Significance and limitations of the VAR figures publicly disclosed by large financial institutions

The value-at-risk (VAR) figures publicly disclosed by the major banks provide useful information on the market risk taken by the banking system. But a number of methodological precautions need to be taken when analysing these figures. These are detailed in this article. The assumptions underlying VAR calculations can, indeed, differ from one institution to the next and are rarely very explicit. The conclusions to be drawn from these figures should therefore be treated with caution. The information provided by disclosed VARs should be corroborated by other indicators and analysed against the general macro-fi nancial backdrop. When they point to greater risk exposures, these figures act as warning signals for conducting more in-depth analyses of the vulnerabilities that could affect financial stability. VAR figures disclosed by financial institutions are closely monitored by central banks and are, for example, often discussed in the overview of the Banque de France’s Financial Stability Review. These figures have contributed to supporting our assessment of a rise in market risk exposures at the time when short-term interest rates were uniformly very low, before the Fed started raising its key rates, which called for heightened vigilance even though the macro-financial context appeared favourable. Central banks’ constant concerns about greater transparency on the part of financial institutions resulted in the publication of the Fisher II report. In this framework, one way of strengthening financial stability would be to encourage credit institutions to be more transparent as regards their methods for calculating disclosed VARs. Credit institutions could, for instance, include more precise and easily comparable methodological explanations in their annual reports. This should not prevent the leading banks from setting up more sophisticated risk management techniques nor impinge on their communication policy. By authorising banks to use – subject to validation – an internal ratings-based approach for calculating their regulatory capital requirements, banking supervisors have actually acknowledged at the international level the diversity of markets and operations carried out by banks. This diversity implies adopting calculation methods tailored to the specificity and management techniques of each bank. Banks should in fact provide analysts with the most relevant information possible. But, in view of current practices, the level of transparency as regards the methods used is still insufficient. In addition, financial institutions other than banks, for example hedge funds, should also be encouraged to disclose their VAR figures. Lastly, financial institutions should disclose their stress tests on a more regular basis, as a methodological complement to the VAR figures, but also in order to prevent a potential homogenisation of behaviours which could result from an exclusive use of VARs in banks’ communication strategies.