When the Basel Committee on Banking Supervision released its minimum capital requirements for market risk in January 2016 — known as the Fundamental Review of the Trading Book (FRTB) — it projected adoption by 2019.
The trend continues in Australia, where the Australian Prudential Regulatory Authority (APRA) published a letter to the nation’s banks notifying them they will not finalize its new market standards until at least 2020, with a one-year implementation.
However, given the operational challenges involved in implementing FRTB, many global banks are making plans to meet the original Basel deadline of 2019. A Bloomberg poll of U.S. and EMEA-based firms, conducted during a recent webinar, found a near even split between banks that anticipate being in a position to meet the original 2019 deadline (41%) and those who expect the timeline to pose a challenge (41%) based on the current status of their FRTB implementation programs.
One of the largest concerns among the banks’ program managers is a lack of adequate preparation for non-modellable risk factors (NMRF). This is especially true in APAC, where in a Bloomberg poll 75% of banks rated compliance with NMRF as the major obstacle to meeting FRTB requirements.
Under FRTB, a risk factor used within the internal model approach (IMA) is only modellable if it is observable, i.e. has 24 real prices over the last year, none further than a month apart. If these criteria are not met, risk factors must be treated as non-modellable (NMRF) and banks will incur additional capital costs.
The complexities involved in meeting this particular requirement will vary by market depending on the availability of trade and pricing data. In the U.S., some trade data is publically accessible from central reporting facilities, such as TRACE. In Europe, more publically available trade data is expected following the implementation of MiFID II next year. The challenge is greater in APAC, where fragmented markets offer few public repositories to consolidate trading data and which lacks a single, base-like currency in the mold of the EUR or USD.
This need for trade data is leading firms to partner with vendors who can help. Bloomberg has completed the first phase of its proof of concept with participating banks, which involves collecting actual trade data, anonymizing contributions, and assessing seasonality and data coverage for specific asset classes.
The results of this trial, which were recently presented to participating banks, show that pooling contributed trade data with Bloomberg’s own data can significantly improve modellability results.
P&L attribution tests
In conjunction with modellability issues, most discussions among regulators and those banks considering running IMA center on passing FRTB’s P&L attribution tests. To get model approval, banks must demonstrate on a regular basis that P&L projected using risk model factors do not deviate significantly from P&L calculated with the market inputs and pricing functions used by the front office.
Banks will need to accept a degree of trade off between passing the test and modellability. Increasing the number of factors will improve the likelihood of passing the test, but as more specific factors are more likely to be non-modellable, may also entail a higher capital charge.
The complexity involved in any IMA implementation means banks will need to have some idea in advance of whether they can pass the required tests. Some banks are better positioned to predict this than others. Those with strong risk infrastructure in place have been using it to forecast which desks would pass P&L attribution and gauge the capital impact of using IMA versus the standardized approach (SA).
For other banks, time is short. They need to begin envisioning their infrastructure and models under FRTB, determine which desks would pass model approval, and quantify what the capital and business impact of not passing the test would be. While regulatory delays may grant them some leeway, many banks need to move quickly in order to make informed decisions on the ultimate impact of FRTB on their business and operations.
While FRTB brings many challenges, banks are also beginning to appreciate its potential for improving risk management. Once implemented, banks stand to reap a range of benefits from the new regulation, including a more risk-sensitive standardized approach, better internal models through the introduction of liquidity horizons and P&L attribution tests, and higher quality market data.