articles Ratings /ratings/en/research/articles/240626-2023-banking-turmoil-global-regulators-reflect-and-react-13163522 content esgSubNav
In This List
COMMENTS

2023 Banking Turmoil: Global Regulators Reflect And React

COMMENTS

Tech Crackdown Could Boost Expenses For South And Southeast Asia Banks

COMMENTS

Asia-Pacific Financial Institutions Monitor 3Q 2024: A Choppier Ride For The Rest Of The Year

COMMENTS

Australian Mutual Lenders: The Magic Number Could Be Less Than 10

COMMENTS

Japan Banks Primed For Market Turbulence


2023 Banking Turmoil: Global Regulators Reflect And React

The 2023 banking sector turmoil in the U.S. and, to a lesser extent, Europe spurred numerous policymaker and regulatory reflections. Key among them is whether further changes are needed to the global and/or national banking regulation frameworks and how supervisors would apply these. Still evolving, these speeches and papers touch on diverse aspects related to banking risks (e.g., governance and risk management), regulatory issues (interest rate and liquidity risk, resolvability), and supervisory shortcomings (the lack of supervisory resources or powers). Among the numerous pronouncements, S&P Global Ratings lists what it sees as key to explaining policymaker and regulators' thinking and possible next steps (see related research below).

Our base case is that this regulatory brainstorming is unlikely to lead to a major reset of global banking rules. Instead, we could see a strengthening of their implementation by banks, national policymakers, supervisors, and resolution authorities--with differences likely to persist across jurisdictions. This would reflect that a principal root cause of the 2023 sector volatility was not the cornerstone rules agreed after the global financial crisis (GFC), but rather a confluence of inadequate implementation of existing rules and failures in bank governance and risk management. Nevertheless, some adjustments are likely--most obviously in aspects of regulation and supervision in the U.S. and Switzerland, and in rules to ensure banks' liquidity resilience.

Here, we draw several key conclusions from this period of reflection, highlight what we perceive to be regulators' likeliest next steps, and explain what this means for our ratings on global banks.

Conclusions

Conclusion 1: The turmoil reflected the interplay of monetary policy setting, regulatory stance, and supervisory effectiveness

Years of easy monetary policy created excess reserves and large deposit inflows into banks. How banks managed this has proven highly important. The regulatory and supervisory frameworks have also played a key role here.

In the U.S., regional banks were, and remain, subject to tailored regulations, less stringent than those faced by their larger peers, and supervision that proved less effective in some cases. Some built up liquidity risk (concentrations on large, uninsured deposits) and asset-liability management (ALM) mismatches over time. When the U.S. Federal Reserve quickly switched its monetary policy stance in response to inflation, this soon revealed banks most exposed to liquidity and ALM risks. These risks crystallized when they were exposed in combination, most notably in the case of Silicon Valley Bank (SVB). For other U.S. banks, liquidity and ALM risks (e.g., unrealized losses) have eased, but remain notable.

In Europe, there was no regulatory loosening of rules and proactive supervision broadly prevented these imbalances from growing in parts of the banking system. In addition, the European Central Bank (ECB) took a more gradual approach than the Fed in its monetary policy normalization, with a later and more limited interest rate shock, and limited action on excess liquidity (so far).

Conclusion 2: Bank failures were largely the expression of classic banking risks

Liquidity runs that sustain themselves are unlikely to be baseless. They rather express, aggressively, market and client concerns about other flaws and weaknesses. Fundamentally, the banking turmoil was the materialization of known idiosyncratic banking risks and the possibility of contagion. SVB was an extreme example in terms of the nature and concentration of its funding profile. Across these banks, weaknesses included:

  • Ineffective governance (e.g., lack of a Chief Risk Officer);
  • Risk management deficiencies (e.g., ALM mismanagement, commingling of IRRBB [interest rate risk in the banking book] and liquidity risks, insufficient reaction to repeated failings);
  • Business models that do not reliably meet stakeholders' expectations;
  • Concentrations, whether for funding or revenue-generation;
  • Impact of media exposure and market indicators on stakeholder confidence (need for appropriate communication plan and execution); and
  • Ineffective supervision, where authorities are unable to exert substantial influence in a timely way.

Nevertheless, the role of social media and digitalization on reputational risks and velocity of contagion give pause for wider thought. Digitally enabled bank runs are not new--digital channels were a prime conduit for the run on U.K.-based Northern Rock in 2008. However, the weight of digitally enabled retail and corporate deposits is now far higher than a decade ago.

Conclusion 3: The agenda for change starts with supervision

The lack of supervisory powers and effectiveness were identified as a key root cause of events in the U.S. and Switzerland. The agenda for change is first focused on national implementation of global rules and supervision--mainly in the U.S., but also in Switzerland.

We see particular focus on the following topics:

  • Importance of supervisory resources--size and expertise of supervisory teams;
  • How to ensure timely and effective supervisory action--the U.S. Barr report highlighted the difference between flagging risks at banks and acting on them;
  • Adequacy of supervisory powers--e.g., to impose fines or to require management to take specific actions; and
  • The appropriate proportionate implementation of rules and supervision, i.e., how to tailor rules and calibrate supervisory resources.
Conclusion 4: Global regulatory policy changes may come in time, but are already arriving at a national level

We see strong policymaker consensus that the March 2023 events did not fundamentally challenge the global regulatory policy framework agreed after the GFC. Policymakers are instead highlighting areas of potential fine-tuning of some calibrations, reinforcement of crisis management, and ensuring full implementation.

National level policy changes are easier to execute—and are already under way in Switzerland, the U.S., and the U.K. However, where changes require political consensus or agreement, as in the EU for framework (level 1 and 2) changes, this will take longer. In the U.S., the complexity of designing updated capital requirements and multiple other potential proposals may also require a lengthy period. Globally agreed changes would take years, based on previous decision-making cycles.

We notably see policy likely to evolve in the following areas: unrealized losses, liquidity coverage ratio (LCR) calibration and contingent liquidity, IRRBB, crisis management, and the role of additional Tier 1 (AT1) capital. This sits on top of a key topic left over from the pandemic--how/whether to improve capital buffer flexibility, notably in Europe.

Conclusion 5: Bank liquidity resilience demands institutional and bank-level actions

The "G30" expert panel noted that the bank failures and interventions revealed many weaknesses in banks and institutional frameworks. Notably, it flagged that a major issue common to all banking systems is that liquidity runs will be much faster due to the speed of information (and rumour) transmission, and technology (real-time banking). Among other things, the report recommends:

  • To limit spillover contagion: enhanced liquidity support mechanisms, a much-improved lender of last resort (LoLR) regime, possible changes to deposit insurance, and a truly workable scheme for resolving large banks.
  • Banks should preposition enough collateral to cover, after the normal haircut for credit risks, all runnable liabilities--i.e., excluding capital, long-term debt, swap liabilities, and insured deposits. Central banks would support this with an efficient collateral management system.

The concept of banks having better contingent funding access is not radical. However, the proposition that all runnable liabilities would in effect need to fund only assets with high liquidity value would dramatically affect credit intermediation, particularly in markets where securitization and covered bonds are little used, and for corporate/SME-focused banks. As a result, a move to this type of narrow banking model appears unlikely to be adopted.

Conclusion 6: The halting of contagion in the U.S., and lack of it in Europe, reflect key differences

In the U.S., contagion was most acute because there were similar weaknesses across several banks, albeit not to the same extent. Regulation did not require the banks that failed, or others of similar or smaller size, to file resolution plans at the group level or to maintain specific levels of total loss-absorbing capacity (TLAC), although they turned out to pose some systemic risks. Ultimately, the authorities stemmed this tide through:

  • The Federal Deposit Insurance Corporation's (FDIC's) extraordinary use of the systemic risk exception in the failures of SVB and Signature Bank (two of the three large failures) to guarantee uninsured deposits while transferring substantially all their deposits and assets to bridge banks prior to sale;
  • A specific bank funding program launched by the Fed against securities at par value; and
  • The use of bigger, stronger, more diversified banks--i.e., a tiered, semi-heterogenous system--to inject interbank deposits in and ultimately acquire First Republic Bank (the third of the large failures).

In Europe, Credit Suisse (CS) had a specific business model and pre-existing confidence problem, exacerbated by a loss of confidence in the U.S.

The absence of a common fact set (banks that had the same combination of underlying troubles as CS, SVB, and Signature Bank) was key to the lack of contagion in Europe. In our view, this was not due to good luck--the European banking system is heterogenous, and some key regulations were more strictly applied (for example, unrealized losses on available-for-sale treasury portfolios are captured in regulatory capital, and the ECB had undertaken significant work to identify banks with outlying IRRBB risk appetite).

Nevertheless, the European banking system remains fundamentally a network of fragmented national-level banking systems. This may ease contagion risk but may also undermine resilience in the broader sense, including the ability to execute transfers of sizable banks in resolution.

Credit spreads have since retightened a lot, and the risk of contagion is now much lower. However, the 2023 episode will remain fresh in the memory, leaving markets highly sensitive to negative surprises.

Conclusion 7: Deposits are special

The FDIC used its systemic risk exemption to avoid enforcing losses on uninsured depositors of SVB and Signature Bank. We think this nods to two realities:

  • There could have been adverse economic consequences of enforcing losses on corporates' deposits, particularly operational ones, even though SVB was not itself deemed systemic; and
  • Deposit bail-in will likely promote contagion and so undermine financial stability. This may be true even if the institution is not among the largest banks, highly connected to other financial counterparties, or involved in critical financial services.

This also underlies the Bank of England's (BOE) decision that it was in the public interest to use resolution (stabilization) tools on SVB's U.K. subsidiary rather than liquidate it. Indeed, while most uninsured bank deposits are bail-inable in Europe (and were forced to take losses in Cyprus in 2014), European authorities tend to find solutions for failing and failed banks that avoid losses for uninsured deposits.

Indeed, the EU's crisis management and deposit insurance (CMDI) proposal (which was already in train before March 2023) and the BOE's response proposal go further. If implemented, they could result in very few insured depositors in failed banks needing direct compensation from a deposit guarantee scheme (DGS), as they would benefit instead from the use of resolution tools that ensure continuity of service.

Conclusion 8: Politicians and the market might not be ready for bail-in

Resolution frameworks are designed to deliver a credible alternative to a bailout of systemic banks--options are enabled but not dictated by resolvability. Whether the resolution option is pursued in a specific circumstance depends on many things, including any alternatives on the table (e.g., the emergence of a last-minute acquirer), the broader environment in which the stress occurs, and the political calculus. For CS, the Swiss authorities' view was that, in the circumstances at that time, resolution (with attendant bail-in of CS' TLAC debt instruments) could have caused a systemic crisis.

That AT1 instruments would take losses when a bank is close to failure is far from new--and yet still there was significant market fallout due to the AT1 write-down, not only because shareholders were not first fully written down. Bail-in resolution has not been tried yet and lacks credibility among some investors who variously doubt the plausibility of (i) using resolution to address the largest systemic banks if they fail and/or (ii) the authorities' willingness to bail-in "senior" instruments that banks use not only to meet loss-absorbing capacity requirements but also funding needs.

This prompts two questions:

  • Whether politicians and regulators are ready to take the (perceived) risk to bail-in a large amount of funding instruments; and
  • Whether the market is ready to see a bail-in of billions of dollars of structurally or contractually subordinated senior bonds (i.e., those included in TLAC or MREL requirements).

In the absence of observable time-consistent responses to stress and failure of major banks, this leaves resolution authorities and policymakers with a challenge to guide market expectations.

Conclusion 9: Accounting differences could continue to shape behavioral differences

Bank behavior is shaped by both regulatory and accounting implications, and the two are not always the same. Notable differences exist between IFRS and U.S. GAAP as regards provisioning--where U.S. GAAP leads to more aggressive loan loss provisioning--but the rise in policy rates highlighted important differences regarding investment portfolios. First, U.S. banks took advantage of U.S. GAAP's accommodation of banks' recategorization of available for sale (AFS) investments to hold to maturity (HTM), which helped them to avoid reporting billions of dollars in unrealized losses. Second, IFRS hedge accounting allows banks to acknowledge the economic mitigation from derivatives hedging (notably via swaps), whereas this is more difficult under U.S. GAAP. This is one reason why U.S. banks' investment portfolios tend to be more heavily invested in government bonds than for IFRS banks--bond portfolios that materially diminished in value after 2022. The accounting bodies have not signaled any review of these standards.

What's On The Regulatory Action Plan?

1 Supervision: Lifting intensity

Switzerland.  FINMA seems likely to gain additional powers to make it more comparable with leading European supervisors, although it's unclear whether it will gain additional resources. The April 2024 Swiss Federal Council proposal envisages expanded powers or changes in the following areas. Timing would depend on whether they can be made under administrative procedures or amended federal law.

  • The ability to impose higher capital or liquidity requirements, as well as limit capital distributions or incentive compensation, for banks with inadequate capital planning, liquidity risk management, or governance and controls;
  • The authority to impose fines;
  • The authority to require management to take corrective actions;
  • The creation of a senior manager regime to increase personal accountability of bank executives; and
  • Stronger influence over bank compensation to set tougher minimum standards and align incentives to risk management.

U.S.   Regulators have not suggested that supervisors lack important powers, but rather that supervision should be more proactive, challenging, and interventionist in the face of identified deficiencies. Faced with dynamic, fast-changing banks, supervision should also be more dynamic and responsive to changing risk profiles. Unlike FINMA, the FDIC said understaffing and resource challenges affected the timeliness and quality of its examinations (notably of Signature Bank). Like FINMA, it's unclear whether it will gain additional resources.

Beyond the U.S. and Switzerland.   We see no significant changes planned aside from hindsight considerations by many regulators as to the question of "could it happen here" with some regulators making announcements. A case in point is the Australian Prudential Regulation Authority (APRA) which in November 2023 proposed targeted changes on liquidity and capital in response to global banking turmoil, specifically reinforcing the importance of minimizing contagion risks.

More generally, these events will have reminded supervisors and policymakers of:

  • The benefits of strong intrusive and corrective supervision, aided by a wide range of tools to correct bad bank behaviors (not only capital requirements, but also business restrictions, board member reassessments and removals, sanctions, and other penalty payments).
  • The benefits of wide-ranging, multiple scenario stress testing.
  • The benefits of monitoring market and other signals of sentiment toward individual banks.
2. Regulation: Selective tightening

Regulatory changes will likely focus on five areas.

Revised application of proportionality within regulations.  Notably in the U.S., which has a "tailoring" approach and thresholds for "systemic" banks subject to resolution preparations, U.S. authorities now explore how to strengthen requirements for any bank with more than US$100 billion in assets.

Unrealized losses.   Many jurisdictions have long required unrealized losses on AFS/FVOCI portfolios (fair value through other comprehensive income) to be recognized in regulatory capital. They seem likely to persist with this, but appear unlikely to extend it to HTM/HTC (hold to collect) portfolios. In this, context it appears counterintuitive that EU policymakers have just approved a new prudential filter under Capital Requirements Regulation (CRR) 3 to allow a short-term waiver through end-2025 for banks to not recognize unrealized gains and losses on government bond portfolios held at FVOCI. In the U.S., regulators have proposed to require a broader set of banks (any bank with more than US$100 billion in assets) to include unrealized gains and losses on AFS securities in regulatory capital on a phased-in basis.

IRRBB.  Following the jump in policy rates, the Basel Committee is updating some of the parameters and stress assumptions underlying the measurement of IRRBB. The EU already fully implemented Basel rules and went further, with thresholds imposed on the maximum permitted impact on net interest income (NII) (in addition to that on economic value of equity). The European Banking Authority IRRBB workplan envisages that supervisors will have greater ability to challenge banks' modelling and hedging practices, and improve consistency in the banks' disclosures. Australia is unusual in that it has long included IRRBB as a pillar 1 risk, but so far no other jurisdiction has indicated plans to follow this regulatory approach. Rather, we see IRRBB remaining fundamentally a pillar 2 topic, but with more onus on supervisors to identify and challenge banks' modelling inconsistencies.

Capital.   None of the banking groups failed due to low regulatory capitalization at a consolidated level, so it's logical that the rulebook would not be rewritten. Nevertheless, the Basel committee is reflecting on the role of AT1 capital, following the Swiss treatment of AT1 instrument holders (relative to shareholders) and the observation that CS management chose not to voluntarily cancel AT1 coupons despite the bank being under stress.

At a national level, APRA is consulting on various related options that envisage reducing the role of AT1 in bank core capital, making design changes to the AT1 contract (e.g., a conversion trigger well above 5.125% and/or based on forward-looking metrics), and introducing mandatory coupon nonpayment if buffers are breached or the bank is lossmaking. APRA may also amend eligibility criteria for AT1 to push the investor base away from retail investors.

Swiss regulators also have their eyes on ensuring that AT1 more reliably absorbs losses on a going-concern basis. They have not announced any specific measure so far but have floated some ideas. These include requiring high contractual triggers for AT1 instruments (e.g., 10% common equity Tier 1) and prohibiting the payment of AT1 coupons in case of consecutive net losses, for example. In addition, they are proposing a selective tightening of requirements on systemic parent banks' foreign participations, prudent valuation adjustments, and intangible assets, and a stronger influence of stress testing on institution-specific Pillar 2 capital surcharges.

There is nevertheless unfinished business on capital in the form of the Basel III endgame rules. Implementation is now agreed in the EU, but remains open in the U.S. The stringency of the initial U.S. proposal, which went beyond the Basel consensus in some respects, was partly a reaction to the events of March 2023, in our view. There would, however, be a potential trade-off with bank competitiveness and intermediation capacity. The final shape of these rules remains unclear and regulators have recently said that their original proposal would change materially.

Funding and liquidity.  We expect considerable reflection on how digitalization (and to a lesser extent social media) has influenced liquidity risks for bank deposits. Switzerland was already tightening regulatory requirements on systemic banks to include a surcharge to pick up additional institution-specific risks.

It's possible that other individual national level jurisdictions could also make associated changes to regulation, but we expect this would more likely be agreed at an international (Basel) level and so take time to gather evidence and develop consensus. The outcome could include:

  • A recalibration of outflow assumptions under the LCR, notably for non-stable retail deposits;
  • An additional/supplementary liquidity ratio, for example, that considers a shorter timeframe than the 30 days under the LCR;
  • A narrowing of what amounts to "stable" deposit funding under the net stable funding ratio (NSFR); and
  • Enhanced disclosures by banks.

Amid a detailed focus on operational liquidity needs for complex banks, we believe that a tighter definition of LCR-eligible liquid assets would avoid the double duty benefit that CS apparently made use of. The Swiss authorities already tightened this aspect of liquidity policy for Swiss systemically important banks (SIBs) before March 2023 (a measure that took effect from January 2024), and others could follow.

While there have been some calls to exclude HTC/HTM securities from LCR-eligible assets, this would be radical and seems unlikely to be pursued in practice. That said, U.S. regulators have indicated that they could propose to restrict the extent that HTM securities could be counted in LCR (rather than completely eliminating from the ratio).

Having liquid assets is necessary but not sufficient for banks to endure acute liquidity stress. Even if policymakers do not go as far as the G30 expert panel suggested, supervisors will press banks to substantially increase operational capacity to monetize as much of their asset base as possible. This also boosts their liquidity in resolution. U.S. regulators recently said they may look to require large banks to maintain borrowing capacity at the Fed's discount window, based on pledged collateral, at some fraction of certain types of uninsured deposits.

Particularly in the U.S. and Switzerland, but also more broadly, central banks are reflecting further on the breadth of their liquidity frameworks and/or on the efficiency of their collateral management systems. Central banks might also attempt to de-stigmatize the use of the discount window.

3. Crisis management and deposit insurance: Modest convergence between the U.S. and Europe?

For well over a decade, U.S. and European policymakers have held a shared conviction that resolution is a critical feature to address the problem that major banks are too big to fail. The implementation of these frameworks has been thorough but varied significantly in scope--reflecting national institutional preferences and the shape of their banking systems. We see changes as likely to be focused on four areas.

Deposit guarantee scheme (DGS) coverage.  Having observed rapid outflows of uninsured deposits, policymakers have pondered whether to expand their DGS. Aside from a targeted widening in the EU, DGS scope and size is likely to remain fundamentally unchanged. The U.S. (FDIC) continues to review DGS coverage, with the most likely change option being to offer different deposit insurance limits across account types. However, a change in the near term seems unlikely since it would likely require an act of Congress.

Optimizing resolution planning for the largest banks.  Bail-in-led resolution will remain the primary option for U.S. and European G-SIBs and D-SIBs if they fail, but in Europe at least these banks will be asked to undertake deeper planning for alternative tools (e.g., business transfer) in case these prove to be a better outcome in the specific circumstances of a bank's failure. Resolution authorities will work with overseas securities regulators to gain greater assurance that debt issued by European banks into overseas markets can be reliably bailed-in and remain compliant with local securities laws. In contrast with the U.S. and Europe, our view for 15 of 19 banking jurisdictions we cover in Asia-Pacific is that direct extraordinary government support for systemically important private sector commercial banks will likely be forthcoming in the unlikely event it was ever needed. Of note, TLAC regimes are well progressed in some of these jurisdictions.

Expanded scope of resolution.  U.S. regulation currently only requires G-SIBs to meet TLAC requirements. Regulators have now proposed to require any bank with more than $100 billion in assets to maintain certain levels of debt that could absorb losses in resolution. They also have proposed new requirements and guidance for resolution planning on those banks, aiming to reduce the systemic risk that can follow the failure of even a regional bank. European regulators have long considered that even midsize banks could pose systemic risk and so merit the use of resolution tools if they failed. They now go further. U.K. regulators envisage a broader use of resolution tools to small banks, but notably also better operational continuity preparedness to avoid deposit/service interruptions and deadweight costs associated with liquidation. The EU CMDI proposal envisages a widening to allow that banks with systemic importance at a regional (not just national) level could also be subject to resolution. It also envisages a more routine use of FDIC-like purchase and assumption transactions (named sale of assets/business in the EU) for smaller banks, aided by funding from DGS and resolution funds. To facilitate this, the EU would introduce a general depositor preference to senior debt, like that already in the U.S., although it would likely maintain the DGS' super-preferential position in the creditor hierarchy.

Extraordinary liquidity support.  The Swiss will codify the Swiss National Bank's new public liquidity backstop into ordinary law, to allow it to lend to a solvent bank without collateral. But they and the BOE might remain relative outliers among U.S. and European central banks in benefiting from this extra flexibility. In the European banking union, Italy needs to sign into law the European Stability Mechanism (ESM) backstop to the Single Resolution Fund for it to become operational. However, this could mobilize only tens of billions of extra lending capacity for the ECB and, while authorities have acknowledged the merits of a larger backstop, there is no concrete proposal.

This points to modest U.S./European convergence in the scope of banks subject to resolution powers, but key differences are likely to remain--not least that pre-emptive solvency support remains legally possible in Europe, but not in the U.S.

What This Means For Ratings

S&P Global Ratings has not announced plans to revise its financial institutions rating methodologies in light of the 2023 banking turmoil. The methodologies provide a systematic but flexible credit assessment framework that blends quantitative measures with a strong qualitative overlay and can be applied globally. Nevertheless, any significant event that affects an industry sector gives pause for reflection.

On banking risks, regulation, and supervision

As we note above, the revealed crystallized banking risks are not new risks--concentrations (assets, funding, revenue base) can rapidly jeopardize a bank's viability. But secular change (e.g., digitalization) and cyclical change (the effects of monetary policy shifts) merit close attention.

For all their other merits from a credit standpoint, banks remain highly leveraged, confidence-sensitive institutions. Confidence sensitivity is further increased when:

  • There are visible indications of adverse market sentiment--low or volatile stock price, spiking CDS, significant short positions, bonds trading far below par; or
  • Substantial unrecognized losses (on securities or loans) or weak business prospects imply that an institution's economic value could be far below its book value.

Adverse market sentiment is unlikely to be self-perpetuating in the absence of fundamental underlying weakness.

As the starting point to our ratings on banks, our banking industry country risk assessments (BICRA) acknowledge that banks operate in an ecosystem. Healthy banks are not immune from the travails of their weaker peers (of size) or the lack of a credible backstop if systemic peers fail. Persistent weakness and poorly managed bank failures undermine confidence and can cause contagion. Bank failures that deplete DGS or resolution funds will be shouldered by the industry after the fact. Last-minute commercial deals (pre-failure or in resolution) might offer good value for the acquirer, but cannot be relied on for the largest banks. Bail-in may spread the loss on a failed bank to investors beyond the local banking system, but the risk premium for other banks will rise.

We continue to reflect in our BICRA institutional framework assessments jurisdictional differences in how regulations are implemented and enforced. Strong regulation rests on effective supervision. All else being equal, enhanced supervisory effectiveness in the U.S. and Switzerland would be supportive for bank ratings. But, even if changes are enacted, we do not see them as a likely catalyst for higher bank ratings, at a system level, particularly given that bank regulation and supervision would remain heavily tiered in these jurisdictions. It's notable, for example, that many of the regulatory changes that the Swiss federal council has proposed would be focused only on systemic banks.

Regulatory and legal frameworks that provide for a reliable prospect of multi-layered and substantial extraordinary liquidity support are evidence of a strong institutional framework that could benefit individual banks on a going- and gone-concern basis, and stem contagion. However, the existence of a supportive institutional liquidity framework is by itself not sufficient. Banks' ability to access these frameworks is idiosyncratic--depending on their risk appetite, preparedness, and shape of their loan books. Particularly in the context of the regulatory LCRs' short 30-day horizon and nongranular assumptions on deposit outflows, we continue to see this as just one of several important liquidity ratios. And we see ratios themselves as only the starting point to a much broader view of a bank's liquidity that encompasses insights into the likely behaviour of its deposit book under stress, the size and nature of other contingent liquidity uses, and the breadth of resources it has available to meet those needs. These features are relevant to our assessment of a bank's potential resilience as a going concern (captured in the stand-alone credit profile), and to our view of the credibility of a successful resolution action on that bank if it failed (captured in our uplift for additional loss-absorbing capacity).

Our view of AT1 capital is fundamentally unchanged--it is highly subordinated and likely to absorb losses on a going-concern basis (particularly under solvency stress), with principal wiped out in a deep stress or failure scenario. We rate these instruments at least four notches below banks' stand-alone credit profiles--our assessment of their intrinsic creditworthiness—to reflect the significant additional default risk associated with these instruments relative to senior obligations. For complex groups, consolidated analysis remains the starting point for our assessment, but, as the high double leverage of CS shows, the distribution of capital around groups merits close attention.

On crisis management

The stresses in the U.S. and Europe provide important insights into the political and technical decisions that authorities may deploy as they deal with future challenges to financial stability. They confirm that financial stability is paramount, and public authorities will react with coordinated, rapid, and pragmatic action. Nevertheless, while the regulatory crisis management toolkit is broad, the use of tools will always be specific to the circumstances.

Faced with an ailing bank, regulators tend to prefer market solutions as opposed to failure and resolution, but they may be hard to achieve in practice. So what's the alternative? Since 2015, we have taken the view that the predictability of solvency support for failed systemic banks is uncertain in the U.S. and Europe. We conclude from events last year that an aversion to bailouts remains in the U.S. and Europe, and, when it's a credible option, resolvability offers an alternative path. The authorities used various tools to address systemic concerns, but direct bank solvency support to prop up failed institutions was not one of them. Rather than backtrack on these reforms, authorities are doubling down on making resolution work--therefore, resolution remains our central scenario for failing systemic banks in Europe and U.S. G-SIBs. Nevertheless, some resolution tools have not been used so far, and we monitor closely the authorities' efforts to further increase the predictability, reliability, and legal certainty of bail-in resolution. We will consider implications for senior creditors of banks if we see the default prospects as having changed due to the authorities' broader planned use of resolution, aided by resource enhancement at bank level.

Related External Research

  • Michael S. Barr, Review Of The Federal Reserve's Supervision And Regulation Of Silicon Valley Bank, April 2023
  • Basel Committee On Banking Supervision, Report On the 2023 Banking Turmoil, October 2023
  • Financial Stability Board, 2023 Bank Failures: Preliminary Lessons Learned For Resolution, October 2023
  • Australian Prudential Regulation Authority, Enhancing Bank Resilience: Additional Tier 1 Capital In Australia, October 2023
  • Bank of England, Bank Failures – Speech By Sam Woods, October 2023
  • FINMA, Lessons Learned From the CS Crisis, December 2023
  • G30 Working Group, Bank Failures and contagion: Lender Of Last Resort Liquidity and Risk Management, January 2024
  • Swiss Federal Council, Report On Banking Stability, April 2024

Related S&P Global Ratings Research

  • Swiss Federal Council Plans To Strengthen The Country's Too-Big-To-Fail Banking Framework, May 29, 2024
  • U.K. Proposal Would Unlock Resolution Funding For Small Banks, March 5, 2024
  • Your Three Minutes In Fintech: Social Media Could Catalyze Bank Runs, Feb. 29, 2024
  • Credit FAQ: How the U.S Proposes to Implement Basel III Capital Rules And The Impact on U.S Bank Capital Ratios, Jan. 11, 2024
  • EU Banking Package: Inconsistencies Temper Framework Improvements, Jan. 9, 2024
  • The Resolution Story For Europe's Banks: Making The Regime Fit For Purpose, Oct. 4, 2023
  • Swiss Public Liquidity Backstop Has Limited Implications For Hybrid Ratings, Sept. 18, 2023
  • How The EU's Bank Crisis Management Reforms Could Affect Ratings, April 25, 2023
  • Crisis Management Observations From Recent European And U.S. Banking Sector Volatility, April 19, 2023
  • European Bank AT1 Hybrids In A Post-Credit Suisse World, March 21, 2023
  • Bank Regulatory Buffers Face Their First Usability Test, June 11, 2020
  • Europe's AT1 Market Faces The COVID-19 Test: Bend, Not Break, April 22, 2020

This report does not constitute a rating action.

Primary Credit Analysts:Giles Edwards, London + 44 20 7176 7014;
giles.edwards@spglobal.com
Brendan Browne, CFA, New York + 1 (212) 438 7399;
brendan.browne@spglobal.com
Nicolas Charnay, Frankfurt +49 69 3399 9218;
nicolas.charnay@spglobal.com
Secondary Contacts:Gavin J Gunning, Melbourne + 61 3 9631 2092;
gavin.gunning@spglobal.com
Stuart Plesser, New York + 1 (212) 438 6870;
stuart.plesser@spglobal.com

No content (including ratings, credit-related analyses and data, valuations, model, software, or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced, or distributed in any form by any means, or stored in a database or retrieval system, without the prior written permission of Standard & Poor’s Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P and any third-party providers, as well as their directors, officers, shareholders, employees, or agents (collectively S&P Parties) do not guarantee the accuracy, completeness, timeliness, or availability of the Content. S&P Parties are not responsible for any errors or omissions (negligent or otherwise), regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data input by the user. The Content is provided on an “as is” basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENT’S FUNCTIONING WILL BE UNINTERRUPTED, OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs or losses caused by negligence) in connection with any use of the Content even if advised of the possibility of such damages.

Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact. S&P’s opinions, analyses, and rating acknowledgment decisions (described below) are not recommendations to purchase, hold, or sell any securities or to make any investment decisions, and do not address the suitability of any security. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a substitute for the skill, judgment, and experience of the user, its management, employees, advisors, and/or clients when making investment and other business decisions. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. Rating-related publications may be published for a variety of reasons that are not necessarily dependent on action by rating committees, including, but not limited to, the publication of a periodic update on a credit rating and related analyses.

To the extent that regulatory authorities allow a rating agency to acknowledge in one jurisdiction a rating issued in another jurisdiction for certain regulatory purposes, S&P reserves the right to assign, withdraw, or suspend such acknowledgement at any time and in its sole discretion. S&P Parties disclaim any duty whatsoever arising out of the assignment, withdrawal, or suspension of an acknowledgment as well as any liability for any damage alleged to have been suffered on account thereof.

S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain nonpublic information received in connection with each analytical process.

S&P may receive compensation for its ratings and certain analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, www.spglobal.com/ratings (free of charge), and www.ratingsdirect.com (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at www.spglobal.com/usratingsfees.

 

Create a free account to unlock the article.

Gain access to exclusive research, events and more.

Already have an account?    Sign in