The Fundamental Review of the Trading Book has proved monumentally difficult to implement, but there could be a positive consequence around the management of data.
What is happening?
Banks are ploughing ahead with the implementation of the Fundamental Review of the Trading Book (FRTB) — the most demanding aspect of Basel III/IV, despite the disruptions caused by the Covid-19 pandemic.
Indeed, some of the big global banks that are particularly active in trading, especially where it involves more esoteric and illiquid instruments, are spending up to $200m on FRTB.
Most banks are confident that they will be ready in time (January 1, 2023), according to the results of two recent surveys — one by Bloomberg and the other by EY. That deadline was extended by one year last year by the Basel Committee on Banking Supervision, due to the pandemic.
Following several years of grappling with FRTB, banks are gaining a better understanding of how to make their models work and how to tailor their business units to its requirements, fundamentally becoming more effective at harnessing the necessary data.
However, a fork in that road is over deciding whether to use models based on the standardised approach (SA) or the internal models approach (IMA).
Why is it happening?
The dilemma over whether to use SA or IMA boils down to capital consumption. The SA is easier to implement but measures risk more bluntly, meaning higher capital charges against positions. The IMA permits more refined risk modelling and hence lower capital charges, particularly against certain positions.
But using IMA comes at a price. It requires vast amounts of data — and validation — and proving to regulators that it is fit for purpose. Then there is always the possibility of a desk slipping from IMA to SA, leading to higher capital charges. This can happen for the simple reason that there are insufficient verifiable observations of transactions for a particular instrument, like a high-yield emerging market bond.
This has tied some banks in knots over how to proceed. Several have decided to try and implement SA and IMA simultaneously, only to find themselves overwhelmed by the workload.
A more common route is to implement SA first — the rules mandate that it has to be there as a fallback and for benchmarking purposes — before attempting IMA.
However, The Banker understands that such are the requirements or the difficulties in obtaining the correct data, that a significant minority of banks have given up, even if that means closing some product lines. Others are forging ahead, often because they can source the necessary data.
The Canadian banks, almost uniquely, have taken a more collegial approach to the IMA data conundrum, pooling their data to allow for sufficient observations to feed their internal models. This has apparently led to significant capital savings under FRTB and kept some desks in business. Elsewhere, banks are largely shunning data pooling for fear of giving their competitors a capital advantage.
What do the bankers say?
Banks are mostly implementing FRTB through gritted teeth. Not only is it demanding and expensive to implement, but for most it will raise their capital costs for trading certain instruments. The lesser traded, more esoteric ones will not only attract significantly higher capital charges, but under some circumstances may not be able to be netted against similar positions, removing another avenue for curbing capital consumption.
However, so fundamental are FRTB’s data requirements that some banks have spotted a silver lining. Completely reworking their data systems (i.e. breaking down inter-departmental silos, building common data bases, data standards and formats) could yield some welcome efficiencies. For instance, the same pool of data that feeds reporting requirements relating to the EU’s Markets in Financial Instruments Directive II could also populate FRTB models, and so on. It is also nudging firms towards greater automation.
Will it provide the incentives?
Possibly one of the main benefits of FRTB, besides better risk management, is that it is forcing banks to take a more holistic view of their data systems. Apart from giving management better visibility of a bank’s position, it may also enable them to identify and exploit market opportunities more effectively than peers who are begrudgingly ticking regulatory boxes.
Banks need to clutch at those silver linings wherever they can find them and make the most of them, even if they probably never will compensate for the huge costs of the post 2007–2009 global financial crisis regulatory reforms.
