance of the interaction and coordination between regulatory agencies and different authorities contribute to systemic risk prevention and financial stability considerations. Some of the main take-aways of the session included:

• - In an integrated model, it is important that the resolution function has sufficient influence.
• - The MPE model is extremely resilient, it limited contagion and contributed to global financial stability.
• - Some specific gaps were identified: sale & merge authorizations fast-track, integration of portfolios on internal capital models of the acquirer, among others.

 

Session 4: Interconnectivity and systemic risk

Introduction to financial networks and systemic risk.
This session discussed an introduction to financial networks and network analysis presenting some types of graphs, topological measures, and other metrics. Also, the interconnectedness of the financial system and the implications for the possibility of cascading failures by measuring systemic importance of participants in these networks with centrality algorithms such as the DebtRank. Some of the main take-aways of the session included:

• - Financial contagion is one of the main components of the systemic risk.
• - The different types of contagion are: default cascades, funding liquidity contagion and the assets fire sales externality.
• - Systemic risk is a consequence of the interconnectedness among institutions in the financial system.

Financial network analysis: applications for financial stability.
This session described the importance of assessing systemic risk through the quantification of systemic risk in multilayer networks for different types of exposures like deposits and loans, security cross-holding, derivatives and foreign exchange. Systemic risk arises from indirect interconnections that occur when financial institutions invest in common assets (overlapping portfolios) as well as from direct interbank exposures (default contagion). Some of the main take-aways of the session included:

• - The need of an intensification of banking supervision.
• - Measuring systemic risk enables better decision making and risk management for financial authorities.
• - The applications show an important underestimation of systemic risk if the information of intra-firm exposures is ignored.

 

Session 5: Stress testing and COVID-related challenges

Introduction to the design of stress tests.
This session explained the key elements of stress tests and the main objectives in crisis and in normal times, which include the recapitalization of banks and risk assessment, respectively. Two fundamental elements to be tested are the financial institutions (banks, non-banks) and exposures (solvency and liquidity stress tests). Some of the main take-aways of the session included:

• - Stress scenarios should be tailored to the objective of the test and need to be tough "enough".
• - Some macroprudential aspects to consider are the interconnectedness, endogenous reactions, contagion, among others.
• - One of the challenges is the way a plethora of new information is communicated properly and how to use them for policy.

Applications and experiences from stress testing.
This session described the architecture of Bank of Spain stress test, where the baseline scenario is aligned with macroeconomic forecasts. Additionally, the main challenges and results of the Spanish experience during the COVID-19 Crisis and the risks associated with climate change. Some of the main take-aways of the session included:

• - Due to the impact on the global economy of the current crisis, it is relevant to estimate the generation of new capital resources in the foreign subsidiaries.
• - Explanatory factors include macro-financial aggregate variables and sectoral gross added value growth and variables related to sectoral financial ratios.
• - The greater granularity, the better results on capturing the effect on credit risk of the different composition of the portfolios.

Panel on financial stability challenges in the COVID Pandemic.
This session described the policy responses to the pandemic in Spain and Germany. Some of the main important measures adopted were fiscal, monetary, prudential and accounting measures. The importance of a coordinate response and authorities’ actions were crucial to avoid and economic and financial disruption. Some financial stability challenges are climate change and digitalization, then we need a resilient financial sector to tackle both of them. Some of the main take-aways of the session included:

• - Vulnerabilities in the financial sector are gathering.
• - Macroprudential policy needs to switch from crisis management to prevention.
• - Monetary and fiscal policy measures prevented a liquidity crisis.