The role of regulators as gamekeeper to the bankers’ poacher has led to an overly rigid definition of risk-based measure which most banks are adopting through programs such as FRTB. The widespread consensus that the regulator knows best eliminates divergence of views on how to measure risk. This conformity reduces the role of independent oversight that challenges firms to adapt to new market conditions and opportunities – a situation that cannot be good for financial stability.
As Risk Managers, we did not cover ourselves in glory during the last crisis. It should not mean that the intelligence of the few in the guise of the Basel Committee should overrule the diversity of a broad set of practitioners. The Basel committee has a depth of quality, however, directing policy from behind the frontline is not optimal.
The Leverage Ratio (LR) under Basel III was meant to act as a backstop capital requirement to complement more sophisticated RWA calculation and was based on a non-risk sensitive risk exposure calculation, initially using the Current Exposure Method (CEM). The punitive nature of the CEM calculation led to the significant over-estimation of low-risk exposure and hence created the perverse incentive whereby banks would prefer to invest in much riskier assets.
The reliance on a single prescribed model for the calculation of both the LR capital requirement as well as CCR for those institutions not running on internal models presents a single point of failure. All models contain weaknesses, and by enforcing universal usage of model prescribed by the regulators, all institutions are becoming exposed to an approach that is not subject to internal and independent challenge.
Under FRTB (“Fundamental Review of the Trading Book”), the regulators are pushing the fullback sensitivity-based approach (SBA), a standard model with a detailed prescription of sensitivities and term structure together with a pre-canned set of correlations. While conservative, SBA lacks granularity so that the Front Office compress their risk to create a view that the regulators recognise but not the one used by practitioners.
The conditions for the use of internal models (IMA) are so high that we believe the vast majority of banks are not willing to spend the resources to move beyond SBA, leading to a net move backwards from aligning FO risk with capital. Even IMA desks are subject to an SBA floor that further reduces the attractiveness of using internal solutions.
By enforcing a singular model for risk exposure, regulators have diminished the benefits from a diversity of modelling approaches. There is a natural resilience in an industry using different approaches which encourages risk transfer between dealers, but it also encourages banks to carry the risk they are most comfortable. The SBA lacks granularity to differentiate between different markets, thus repeating the adverse incentives of the LR.
Such diversity leads to certain banks being stronger than others, with some of them possibly failing.
A series of QIS (Quantitative Impact Studies) carried out by global regulators collected risk measurement results from the largest banks. The results highlighted that estimation of potential losses varied by orders of magnitude that convinced regulators there were aspects of risk missing from specific models — however, operational issues where practitioners start with a conservative view of risk based on sparse data have hampered the effectiveness of the results.
By concluding that the only path is to force all banks onto a standard model, the regulators are exposing an entire industry away from weakness from a single failing bank to singular weakness shared by all banks.
A weary acceptance by the banks to swallow the hike in capital charge and complacency from the regulators brought on by a lack of challenge creates systemic vulnerability and unknown consequences.
These vulnerabilities hide in the medium-term in an environment where governments have underwritten the markets with borrowings at near-zero rates. However, with rates increasing the seeds of model weaknesses are sprouting.
Given the Basel committee sit remotely, the weakness can hide within the rollout of FRTB (2022-2025) and expand unchecked. At some point, with history repeating itself, the industry is likely to suffer the consequences of single model failure.