The Fundamental Review of the Trading Book: Implications for Portfolio and Risk Management in the Banking Sector

• Published date: 01 October 2023
• Author: ORLA MCCULLAGH,MARK CUMMINS,SHEILA KILLIAN
• Date: 01 October 2023
• DOI: http://doi.org/10.1111/jmcb.13022

DOI: 10.1111/jmcb.13022

ORLA MCCULLAGH

MARK CUMMINS

SHEILA KILLIAN

The Fundamental Review of the Trading Book:

Implications for Portfolio and Risk Management in

the Banking Sector

The Fundamental Reviewof the Trading Book (FRTB) is the promised over-

haul of bankmarket risk regulation. FRTBretains the authorized use of pro-

prietary risk models, however,it introduces two additional criteria: (i) P&L

attribution (PLA) tests and (ii) desk-level backtests. We examine empiri-

cally whether these additional criteria inuence risk management and port-

folio management practice, specically portfolio construction and choice of

risk model. We nd that the PLA tests demand signicant alignment with

risk factors, however, the backtests do not incentivize use of superior risk

models. This has important implications for the efcacy of the capital-based

regulatory system.

JEL codes: G11, G17, G21, G28

Keywords: Basel III, Fundamental Reviewof the Trading Book, market

risk, portfolio management, value-at-risk, P&L attribution tests

T   -  management were laid

bare in the events of the nancial crisis (2007-09). Indeed, during this turbulent time,

bank capitalization proved an important factor for resilience. Demirguc-Kunt, De-

tragiache, and Merrouche (2013) provide evidence that better capitalized banks ex-

O M is with the Department of Accounting and Finance, KemmyBusiness School, Uni-

versity of Limerick (E-mail: orla.mccullagh@ul.ie). M C is with Financial and Operational

Performance Group, Business School, Dublin City University. S K is with Department of

Accounting and Finance, Kemmy Business School,University of Limerick.

Received September 11, 2020; and accepted in revised form March 31, 2022.

Journal of Money, Credit and Banking, Vol. 55, No. 7 (October 2023)

© 2023 The Authors. Journal of Money, Credit and Banking published by Wiley Periodicals

LLC on behalf of Ohio State University.

This is an open access article under the terms of the Creative Commons Attribution-NonCom-

mercial-NoDerivs License, which permits use and distribution in any medium, provided the

original work is properly cited, the use is non-commercial and no modications or adaptations

are made.

1786 :MONEY,CREDIT AND BANKING

perienced higher stock returns during this period, with higher quality capital hold-

ings (Tier 1 capital and tangible common equity) being more relevant. The Funda-

mental Review of the Trading Book (FRTB) is the regulatory response of the Basel

Committee for Banking Supervision (BCBS) to these market risk management fail-

ings. The headline change to the market risk regulatory framework under FRTBis the

replacement of value-at-risk (VaR) by expected shortfall (ES) for calculating capital

requirements, although VaR remains a central metric for backtesting. As described

by Gordy and McNeil (2018, p. 3) of the Federal Reserve: “although estimates of ES

will be the cornerstone of the risk capital calculation, the risk model approval process

will continue to be based on VaR estimates and VaR exceedances.” FRTB will sig-

nicantly increase regulatory capital requirements, primarily to ensure the stability

of the nancial system and substantially act as a valve for the availability of credit

to the economy. Our paper contributes to the emerging discourse on FRTB through

a quantitative analysis of its impact on bank-market risk management and portfolio

management practice.

The prerogative for banks to develop their own internal models, including their

choice of risk-forecasting models, remains central to FRTB, with the BCBS arguing

that this is essential to enable a level playing eld between banks in different jurisdic-

tions (BCBS 2019). A key concern with the continued use of internal models across

the banking system is the level of variation in the results from banks’ proprietary

internal models (Beder 1995, Pritsker 1997, Berkowitz and O’Brien 2002, BCBS

2018). Indeed, VaR became a cynosure for criticism of bank risk taking and under-

capitalization in the 2007–09 nancial crisis (Nocera 2009, O’Brien and Szersze´

n

2017) because of (i) the alarming variability of VaR-implied capital requirements,

(ii) the perceived ease with which VaR can be gamed (Danielsson and Zhou 2016,

Armstrong and Brigo 2019), and (iii) the issue that VaR is a point estimate that does

not quantify losses in the tail. FRTB aims to address these key concerns while os-

tensibly retaining the role of proprietary internal risk models as a means of enabling

level playing eld competition and risk-sensitivecapital. However, it undergirds their

use with additional restrictions.

FRTB introduces two additional criteria for the qualied use of an internal model

approach (IMA) in determining market risk regulatory capital. These two criteria are

(i) P&L attribution (PLA) tests and (ii) desk-level backtests. These additional criteria

have the potential to change the nature of the use of proprietary risk models in the

determination of market risk capital. In this context, we formally assess these two

criteria and provide new insightsof relevance for banking practitioners and regulators.

The 2018 McKinsey report on FRTB (Azoulay et al. 2018b) depicts practitioners’

views of FRTB as requiring merely a change of statistical metric, namely, a move

from VaR to ES. Contemporary nancial media, in their evaluation of FRTB, have

focused much of their attention on the PLA tests (Nield 2017), which have been re-

vised since the original FRTB publication1(Mahfoudhi 2018). However, as impact

1. The original specications of the PLA test incorporated the joint test of mean and variance of the

RTPL and the HPL: (BCBS 2016)

ORLA MCCULLAGH, MARK CUMMINS, AND SHEILA KILLIAN :1787

analyses are published (EBA 2019), the seismic shift implied by FRTB is being rec-

ognized. According to the European Banking Authority, the expected averageimpact

of FRTB on Pillar 1 market risk capital is an increase of 81%, with an interquartile

range of 32% to 140% (EBA 2019). This excludes the application of the output oor.2

Despite this context, there are a limited number of studies examining FRTB to

date. This is likely due to its recent development and postponed implementation. In

March 2020, the Basel Committee’s oversight body, the Group of Central Bank Gov-

ernors and Heads of Supervision (GHOS), approved a revised implementation sched-

ule (now January 2023) to facilitate banks and supervisors in the management of -

nancial stability issues arising from the impact of COVID-19 on the banking system

(ICMA 2020). Soobratty, Stern, and Cheng (2020) nd that regulators viewthis post-

ponement as a pragmatic respite, though assert that banks must move forward with

their implementation plans. They further argue that the feasibility of banks’ adoption

of IMA hinges on their preparedness, including quantitative analysis and backtesting.

This reinforces the relevance, timeliness, and prescience of our quantitative impact

study of the FRTB framework.

One of the rst FRTB studies is that of Thompson, Luo, and Fergusson (2016,

2017)3who examine the theoretical failure rate of the PLA tests (based on the Jan-

uary 2016 specication). Subsequently, Farag (2017) argues that alignment between

front ofce pricing models and middle ofce risk models will be key to passing the

PLA tests. Correspondingly, Farag (2018) examines anomalies and asymmetries in

the proposed FRTB SA and IMA framework that could be problematic upon imple-

mentation. Additionally, Pederzoli and Torricelli (2019) provide a useful illustration

of the impact of both FRTB’s standardized approach (SA) and IMA on a stylized

portfolio. The calculative requirements for SA are radically overhauled in FRTB,es-

tablishing greater risk sensitivity in the capital calculations. FRTB requires that IMA

and SA are ran in parallel, which presents signicant challenges to banks’ information

technology systems. Indeed, Li and Xing (2018) examine different computational ap-

proaches to align internal capital allocation to FRTB regulatory capital calculation.

In the context of the postponed implementation of FRTB due to the COVID-19 pan-

demic, Lazar and Zhang (2020) quantitatively analyze the impact of market reaction

to the COVID-19 pandemic under a stylized interpretation of the FRTB-prescribed

ES measure. They nd that the measure generally overestimates the risk, driving in-

creases in regulatory capital, and that assets with longer liquidity horizons display

disproportionately high capital requirements.

2. The output oor limits the benets achieved under the IMA relative to the SA (SA). BCBS argue

that the output oor will strengthen the principle of the level playing eld between SA and IMA banks,

improve the comparability of disclosures, and enhance the credibility of capital calculations (BCBS 2017).

Capital requirements will be calculated as the higher of: (i) capital calculated using the IMA (where the

bank has approvalfor their use) and (ii) 72.5% of the capital requirements calculated under the standardized

(or simplied standardized, where appropriate) approach.

3. Thompson et al. (2016) working paper was used for immediacy and subsequently augmented by

the 2017 published paper.