Future of RAF No.2
Some lessons we learned from the recent bank failure events
In March this year, for the first time since the last GFC, the market was enveloped in tension bracing for a storm because bank failure events occurred one after another in the U.S. and Europe. First, in the U.S., Silvergate Bank, well-known in crypt-currencies businesses, decided to suspend operations due to the unstoppable outflow of deposits. The outpouring of deposits then spread to Silicon Valley Bank (SVB) and Signature Bank, and these two banks suffered bankruptcy (although their deposits were fully protected). In addition, several other regional banks that have recently experienced rapid deposit growth, such as First Republic Bank, faced a bank run but have managed to hold on to their minimum liquidity to this point, thanks to liquidity support from the authorities and major banks.
SVB's assets totaled $209 billion, making it the largest U.S. bank failure after the GFC and the second largest in total after Washington Mutual, which failed during the GFC (Kodachi (2023)). In addition, the federal government enacted a "systemic risk exception" to protect all deposits in the event of regional bank failures. The FRB and others provided emergency liquidity support to other regional banks. These events surely reminded us the ghosts of GFC, which were once almost concurred by Basel III, have not yet gone.
To make matters worse, the regional bank crisis in the U.S. sparked one of Europe's G-SIBs, among the top 30 banks that are too significantly important to die disorderly. The bank that caught fire was Credit Suisse, based in Switzerland. Despite its long history as a prestigious Swiss bank, the bank has recently been plagued by misconduct events and poor earnings. In this sense, the outflow of deposits began even before the U.S. regional bank crisis. Still, even though it was not directly related to the U.S. regional bank crisis, the outflow of deposits accelerated simply because of the heightened sense of uncertainty about the bank management. Furthermore, the fact that the Swiss authorities provided liquidity support to the bank led them to conclude that it was facing an emergency overpassing the PONV (point of non-viability) and forcing it to sell to its peer UBS thereby avoiding its failure. The controversial twist at this time was the total loss of AT1 bonds (the most senior bonds, but liabilities that were originally expected to absorb losses only after equities) before the total loss of equities.
The U.S. authorities' decision to provide full deposit protection to regional banks, as well as the Swiss authorities' very early decision to give up on Credit Suisse management and the measures taken to avert the bank's bankruptcy, which reversed the expected order of repayment, have been criticized in various ways from the perspective of policy adequacy. However, their quick responses to the events should be highly appreciated. These immediate actions contrasted favorably with the U.S. authorities' lukewarm response in fear of criticism from politicians to the failure of Lehman Brothers in 2008, which triggered the global financial crisis. Regardless of what was discussed in advance and what responses were prepared, the situation on the battlefield changes constantly, and the decision by the authorities to respond flexibly to this situation and avoid the spread of the crisis is commendable.
Analysis of the Causes of the Bank Failure Events
Even if the authorities’ responses were quick and right, whether the existing framework to manage them was correct is another issue to be discussed. For this, it is necessary to analyze the causes of the current crisis and review how the current supervision and regulation address them accordingly. The causes of the crisis can be divided into the U.S. and Switzerland and summarized as follows.
The cases of the United States
Direct causes (primary): An unusually rapid outflow of deposits against a backdrop of anxiety spread via SNS. The speed is so rapid that the authorities had difficulties in preparing liquidity support in time.
Direct causes (secondary): The rapid accumulation of deposits in a money glut over the past few years and the resulting increase in the security-to-deposit ratio, as well as the rapid increase in interest rates since last year, which led to higher valuation losses on securities holdings and outflows of deposits into MMF.
Indirect causes: Deregulation under the Trump administration for medium-sized banks with assets ranging from $1,000m to $2,500m, delays in implementing Basel III, and possibly weak supervision by U.S. supervisors (e.g., S.F. Fed) over the mid-sized banks.
The cases of Switzerland
Direct cause (primary): Accelerated outflow of deposits triggered by growing concerns about banks in the wake of the U.S. regional bank crisis (the largest shareholder's refusal to contribute additional capital added fuel to the fire). While providing liquidity support, the Swiss authorities quickly gave up on the bank's management (even though its capital adequacy ratio, LCR, and other liquidity indicators were well above the required minimum levels) to avoid the worst-case scenario of a bank failure.
Direct causes (secondary): The bank's bad reputation resulting from poor management and a series of misconduct events over the past few years, as well as "material weaknesses in internal control over financial reporting" announced a few days before its crisis, created a strong image that was vulnerable to crisis.
Indirect causes: The Swiss authority lacks enough capacity to manage the crisis of banks with assets close to the size of the GDP.
Assessment of Policy Responses Based on Cause Analysis
As a next step, based on the above cause analysis, let me assess the policy responses and any remedies to overcome their possible weaknesses.
First, we need to address the rapid outflow of deposits, which was cited as a direct cause (primary) in both cases in the U.S. and Switzerland. Basel III had already focused on liquidity risk, so the LCR and NSFR were introduced. Unfortunately, the LCR and NSFR had yet to be introduced for medium-sized regional banks in the U.S. However, even if they had been introduced, they would have been ineffective against such a rapid outflow of deposits (in the case of SVB, it is estimated that as much as 80% of all deposits could have been withdrawn in just two days (Nihon Keizai Shimbun (2023)). In addition, the fact that both the LCR and NSFR were applied to Credit Suisse and were well above the minimum required levels until just before the crisis suggests that the current regulations may not be able to handle a modern-style liquidity crisis when it occurs.
Some believe the LCR and other parameters should be slightly adjusted. For example, it is said that most of the deposits quickly drained this time are uninsured corporate deposits (especially those that have recently increased). So it might be better to assume a much lower stickiness of such deposits for calculating the LCR. However, the degree of stickiness of such deposits may vary greatly from business to business and country. It would be not easy to factor them into Pillar I of Basel III, a common standard for all countries. Rather, it may be preferable to consider them into the Pillar II through strengthening stress testing.
In addition, BOE Governor has hinted at raising the maximum amount of insured deposits in the United Kingdom (Bailey (2023)). However, this adjustment could lead to moral hazard among depositors, making it difficult to derive an optimal level of the insured deposit amount.
In this context, I am personally interested in the idea of the minimum balance at risk (MBR) proposed by the staff of the New York Federal Reserve Bank (Cipriani et al. (2023)). This concept is intended to reduce the motivation for "runs" and allow the authorities to "buy time" during a crisis by setting a certain penalized limit on the speed of deposit withdrawals concerning uninsured deposits. For example, if a depositor tries to withdraw all of his uninsured deposits, up to 95% can be withdrawn immediately (relative to the average balance over a certain period), while the remaining 5% can only be withdrawn after a certain period. The 5% of deposits would then be earmarked so that (if the bank fails before a certain period) they would be subordinated to other normal uninsured deposits in terms of repayment in the event of failure. These ideas have already been incorporated into the MMF liquidity crisis response promoted mainly by the U.S. and the FSB.
For example, we may soon see an era in which A.I. locates bad rumors of a certain bank on SNS and automatically executes the withdrawal of deposits. It is necessary to build an "incentive compatible" system now that can withstand such an era.
The IRRBB and stress tests under Basel III should address the increased interest rate risk in the banking account associated with a higher security-to-deposit ratio (direct causes in the U.S.). The IRRBB is not applied to medium-sized regional banks in the U.S., and stress tests are conducted once every two years due to deregulation. Still, even in this environment, some proactive measures could have been taken if supervisors had kept a closer eye on the excess risk. The risk of regulatory capture (indirect causes in the U.S.) could be a reason for concern.
Next, let me look at the direct cases (secondary) in Switzerland. They are a governance problem at Credit Suisse. Such a problem has persisted for such a long period due to the same regulatory capture as in the U.S. or the limited capacity of the authorities themselves (indirect causes in Switzerland). If the authorities are "too small to manage G-SIBs," it may be necessary for the bank to change its home country and is subject to the supervision of another country’s authority or try to split up the bank in question. Unfortunately, the merger between UBS and Credit Suisse, while helpful in overcoming the current crisis, is not conducive to the sustainable management of G-SIBs in the future (particularly considering that the combined bank's assets will be more than twice the size of Switzerland's GDP).
(References)
Kei Kodachi, "U.S. SVB Failure and Banking System Instability: Background Analysis and Tentative Discussion Points," Nomura Institute of Capital Markets Research, March 24, 2023. The paper was written in Japanese and the title was translated into English by the author.
Nihon Keizai Shimbun, "Digital version of 'mounting racket': Impact of 80% outflow of deposits in two days," Nihon Keizai Shimbun, April 17, 2023. The paper was written in Japanese and the title was translated into English by the author.
Bailey, Andrew “Monetary and Financial Stability: Lessons from Recent Times – speech by Andrew Baily” Bank of England, 12 April 2023
Cipriani, Holscher, McCabe, Martin and Berner,“Mitigating the Risk of Runs on Uninsured Deposits: the Minimum Balance at Risk” Federal Reserve Bank of New York,14 April 2023
