BoE Paper on Machine Learning Approach to Predict Financial Crisis
BoE published a working paper that details the early warning models developed for financial crisis prediction using machine learning techniques on macro-financial data for 17 countries over 1870–2016. The paper shows that machine learning models mostly outperform logistic regression in out‑of‑sample predictions and forecasting. Due to their greater flexibility, machine learning models have the advantage that they may uncover important non-linear relationships and variable interactions, which may be difficult to identify using classical techniques. The results help policy makers to identify the risk of financial crises in advance and potentially act on these signals.
The paper describes the dataset and reviews literature on the variables that the authors choose as predictors. It then presents the benchmark logistic regression, outlines the methodology, and provides a brief description of the different machine learning models applied and the Shapley value framework. Next, the paper compares the predictive performance of all models and investigates the importance of the predictors using Shapley values. Finally, the paper looks at the robustness of indicators in a macro-prudential policy context and investigates in detail the role of the yield curve.
This paper shows that machine learning models outperform logistic regression in predicting financial crises on a macroeconomic data set covering 17 countries between 1870 and 2016 in both out-of-sample cross-validation and recursive forecasting. The gains in predictive accuracy justify the use of initially more opaque machine learning models. All models consistently identify similar predictors for financial crises, although there are some variations across time reflecting changes in the nature of the global monetary and financial system. While the crucial role of credit is an established result in the literature, the predictive power of the yield curve has obtained far less attention as an early warning indicator. The authors also inspect non-linearities and interactions identified by the machine learning models. Global credit shows a particularly strong non-linearity—only very high global credit growth beyond a certain point influences the prediction of the models. Interactions are particularly strong between global and domestic indicators.
Overall, the findings suggest a combination of low risk perception, search-for-yield behavior, and strong credit growth in the years preceding a crisis. The results highlight the potential value of machine learning models for broader economic policy making in two key dimensions. First, the approach illustrates how machine learning techniques can uncover important non-linearities and interactions that facilitate superior out-of-sample prediction and forecasting even in situations characterized by relatively small data sets with limited observations of the event of interest. Second, the novel Shapley value approach demonstrates how the black box concern linked to the practical policy application of machine learning models may be overcome. By providing a mechanism to identify the key economic drivers of the predictions generated by such models, the approach allows insights from machine learning models to be integrated into a broader decision making framework while preserving the transparency and accountability of any resulting public policy decision.
Related Link: Working Paper
Keywords: Europe, UK, Banking, Machine Learning, Yield Curve, Financial Stability, Dataset, Research, Early Warning Model, Fintech, BoE
Related Articles
ECB Finds Banks Unprepared for Pillar 3 Climate Risk Disclosures
The European Central Bank (ECB) published results of the 2022 supervisory assessment of climate-related and environmental risk disclosures among significant institutions (103) and a selected number of less significant institutions (28).
NCUA Assesses Credit Union Exposure to Climate-Related Physical Risks
The National Credit Union Administration (NCUA) released a Research Note that examines the exposure of credit unions to climate-related physical risks. In a related development
EBA Issues Multiple Regulatory and Reporting Updates for Banks
The European Banking Authority (EBA) is seeking comments, until July 31, 2023, on the draft Guidelines on the proposed common approach to the resubmission of historical data under the EBA reporting framework.
EC Adopts Regulation on Own Funds, Issues Other Updates
The European Commission adopted Delegated Regulations on own funds and eligible liabilities, on requirements for the internal methodology under the internal default risk model
CDP Platform to Report Plastic-Related Impact, Issues Other Updates
The Carbon Disclosure Project (CDP) announced that its global environmental disclosure platform has enabled reporting on plastic-related impact for nearly 7,000 companies worldwide
IASB to Enhance Reporting of Climate Risks, Proposes IFRS 9 Amendments
The International Accounting Standards Board (IASB) updated its work plan to enhance the reporting of climate-related risks in the financial statements,
BIS Addresses Data Gaps and Macro-Prudential Policy for Climate Risks
The Financial Stability Institute (FSI) of the Bank for International Settlements (BIS) published a brief paper that examines challenges associated with the use of macro-prudential policies to address climate-related financial risks.
FCA Sets Out Business Plan, Launches TechSprint on Greenwashing
The Financial Conduct Authority (FCA) published its business plan for 2023-24. The plan sets out details of the work planned for the next 12 months to achieve better outcomes for consumers and markets
UK Committee Sets Out Recommendations for Next Phase of Open Banking
The Joint Regulatory Oversight Committee (JROC), comprising the Financial Conduct Authority (FCA) and the Payment Systems Regulator (PSR) as co-chairs and the HM Treasury and the Competition and Markets Authority (CMA) as members
ECB Publishes Multiple Regulatory Updates for Banking Institutions
The European Central Bank (ECB) published the results of the 2022 climate risk stress test of the Eurosystem balance sheet,