Data management and business complexity are now fundamental drivers of RWA and returns
Basel IV (FRTB): It’s All About Data, Pt. 2
Two weeks ago, we touched on a few points where data practices will be impacted with the introduction of FRTB. Now, we will continue that discussion and delve even further into several areas where business practices must adapt to reach compliance.
Lack of Common Data Could Impair Performance
With FRTB, the Risk and Finance department structure and how data is managed will likely change. Most banks currently remain highly siloed and rely on vertical technology stacks. This inevitably results in a lack of common data and the ability to create and derive various risk factors.
Risk and Finance are the primary areas where this data needs to be aggregated, but they need to coordinate data capture to ensure consistency and reduce reconciliations. This slows down RWA production, which in turn drives up capital requirements, impairing business performance and returns.
Data storage and replication needs to be kept to a minimum as the more times data is copied and ‘touched’ the more likely it is that its traceability will be lost and it will need to be reconciled. FRTB now introduces a potential capital cost to this inefficient way of working.
A Sensitivity Approach Necessitates Data Management Upheaval
For smaller banks currently using a simplistic standardized approach, moving to a sensitivity based approach might be difficult if they lack the appropriate models that provide optimal sensitivities. The current standardized approach consists of balance sheet based inputs, such as mark-to-markets and notionals, which are risk-weighted according to asset class and maturity.
Since the new standardized approach is sensitivity based, it presents an immediate data capture and storage challenge that requires new data management capabilities. This is something that can be done in a Data Lake-type big data store. Using different technologies could facilitate any transformation and enrichment. In addition to this, the introduction of a market sensitivity based model will see a shift of ownership within banks from Finance to Risk.
In some cases, the introduction of FRTB will force banks to review the product set that they trade and could be become a barrier to entry for smaller players. Even larger players face additional overheads in terms of calculations and data management if they have to maintain a huge library of calculation models and data sets sufficient to support multiple desks and products.
All firms impacted by FRTB will have to weigh the pros and cons. Desks trading multi-asset products, index products and those with non-linear (curvature) risks are likely to face higher capital charges. Complexity also implies that more risk factors will be non-modelable and push up regulatory capital costs.
Of course these could be offset through investment in more sophisticated models and data management but if the Committee sets the RWA floor at around 70 to 80 percent of the standardized approach then there is a limit to how much an the IMA can model in order to reduce RWA.
FRTB will almost certainly push up capital charges for most banks. The amount will depend on each bank’s individual business mix, but 25 to 30 percent may be realistic. As such, each desk will be assessed on its RWA consumption and this will impact market activity and pricing.
Tackle FRTB Data Standards Head-on
Banks need to start assessing their current desk structure and determining which desks can qualify for the IMA and whether an appropriate model can be used and will deliver a stable performance. Analysis need to be done to determine what proportion of the risk factors are modelable. Any data gaps must be reduced. This leads to another data, as individual banks may not be able to ascertain model-ability if they only have their own transaction data to rely on. This opens the way for banks to start collaborating and sharing data via some type of data utility or data exchange.
To wrap it up, FRTB or Basel IV places business complexity and data management at the center of the market risk capital framework. A sub-optimal desk structure, complex products, data gaps, and poor data management will all lead to higher RWA and therefore capital requirements. Business managers will want to assess desks in terms of their return on RWA and allocate capital accordingly. Those desks that don’t achieve internal mandates and hurdle rates will not survive. There needs to be much more focus on inputs into risk models and issues such as data lineage, trusted sources, and single data dictionary will now become financial KPI’s and MI.