New research develops a model to measure risk of European banks (Reuters)
When Jean-Claude Trichet, ex-president of the European Central Bank, reflected on the financial crisis, he was at a loss to explain how no one saw it coming.
Although intelligent, widely experienced and decorated with countless honours, Trichet was a major official who simply failed to see the financial tsunami that engulfed Europe.
Therefore it is fitting that a research paper from the ECB has been published that tries to address the intellectual blind spots of policymakers and economists.
It is called Predicting Distress In European Banks and is authored by Frank Betz, Silviu Oprică, Tuomas A. Peltonen and Peter Sar.
They have developed an early warning model to predict vulnerabilities that could lead to further stresses in European banks.
The paper is significant as it could inspire a new branch of research that is needed given the depth and breadth of the eurozone's sovereign debt crisis.
The paper focuses on two key issues, the main sources of bank weakness and to what extent indicators in the model can reveal these vulnerabilities.
The authors surveyed the academic literature on banking failures and observed most of it was centred on the United States.
"Most papers analysing individual bank failures or distress events focus on US banks or a panel of banks across countries, while there are only a few studies dealing with European banks," they write.
Although recent studies have begun to examine Europe's banking problems, they have "no focus on the entire European banking system, in particular the core European countries" which the paper attempts to correct.
Methodology And Goals
To get around these pitfalls, the model takes a "signal" approach to detect banking failures.
The model tries to resolve what it calls the tension between "type 1" and "type 2" errors in detecting and warning about banking problems.
Missing a banking crisis is the former and false alarms are the latter.
The authors' model tries to minimise "the noise-to-signal ratio, given by the number of false alarms relative to the correct calls" when policymakers confront weak banks.
The research relies on a rich sample of information from balance-sheet and income-statement data from Bloomberg.
Some 546 banks with a minimum of €1bn in total assets are analysed in the paper.
Their timeframe starts in the first quarter of 2000 and ends in the second quarter of 2013.
The banks sampled are in all of the European Union nations except Estonia, Cyprus, Lithuania and Romania.
The paper finds that a policymaker who deals with a banking crisis should be more concerned about missing the fact the bank is threatened rather than issuing a false alarm.
Most importantly, the authors claim their results show that their model can predict failures in the European banking system.
They write their "early-warning model based on publicly available data yields useful out-of-sample predictions of bank distress during the global financial crisis".
Furthermore, the authors conclude that their results confirm the usefulness of the vulnerability indicators introduced recently via the EU Macroeconomic Imbalance Procedure (MIP).
They find that the more accurate the risk measures are, the better inputs there are into policy to encourage banks to take actions to solve their problems rather than rely on governments.
However, the authors acknowledge that there are a number of factors that are "beyond the scope" of the study and do not factor in their model.
These are described as "political economy aspects" or the way politics can frustrate a lawmaker's ability to warn that a bank has problems due to political reasons.
When a policymaker knows it is the correct time to raise the alarm about a bank in distress but might not want to due to political circumstances can compromise the measure in any model, said the authors.
The paper does not represent the views of the ECB.
To report problems or to leave feedback about this article, e-mail:
To contact the editor, e-mail: