The Self-Organizing Financial Stability Map (SOFSM) can be a valuable tool for mapping the state of financial stability and visualizing possible sources of systemic risk. Beyond its visualization capabilities, the SOFSM can be used as an early warning model that can be adjusted to accommodate policymakers’ preferences so they do not miss a systemic financial crisis or issue a false warning.
The recent global financial crisis exposed the importance of understanding and detecting potential sources of domestic and global vulnerabilities that may build up and lead to a systemic financial crisis. With this in mind, the authors’ goal is to present and promote a greater awareness of mapping techniques and financial stability surveillance. Mapping techniques in financial stability surveillance are demonstrated by presenting a methodology for mapping the state of financial stability and visualizing potential sources of systemic risk.
How Is This Research Useful to Practitioners?
The Self-Organizing Financial Stability Map (SOFSM) provides a technique for financial market participants and macro-prudential policymakers to better understand, through visualization, the state of financial stability and the potential sources of domestic and global vulnerabilities that may lead to a systemic financial crisis. The authors’ research supports the idea that the SOFSM can be used as an early financial stress identification tool that could allow policymakers to target their actions to reduce or avert further buildup of susceptibilities in the financial system.
The SOFSM introduces the concept of the financial stability cycle. As such, the SOFSM can be used to monitor macro-financial exposures by determining a country’s status in the financial stability cycle (pre-crisis, post-crisis, or tranquil state). The authors describe how the SOFSM can be used to evaluate the potential for contagion through similarities in macro-financial conditions across countries and to visualize scenario analysis results.
The SOFSM performed well in classifying in-sample data and predicting out-of-sample the global financial crisis that began in 2007. It is noteworthy that the SOFSM makes an out-of-sample prediction identifying the financial crisis as early as the first quarter of 2006, indicating the potential efficacy of the model for financial stability surveillance.
The authors highlight that although the model is evaluated as an early warning model, the SOFSM should be viewed as a complementary tool and should be treated as a starting point rather than an ending point for financial stability surveillance.
How Did the Authors Conduct This Research?
The authors’ objective is to promote a greater awareness of mapping methods in the field of finance among the policymaking community in general and financial stability surveillance in particular. Their aim is met by demonstrating the use of mapping techniques in financial stability surveillance through a methodology based on data and dimensionality reduction for mapping the state of financial stability and visualizing potential sources of systemic risks. The authors’ research provides a basis for a wide scope of applications of mapping techniques and establishes the necessity for further research.
The authors begin with an explanation of the five elements that are required for constructing the SOFSM: (1) data and dimensionality reduction based on the self-organizing map, (2) identification of a systemic financial crisis, (3) macro-financial indicators of vulnerabilities and risks, (4) a model evaluation framework for assessing performance, and (5) a model training framework.
The authors describe how a systemic financial crisis is identified and how the class variables are defined for enabling assessment of the financial stability cycle. The identification of a systemic financial crisis consists of a database of systemic events and a set of vulnerability and risk indicators that include a quarterly dataset of 10 advanced and 18 emerging economies for the period of 1990–2011. The data are retrieved from Haver Analytics, Bloomberg, and Datastream.
The authors follow with SOFSM training and evaluation, robustness checks, and details regarding how the SOFSM can be used for spotting signs of vulnerabilities and potential for contagion, in addition to mapping the state of financial stability over time, across countries, and for different levels of aggregation. Lastly, they show how the SOFSM can be used for visualizing scenario analysis results.
The global financial crisis uncovered the need to better understand systemic risks that underlie the financial system as a whole. But the complexity of the potential sources of vulnerability makes identifying larger trends in the system difficult. Macro-prudential policymakers, regulators, and central bankers could benefit from using mapping techniques, such as the SOFSM, as financial stability surveillance tools. Visual mapping models that provide early warning signals of systemic risks would potentially allow policymakers to mitigate the harm from cycles in the financial system by the timely implementation of targeted policy actions.