Written by Kristian Lajkep, Consultant
A proverb says generals always fight the last wars. Apparently, the regulators always address the last crises. The post 2007-2009 crisis Basel standards (Basel III) were ideal at handling the kind of crisis that occurred in the late 2000s. The COVID 19 crisis was very different in its origins, nature, and outcome. Whereas in the great crisis of 2007, the banks were the source of the problems, during the current pandemic, banks remain relatively stable: no major bank has suffered a default or has required substantial government support. On the contrary, the governments are explicitly relying on the banking sector to fund the recovery. In this crisis, the Banking sector is neither the villain, nor a sick man, but a veritable protagonist.
This paper draws on the report released by the BCBS and asks to what extent has the bank resiliency been improved by the post-crisis Basel standards. In its report, BCBS has been able to compare figures from several different banks in a variety of different countries and conducted several quantitative analyses. The results are presented in this article.
In early 2020 a new respiratory pandemic (COVID 19) started to force the governments to impose precautionary measures, resulting in a closure often temporary, but sometimes permanent, of many businesses. The unemployment spiked and households decreased their consumption. Business owners started to draw on their existing credit lines to acquire liquidity. The demand for liquidity sored and the overall market liquidity deteriorated. A high selling pressure appeared even in the most liquid markets including those for the government bonds.
In response to this, the central banks have introduced numerous measures to contain the imminent crisis. Where possible, the rates were lowered, asset repurchase programmes extended, and liquidity support provided. The national governments have also rushed in to help with schemes to cover the lost salaries and loan guarantees. The regulators have sought to improve the ability of banks to survive and address the emerging crisis through:
By the end of Q2 2020, the financial markets were stabilized despite the ongoing closures in the real economy. Whilst the most volatile period of financial instability seems to be over, the pandemic and resulting uncertainty persists and will likely continue to the near future; therefore, we can’t say that the COVID crisis is entirely over yet. The banks will keep encountering the consequences of this pandemic for the years to come.
Compared to 2008, the banks entered the pandemic significantly better capitalised and with greater liquidity ratios. When provided the state guarantees they proved sufficiently prepared to accommodate a surge of demand for loans that occurred in first half of 2020 and thus to provide companies a bridge across the crisis. The payment services remained intact.
On the other hand, the bank equity value has dropped sharply in early 2020. The banking sector already entered the crisis with persistently low profitability, which was further worsened as the already meagre net interest margins were further decreased by loss provisioning. The credit spreads widened. The liquidity came under stress as the wholesale funding costs increased – particularly for lower-rated institutions. Most banks however had stable deposit franchises mitigating the wholesale funding costs.
The provisioning rates and concerns about the credit quality normalised by Q3 2020. Valuations and spreads returned close to normal by the end of 2020, although their recovery lagged behind that of the general market.
The bank resilience has been bolstered by public authorities’ support, when it will be withdrawn, more losses will emerge. Since much of the help the banks benefited from has been indirect (through the clients) it is hard to ascertain how important this help was for the banking institutions, and it could be underappreciated. Additionally, the government aid did not come free of costs, higher corporate and government debt will increase the aggregate level of risks in the future.
No major bank has defaulted as the result of the pandemic and banks’ capital, leverage and liquidity positions had remained strong. Whilst the governments and the central banks have been providing exceptional help to the economy, the major institutions themselves required no help. Banks were mostly aided through the help offered to their clients. The BCBS is of the opinion, that this is largely because the resilience of financial institutions has significantly improved since the adoption of Basel post-crisis reforms. It presents figures to support this assertion.
Between 2013 and Q4 2019 the total capital ratios and particularly the CET 1 have improved considerably for the sample of banks analysed by BCBS. The CET1 has increased on average by 300 bps since 2013, whilst the pandemic only managed to diminish it by less than 30bps by 30th June 2020. The Tier 1 capital thus decreased from a significantly higher position by relatively little. In Europe, the CET 1 ratio kept increasing at no slower rate and often faster. A large part of this must be attributed to payout restrictions, the reduced riskiness of portfolios due to loan guarantees, and differences in accounting practices regarding loan-loss provisioning. Likewise, the liquidity (measured through NSFR) Improved on average by 15% between 2013 and 2019 and kept improving, apparently unhindered by the pandemic.
A different and less optimistic view is acquired when considering the market-based capital ratios. The marked based capital has not seen much of an increase since 2013. This is likely due to the persistently low profitability of the banking sector. The onset of the pandemic instigated a large drop in the value of bank equity, which sent the market-based capital ratios plummeting down. The recovery of the equity value, together with some fiscal and monetary support allowed some losses to be retrieved. This may reflect a temporary sway in the investors' preferences but perhaps is more indicative of bank resilience and vulnerabilities than regulatory measures.
There was not a great difference between jurisdictions that have, and that have not yet implemented the Basel standards. The level of regulatory capital is only weakly correlated to the implementation stage and the level of liquidity was persistently high regardless of Basel implementation. However, the figures were collected from internationally active institutions that often try to meet the Basel requirements regardless of the regulation in their respective jurisdictions.
COVID 19 has to date not resulted in the failure of any bigger financial institution; BCBS has thus turned to CDS spreads to measure the resiliency of the banking sector in a more detailed analysis. The goal was to establish the relationship between the regulatory measures and changes in CDS spreads using the regression analysis. The regression analysis focuses on the increase in CDS spreads that occurred in March/April 2020.
Regardless of whether the Basel framework was implemented, the CDS spreads increased sharply as the pandemic begun and decreased back once authorities started intervening. The regression discovered that higher levels of CET 1 have indeed been related to lower increases of the CDS spreads. An increase of CET 1 by one standard deviation was related to a decrease of CDS spreads by approximately one-quarter of standard deviation. Interestingly, further analysis shows that CET 1 headroom is the main driver of this phenomenon. This means that banks that have capital above what is required are perceived as safer whilst institutions that dip into their required capital are viewed as risky regardless of the actual level of capital.
Conversely, there was no significant relationship between changes in spreads and either the leverage ratio or the LCR. This is likely because the major institutions did not come under a shortage of liquidity, which was readily provided by governments.
The analysis also shows that the retail-oriented business models were less impacted by the pandemic than banks with more capital markets-oriented business models. This likely relates to the increased costs of wholesale funding.
The banks’ vital role in the pandemic was to continue providing loans throughout the crisis. Particularly, the first half of 2020 was marked by a sharp increase of outstanding loans, driven perhaps by the borrowers drawing on their credit lines for precautionary reasons. The ability of banks to cope with this increased demand for loans was reinforced by extraordinary intervention measures of authorities in the form of loan guarantees or cash assistance.
The analysis reveals a correlation between higher initial capital levels (particularly CET 1) and the ability of a bank to sustain additional levels of lending during the pandemic. This additional lending usually takes the form of new lending, rather than the drawdown on existing credit lines. The cross country examination also showed that this sensitivity of lending to bank capital declined with increased fiscal support.
The advantage of capital buffers is that dipping into them should have relatively little effect on the ability of banks to operate. This should provide a relatively costless reserve for the banks to draw onto before needing to access their regulatory capital. The intended purpose of the buffers is to allow banks to provide their lending services during economic distress even when that distress starts eating into the capital.
In practice, during the covid pandemic, not very many banks made use of their capital buffers. If anything, banks in some jurisdictions even increased their capital headroom. In part this was because buffers were not necessary; the support measures, as well as bans on the distribution of dividends, left the institutions with abundant capital. In the rare cases where the capital was lacking, institutions still showed a peculiar restraint from dipping into their buffers.
Why the banks didn’t want to draw upon their capital buffers and whether they would if the situation was worse remains an open question. Going deep into capital buffers may be interpreted as signalling weakness. Additionally, the banks do not know how long the distress is going to persist or how quickly would the regulator require them to rebuild their buffers, so they may seek to dipping into their buffers as late as possible. For the same reason, banks tend to maintain headroom above their minimum requirements and buffers. Regulation offers only a part of the story, banks also consider non-regulatory aspects such as revenue targets, franchise value and risk appetite when constructing their capital position, which may provide a backstop well before the capital buffers are threatened.
The analysis shows that the banks with smaller headroom before the pandemic elected to restrain lending activities during the pandemic. The longer the pandemic persisted, the more apparent this phenomenon was. In the Euro area, all banks lowered their lending rates, but the amount by which the lending was reduced correlates with the capital headroom. Consistently the customers were switching from the banks that offered relatively high lending rates.
Several jurisdictions entered the crisis with positive countercyclical buffers that they could relax when the crisis became serious. Other jurisdictions were caught in the process of newly implementing or increasing countercyclical buffers, in such cases, the implementation was terminated. Yet other jurisdictions found different ways of lowering the cyclical buffers including lowering the required buffer for systemically important institutions or pillar 2 requirements. In some cases (ECB) the relaxation of buffers and particular of pillar 2 requirements was further compounded by enabling banks to meet a substantial portion of their pillar 2 requirements with a higher proportion of AT 1 and T2 capital. The purpose of these steps was primarily to increase the regulatory headroom of institutions so they would be more willing to expand credit supply.
The credit supply has materially (albeit modestly) risen in the jurisdictions that had previously implemented a higher countercyclical buffer and relaxed it during the pandemic. This is supported by the quantitative analysis in which BCBS found that an overall decrease in required CET1, whether due to relaxation of buffer or Pillar 2 requirement was related to a positive increase in lending. A relaxation of planed buffer implementation had no effect on the credit supply.
As only the Countercyclical buffer is explicitly releasable, it was probably the most efficient way of increasing the capital headroom during the crisis.
The liquidity buffers have been introduced to allow the banking sector to absorb shocks caused by financial and economic stress, thus reducing the risk of spill-over from the financial sector to the real economy. In the current pandemic, the cause and effect were reversed: it was the closure of the real economy that was a potential threat to the stability of the financial sector. In this sense, the pandemic gives no particular indication of the potential effectivity of liquidity buffers in their intended purpose.
Nevertheless, some banks have withstood short-lived pressure on their liquidity positions. Others perhaps operated securely in knowledge that they have sufficient buffers of high-quality liquid assets. At the climax of the pandemic, banks were experiencing downward pressure on their liquidity positions with their clients drawing on their credit lines to raise cash. In some cases, the drawdowns surpassed the prescribed LCR outflow rates. However, often, this pressure on LCR was mitigated by an increase in high-quality liquid assets as corporate entities have re-deposited the proceeds.
In the third quarter, there was an increasing pressure to buy back debt originating from disruption in financial markets as investors shifted away from investment vehicles that invest in bank liabilities. These vehicles were thus less willing to roll-over the short-term investments, which had a negative impact on bank funding. Moreover, these vehicles were unprepared to meet the surge in redemptions and turned to bank issuers to repurchase their debt. Although under no obligation to do so banks often met those requests to preserve the relationship.
Two-thirds of jurisdictions noted that the deposit inflows, due to the “flight to quality,” have counteracted some of the liquidity outflows. Another significant mitigant were the actions taken by the central banks such as asset purchases, lending facilities and the reduction in reserve requirements.
The extent to which the institutions felt the downward pressure was dependent on banks’ funding models, exposures to stress factors, and the extent to which they were positioned to benefit from mitigating factors. The banks that relied on unsecured wholesale money markets to fund corporate loan portfolios could experience notable pressure, whilst the banks that had a large deposit franchise (particularly if in the same currency as their loan portfolio) were not affected. After the central banks have engaged in liquidity measures, the liquidity pressure gradually subsided.
Banks proved reluctant to draw on their liquidity buffers, especially if it would lead to a drop in the LCR below 100%. Their response to the relatively minor liquidity pressure was not entirely proportional: the banks were observing their own internal liquidity targets that were part of their liquidity risk management frameworks and were significantly stricter than those required by the regulator. This overreaction however did not appear to have a significant adverse effect. Most commonly, they borrowed extra money from the central banks via standing or extraordinary facilities. The wholesale or retail sources still performed their function for albeit at higher rates. Some of these actions, such as restrictions in credit supply or spreading the concern in the interbank market could have been potentially harmful, nevertheless, they had little effect on bank’s ability to address the dash for cash.
Unlike in 2008, during the COVID pandemic, the crisis struck at the real economy through which it affected financial institutions. A number of current safeguards were designed to prevent spill-overs in the opposite direction: from the banking sector to the real economy. As such, their adequacy was not tested by this crisis.
Other safeguards, capital buffers and liquidity ratios were put in place to ensure that the economic distress will not send the bank to a downward spiral at worst and at best would not interfere with day-to-day operations. They performed well during this crisis, however, it should be noted that due to the concerns about internal ratios and reputational loss as well as loss of relationships with their investment vehicles the banks have displayed some ability to send themselves down the downward spiral, even if there is no regulatory or financial reason to.
It seems that the internal policies and reputational concerns amidst banking institutions as well as the pressure in the wholesale interbank market are often just as important, just as dangerous, and far less forgiving than the regulation.