The Long FRTB Road Ahead

Jan 30, 2017 Share this! LinkedIn logo Facebook logo Twitter logo Reddit logo Google+ logo
Mon, 30 Jan 2017 09:30 America/Chicago

Banks have already started to lay the foundations but there will be many twists and turns ahead. DerivSource reporter, Lynn Strongin Dodds looks at the obstacles to overcome.

Although the Fundamental Review Trading Book (FRTB) does not see the light of day until 2019, banks especially the tier one players have cranked into gear. Projects have moved from the initial quantitative blueprints into actual design and implementation phases but challenges lie ahead not only in the restructuring of systems but also the higher than expected outlays.

In other words, implementing the FRTB is a bigger undertaking than was forecast twelve months ago when banks were in their early stages of planning.  This is perhaps not surprising given that new IT architecture and risk systems that are required to enable trading desks to manage and monitor risk in a much more detailed way.  They will be obliged to calculate their capital requirements using the internal model approach (IMA) or the more punitive standardised approach (SA).

The latest figures reported from Oliver Wyman, which assessed the plans of 20 European, US and Asian banks with large trading businesses found that they will have to shell out between $100m to $200m compared to the $43m to $129m originally forecast by the consultancy.  Putting this is into context, these numbers still represent a small percentage of their total annual cost base of around $60 bn.

The industry wide expenses are more eye watering with Oliver Wyman reportedly estimates implementation to be about $5bn, including over $500m that has already been spent, mostly on planning and impact studies.  The firm predicts banks will have to add 2,000 new staff to their FRTB programmes, through a combination of hiring new people, reassigning existing employees and using consultants.

This is likely to trigger further restructuring. Many banks have already streamlined their employee ranks, culled unprofitable business lines and re-focused their attention on their most profitable clients due to Basel III’s more onerous leverage and liquidity constraints. However, under the FRTB,  there could be further consolidation among smaller bank trading operations because they may decide that the costs of FRTB and, in particular, using the internal model approach for smaller operations is too high and does not outweigh the capital relief benefits,” says Martijn Groot, VP Product Management at Asset Control. “For example, this may lead to a shrinking number of banks trading certain types of derivatives.”

One solution to ease the burden would be to build large tick data bases to be shared throughout the bank or asset management firm, according to Ian Hillier-Brook, CEO of MCO, McObject European partner. “This will take time to develop because under FRTB the reporting is at the desk level. At the moment what we are seeing is some of the large banks running the projects internally, and some working with consulting and software partners while the smaller ones are looking at cloud based solutions,” he says.

The modifications will vary according to the models they select. Overall, industry estimates show the capital bill for a SA model could be from two to 6.2 times higher than current levels while the revised IMA could result in a smaller 1.5 times hike. While it may be tempting to opt for the IMA, there are difficult hoops to be jumped through including a P&L attribution test which is still being clarified as well as back testing.

Regulators are still debating whether a more stringent version of the P&L attribution test, proposed in the glossary of the final rules, should apply over an easier-to-pass draft in the main text. 

Although some banks are waiting for the final edict, the larger ones are ploughing ahead because regardless of the outcome, the new rules will require “significant changes to the IMA and a complete rewrite of the calculations required under the standardised approach,” says Nicola Hortin, Head of EMEA Regulatory Analysis at AxiomSL. "In addition, regardless of whether they currently use the standardised or internal models approach, banks will need to review their portfolios to determine if existing classifications of desks as trading or banking book are still applicable under the new trading book definitions or whether a revision of desk structure is required.” 

She adds that banks using the IMA model will need approval at desk level for the first time but they will also be required to implement the SA for desks not gaining approval or for desks which fail the tests and are forced to switch from IMA to the standardised approach.

Sven Ludwig, MD risk and analytics EMEA, Institutional and Wholesale. FIS, notes the decision will clearly depend on the business of an individual bank. “The use of IMA will be a pure business case with banks asking if they can save capital and whether this justifies the cost.  We conducted a survey at the end of last summer and found that IMA is still on the radar screen.”

Fraser Schad, Senior Strategist, Financial Risk Product Management, IBM Industry Solutions, agrees adding, “some smaller banks may choose to accept the higher capital charges that come with sticking to the standarsed approach, and may only be looking for an efficient FRTB reporting solution.  “Most large banks will have a much stronger business case under FRTB to rethink their operations, and will take the opportunity to rebuild a much more modern infrastructure from the ground up that will reliably serve the bank in many other ways,” he adds.  

Flexibility is key, according to Zeshan Choudhry, Partner, Risk and Regulation, Capital Markets at Deloitte.  “Over a number of years, banks have had to deal with multiple technology systems influenced by business size and type,” he adds. “Meeting regulation as an individual initiative is not sustainable and as a result we are seeing senior executives look at regulation from an aggregated perspective to deal with compliance effectively and strategically. FRTB is quite prescriptive and should be implemented strategically, which would place banks in a good position to meet additional future regulation.” 

On the big picture level, Srikant Ganesan, Head of Risk and Trade Solutions at Risk Focus identifies four main challenges of FRTB. They range from the acquisition of trade and market data from multiple sources, to cleansing, validation and aggregation of that data to support computations and visualization of the results.

Moreover, the data required will have to be much more granular and complex. This is particularly true with the new calculations under the SA models. "For example, banks will need to calculate deltas and vegas for prescribed risk factors and will also need to run both downward and upward shock stress scenarios in order to provide the data required as inputs into the standardised calculations,” says Hortin.

The non-modellable risk factors (NMRF) framework will also force trading desks that use an internal models-based approach to be more detailed. They will need to include a capital add-on for each risk factor that is supported by what is considered insufficient data. It requires risk factors to be based on data from at least 24 representative transactions in a given year, with a maximum of one month between two consecutive trades.

One solution to ease the burden would be to build large tick data bases to be shared throughout the bank, according to Hillier-Brook. “This will take time to develop because under FRTB everything is at the desk level. At the moment what we are seeing is the large banks running the projects internally while the smaller ones are looking at cloud based solutions.”

Ludwig notes that “larger banks have the resources and are focusing on the internal model approach. One reason is that the   risk is unique and they don’t want to be vulnerable to outsourcing. Also strong risk management can be seen as a key differentiator going forward.”

Sadiq Javeri, Head of Product Strategy, Capital Markets and Enterprise Risk, Misys, notes that some of the bigger players who have outdated technology may hand over to a third party vendor to fill in the blanks. Overall, “what we are seeing is that it is only some of the very largest global banks that are committed to implementing this significant regulation entirely in-house, while the remainder are exploring options to outsource given their own internal resource constraints, execution complexity and risks.”

Schad echoes these trends and says some larger banks are ahead of the curve, having actively invested in modernising their risk analytics infrastructure over years, and are only looking for a vendor to provide an FRTB reporting solution.  “However, most mid-to-large banks with internal model approval recognise the need for a fundamental upgrade in their analytical approaches to be capable of running both the revised standardized approach and internal models in parallel,” he adds.

Groot also believes that banks will move to cloud based solutions but envisions utilities as part of the solution. “We could see the increased data and reporting requirements of regulations such as FRTB lead to more shared middle office capital markets infrastructure if there are enough banks with similar requirements who are willing to participate,” he says. “We have seen this work already within countries such as Denmark, where BankData is owned by 11 Danish banks. New regulation could move such services to an EU or global level.”

Article Type: