VoxEU Column COVID-19 Financial Regulation and Banking

Regulatory convergence and market confidence

Regulatory divergence in the banking sector, including differences in how jurisdictions apply Basel III and other regulatory standards, could be a significant source of market fragmentation. Unwarranted divergence in prudential requirements may discourage cross-border activities, international capital flows, and global risk-sharing. This column identifies three main sources of regulatory fragmentation in the post-Basel III era and proposes policy measures to reduce excessive discrepancies in the application of global banking rules. Maintaining trust in banks’ financial health is crucial in periods of economic and financial disruption – such as the COVID-19 pandemic.

Global banks have long sought to operate under a single set of globally harmonised rules to enhance their efficiency, reduce costs, and to facilitate market entry across a range of jurisdictions. Standard-setters, such as the Basel Committee on Banking Supervision (BCBS), also have a vested interest in the principle of ‘global rules for global banks’, to foster financial stability and to promote greater confidence in the financial system.

The introduction of Basel III – together with the implementation monitoring programmes of standard-setters, including the BCBS and the Financial Stability Board – constitutes a step forward to a more consistent application of rigorous banking rules across jurisdictions. While Basel III has significantly strengthened banks’ financial buffers to better withstand unexpected shocks, unwarranted divergence in its implementation reduces the positive impact of the new standards, which, ultimately, has a bearing on global financial stability. These differences also undermine the comparability of key prudential ratios across banks and jurisdictions, thus affecting market participants’ ability to make evolving judgments on whether global banks have sufficient capital and liquidity backstops to withstand a sharp spike in non-performing loans, a likely consequence from the fallout of COVID-19.

In a recent paper (Coelho et al. 2020), we examine the issue of regulatory driven fragmentation under Basel III, focusing on its prudential implications. We find that even if domestic regulations are assessed as ‘compliant’ with applicable Basel standards under the BCBS’ regulatory consistency assessment programme, prudential regimes may still diverge and lead to different outcomes across jurisdictions. These differences can be attributed to at least three factors: (i) varying practices in asset valuations that may impact the measurement of banks’ regulatory capital, (ii) differences in the scope of application of Basel III’s minimum regulatory requirements under Pillar 1, and (iii) distinct practices of the supervisory review process under Pillar 2 of Basel III.

Varying asset measurement practices and regulatory capital

Regulatory capital is based on the difference between the value of assets and liabilities. Valuation practices, therefore, materially impact a bank’s reported capital figure.

The measurement of loans, as they comprise the largest asset class in many banks, drive the reported capital of banks. While loans are carried at cost, banks are required to reduce the loan’s carrying value to its estimated recoverable amount if they do not expect to receive the amount due. These valuation adjustments – which are taken through the credit loss provisioning process – are inherently judgmental and may lead to different outcomes across banks and jurisdictions.

The magnitude of credit loss provisions is largely driven by how banks measure losses on non-performing exposures (NPEs), for which there is no single accounting or global prudential standard. In addition, the introduction of expected loss provisioning under International Financial Reporting Standards 9 – which requires banks to provision for expected future losses even if a loan is not past-due – significantly expands the scope for judgment and dependency on internal models. Variations in how this standard is implemented, along with its implications for regulatory capital, will become apparent as the consequences of the COVID-19 pandemic plays out.

Other hard-to-value assets also affect a bank’s capital figure. Accounting rules require certain financial instruments to be measured at fair value (FV), with the unrealised gains or losses immediately impacting Common Equity Tier (CET) 1 capital. Some FV assets, such as so-called Level 2 and 3 assets, are not traded in an active market, necessitating modelling assumptions. These valuations are assumption-dependent and can lead to varied practices, particularly in periods of market volatility. Table 1 highlights the materiality of Level 2 and 3 asset holdings of 23 reporting global systemically important banks (G-SIBs). The median ratio for the reporting G-SIBs indicates that a 10% decline in the estimated FV of their combined Level 2 and 3 asset holdings would eliminate half of their CET 1 capital.

Table 1 Level 2/3 asset holdings of 23 G-SIBs as of end-2018

Note: the information above is based on end-2018 data for 23 G-SIBs as reported in SNL financial. Information on seven G-SIBs domiciled in China and Japan was not available.

In order to constrain the judgmental component of asset measurement, some national authorities control the valuation approaches by banks through a range of bespoke regulatory measures. The lack of sufficiently prescriptive global standards on the matter opens the door for diverging national practices that may make reported capital figures insufficiently comparable across jurisdictions.

Differences in scope of application of Pillar 1 of Basel III

Basel III’s minimum regulatory requirements under Pillar 1, such as quantitative capital and liquidity ratios, are intended for internationally active banks. However, there is no commonly agreed definition of an internationally active bank and each jurisdiction is free to apply its own interpretation. In some countries, a bank is considered internationally active if its size or cross-border activities exceed a certain threshold. In others, a bank is deemed internationally active if it has branches and subsidiaries in foreign countries. Table 2 shows the value of total assets used explicitly or implicitly by jurisdictions to determine the scope of application of the Basel framework. For example, a bank with $20 billion of assets is considered internationally active in one jurisdiction and therefore subject to Pillar 1 requirements, while in another jurisdiction a bank that is ten times bigger would not fall under this category.

Table 2 Implicit size thresholds for the application of the Basel framework

Source: Castro et al. (2017)

BCBS standards constitute minimum requirements and BCBS members may decide to go beyond them, which includes expanding the scope of application to banks that are not internationally active. This is the case, for example, in the EU, where, in general, quantitative capital and liquidity requirements are applied to all banks, and not limited solely to those classified as internationally active.

However, in most BCBS jurisdictions, non-internationally active banks are subject to tailored requirements. Since countries differ in their definitions of ‘internationally active’, the scope of banks subject to tailored prudential requirements can vary substantially across jurisdictions, resulting in additional sources of regulatory divergence.

Different interpretations of Pillar 2 of Basel III

The goal of the Pillar 2 supervisory review process is to ensure that banks have adequate capital and liquidity to support all the risks in their business. In practice, this requires banks and supervisors to assess whether the minimum regulatory requirements under Pillar 1 are commensurate with an entity’s overall risk profile. Therefore, actual capital and liquidity requirements of banks are determined by the combination of minimum Pillar 1 rules and Pillar 2 add-ons.

While minimum Pillar 1 capital and liquidity requirements are clearly prescribed in the Basel standards, Pillar 2 is principles-based so that it could be tailored to the idiosyncrasies of the respective jurisdictions.

This flexibility gives rise to a wide range of Pillar 2 frameworks across jurisdictions. For example, countries require a diverse set of banks to submit a self-assessment of their own capital adequacy, setting the context for the supervisory review process. Authorities also apply various approaches to determine capital add-ons and stress-testing exercises, including where the add-ons, if applied, belong in a bank’s capital hierarchy (Figure 1). Finally, supervisory expectations for liquidity add-ons are not specified in Pillar 2.

Figure 1 Capital hierarchy: Range of practices

Source: BCBS (2019)

Therefore, even if Pillar 1 requirements were fully harmonised, capital and liquidity requirements would not be entirely comparable since differences in the implementation of Pillar 2 will lead to different Pillar 2 add-ons across jurisdictions. While some discrepancies could be justified to reflect differences in national financial systems, others are likely to result from insufficient convergence of national approaches towards sound practices.

Policy implications

Given the unprecedented economic carnage caused by COVID-19, the immediate focus of all prudential authorities is to devise policies to support the resilience of the global banking system, so they are better positioned to support the real economy. In this context, and at an appropriate time, addressing the underlying causes of excessively divergent practices should be considered as a tool to further bolster not only market integration but also investor confidence.

So, what can be done to promote greater regulatory convergence across internationally active banks?

  • First, to reduce excess variability in asset measurement, the use of harmonised prudential backstops may be considered in the calculation of regulatory capital. For NPEs and performing loans, such backstops might entail the development of regulatory provisioning backstops (e.g. minimum provisions required to cover incurred and expected losses arising from such exposures). In regard to the measurement of Level 2/3 assets, there is merit in considering whether supervisory-imposed valuation haircuts can reduce unwarranted variability.
  • Second, policy measures can be taken to ensure that Pillars 1 and 2 of the Basel framework are consistently applied across jurisdictions. A common definition of ‘an internationally active bank’ can help to ensure that the framework’s scope of application is uniformly applied across jurisdictions. In addition, greater clarity on certain features of Pillar 2 can also promote more regulatory convergence. 

In times of heightened market uncertainty, trust in the global banking system remains the oldest and most important game in town. A robust and consistent application of global banking standards across jurisdictions can enhance the proper functioning of the global financial system and boost confidence. This, in turn, can provide comfort to society that banks remain viable and can help to sustain global economic activity in these extraordinary times.

Authors' note: The views expressed in this column are those of the authors and do not necessarily represent the views of the Bank for International Settlements or the Basel-based standard setters.   


Basel Committee on Banking Supervision (2019), “Overview of Pillar 2 supervisory review practices and approaches”, June.

Baudino, P, J Orlandi and R Zamil (2018), “The identification and measurement of non-performing assets: a cross-country comparison”, FSI Insights on policy implementation, no 7, April.

Baudino, P and R Zamil (2019), “Same but different: comparing non-performing asset measures across countries”,, 30 September.

Baudino, P, R Goetschmann, J Henry, K Taniguchi and W Zhu (2018), “Stress-testing banks - a comparative analysis”, FSI Insights on policy implementation, no 12, November.

Castro Carvalho, A, S Hohl, R Raskopf and S Ruhnau (2017), “Proportionality in banking regulation: a cross-country comparison”, FSI Insights on policy implementation, no 1, August.

Coelho, R, F Restoy and R Zamil (2020), “Convergence in the prudential regulation of banks – what is missing?”, FSI Insights on Policy Implementation No 24, February.

Duckwitz, V, S Hohl, K Weissenberg and R Zamil (2019), “Proportionality under Pillar 2 of the Basel Framework”, FSI Insights on policy implementation, no 16, July.

Hohl, S, M Sison, T Stastny and R Zamil (2018), “The Basel framework in 100 jurisdictions: implementation status and proportionality practices”, FSI Insights on policy implementation, no 11, November.

Restoy, F and R Zamil (2017), “Prudential policy considerations under expected loss provisioning: lessons from Asia”, FSI Insights on policy implementation, no 5, October.