With any jigsaw puzzle, it takes time before the full picture starts to become visible. Look at any single piece in isolation, and the picture is unrecognizable. Slot several of the pieces into place, and the image slowly starts to take shape.
A comparison of sorts can be made with the package of capital, leverage and liquidity reforms being introduced by the Basel Committee on Banking Supervision. The Group of 20 (G-20) has set out the picture it wants to end up with: a Basel III framework with an increase in the level and quality of capital banks must hold compared with the pre-crisis Basel II.
But the G-20 has also decreed that any work to refine and calibrate elements of the Basel III rules prior to their finalization and implementation should be made without further significantly increasing overall capital requirements across the banking sector. This is where it’s hard to see how the pieces come together.
The latest segment of the capital jigsaw to be slotted into place is the Fundamental Review of the Trading Book (FRTB), an initiative to overhaul market risk requirements. In its January publication of the final FRTB framework, the Basel Committee estimated the revised standard would result in a weighted mean increase of approximately 40% in total market risk capital requirements. That estimate, though, was based on a recalibration of quantitative-impact-study data from an earlier version of the rules.
As a result, ISDA decided to lead an additional industry study based on data from 21 banks to determine the impact of the final requirements – and the results were unveiled at ISDA’s 31st annual general meeting in Tokyo last week.
The study shows an overall increase in market risk capital of between 1.5 and 2.4 times compared to current market risk capital. The lowest estimate of 1.5 times assumes all banks will receive internal model approval for all desks. If all banks fail the internal model tests for all trading desks, market risk capital would increase by 2.4 times. ISDA believes the end result will be somewhere in between, but this will depend on two key variables: interpretation of rules on a so-called P&L attribution test and whether the calibration of capital floors applies to market risk.
The former is particularly important – and currently problematic. Under the FRTB, banks have to apply for regulatory approval to use internal models for each trading desk, with approval dependent on passing a P&L attribution test (essentially comparing internal capital systems with front-office models). But there is currently a lack of clarity over how this test will work in practice, while banks have not had time to develop the infrastructure that would enable them to produce the data required for the test.
Without more certainty on the methodology, and without knowing whether or at what level capital floors will be set, it is difficult to accurately estimate the ultimate impact. But it is unlikely all banks will receive internal model approval for all desks, meaning the end result may be closer to 2.4 times than 1.5 times.
Crucially, the study shows the final FRTB framework hasn’t eliminated a cliff effect between standardized and internal models. If a particular desk loses model approval, capital requirements could immediately increase by multiple times. This had been something the Basel Committee had wanted to eliminate.
The FX and equity markets are most affected. Losing internal model approval under the new rules would result in a 6.2 times increase in capital for FX desks and a 4.1 times increase for equity desks.
These are big increases, and come on top of the jump in capital requirements already envisaged in Basel III. The question is whether this single piece of the jigsaw suggests the final picture will be out of line with what the G-20 expects. To put it more simply, will this piece, when combined with other changes in the capital framework, ultimately result in further significant increases in capital across the banking sector? The honest answer is that no one knows.
We do, however, know that large increases in capital could mean certain business lines end up becoming uneconomic. This could severely affect the ability of banks to provide risk management services and reduce the availability of financing for borrowers. At a time when some jurisdictions are increasingly focused on initiatives to generate and sustain economic growth, that’s a concern.
 These numbers exclude the so-called residual risk add-on, non-modellable risk factors and diversification across risk classes under internal models