To further safeguard against another financial crisis, the Bank of International Settlement (BIS) introduced Financial Review of Trading Book (FRTB) guideline to ensure more current and accurate valuation by banks of their banking and trading books. Banks must comply with FRTB by the end of 2019.
- Banking and Trading Books: FRTB will reduce a bank’s ability to reduce capital reserve requirements by moving assets to Banking Book, and using the potentially less accurate Mark-to-Model to value them.
- Higher capital requirements: Expected Shortfall (ES) replaces Value At Risk (VaR) for measuring long-tail risk during periods of extreme economic stress. Under ES, the Basel Committee estimates that total market risk capital requirements will increase by a 40% weighted average. FRTB also requires more comprehensive data sources to calculate ES.
- Internal Model / Standard Approach – Banks using the Internal Model Approach (IMA) now require compliance enterprise-wide at the trading desk level rather than at an entity level. Along with revised guidelines on determining sensitivities to risk factors in the Standard Approach (SA), FRTB enforces a more consistent way to measure Market Risk enterprise-wide.
- Real, Live Transaction Data – Requires real-time data to be readily available for calculations.
- Risk Factors and Data Adequacy – The FRTB regulation introduces Non-Modellable Risk Factor (NMRF). A NMRF is one where real, continuously available prices (general defined as those with 24 or more observable prices from an independent 3rd party over the current ES period with a maximum lag of one month between and two consecutive observations) are NOT available. For NRMF, firms may need to use longer time horizons that results in higher capital reserve requirements.
- More Time Horizons for Calculations– FRTB augments 10-day time horizon for calculations with 20-, 60-, 120-, and 250-day horizons.
Impact on IT and development
Under FRTB, banks will need more compute and storage capacity given the multiple time horizons, greater model complexity, and larger data sets. This need is not surprising. FRTB along with a whole host of other regulations like General Data Protection Regulation (GDPR) demand greater operating transparency, accountability, and reporting which translates to more analyses of more data more often.
At the same time, bank ITs’ shrinking space, budgets, and manpower is accelerating adoption of open source technologies and architectures for analytics on-prem and in the cloud. With that in mind, the following technologies have or are proving to be useful in IT’s quest to provide the requisite analytical capabilities to their customers.
- Late majority: GPUs, FPGAs, flash storage, high performance file systems; private cloud
- Early majority: Distributed computing; hybrid cloud computing when applicable including bursting and cloud-based API development; data lifecycle management; 1st generation big data
- Early Adopters: Software defined infrastructure; 2nd generation big data; and containers
- Innovators: AI/DL, and blockchain
These more agile, modern technologies offer the scalability, openness, and efficiency to support FRTB and whatever new regulations that may appear in the future. Technologies such as software defined infrastructure provide greater agility and efficiency for hybrid cloud and/or Hadoop-based environments. High performance multi-tenant grid computing, and secure cloud bursting deliver scalability and cost savings for compute-heavy risk simulations. While data-aware job schedulers and life-cycle management tools can optimize data movement (minimizing latency impact), access, and storage choice.
To learn how IBM can provide you with the ability to access and manage data more efficiently, click here.
To learn how IBM simplifies and accelerates analytics of all kind, click here.