Nine months since the start of the pandemic, we argue that despite increasing provisioning costs and declining net interest income, the banking sector is relatively better placed to withstand the potential economic fallout of Covid19 than it was the financial crisis of 2007-12.
Starting with Capital Adequacy, the most fundamental measure of a bank’s ability to withstand stress, bank core equity tier one (CET1) ratios pre-crisis averaged 7.5% of risk weighted assets in Europe in 2007, but for GSIB’s and other large banks the ratio was lower, at around 6%. Due to Basel III’s reforms these ratios are today more likely to be at least 12%, and in many cases 14% and upwards.
Key Stress Test Assumptions 2018-19
|National GDP fall||8%||8.3%||4.7%|
|House price falls||25%||19%||33%|
|CRE price falls||35%||20%||41%|
|Equity price falls||50%||22%||50%|
Complementing the much-improved capital ratios of the sector are the periodic stress tests performed by regulatory authorities in the USA, Europe and elsewhere. The objective of these annual exercises is to ensure that large banks can survive stressed environments whilst retaining minimum levels of capital. Examples of the key stress test assumptions are provided in Figure 1. The severity of these assumptions is often based upon the 2008-10 recessionary period. However, some variables, e.g. GDP growth and in some countries unemployment, look as though they will be quickly exceeded by the economic fallout resulting from Covid-19. In response to this weaker than expected environment, the sector appears in some cases to be taking prudent action to further buttress capital levels. For example, in the US & UK bank supervisors have restricted dividends and share buybacks for 2020.
To back up enhanced capital levels, formal bank resolution regimes have also been established in many parts of the world to facilitate orderly recapitalization or wind-down of failing firms. In Europe the Bank Resolution and Recovery Directives (1&2) and in the USA the Orderly Liquidation Authority section of the Dodd-Frank Act, provide resolution authorities with clearly defined powers to resolve failing firms whilst maintaining financial stability. In parallel resolution eligible liability requirements, (although not fully complete in many parts of the world, particularly for smaller firms) have been in place at a transitional level for globally systemic important banks (G-SIB’s) since 2018. Such liabilities are likely to be ultimately at least equal to loss-absorbing capital for international and domestically systemically important firms, providing a clearly identified pool of liabilities to provide recapitalization in the event of failure.
Most commentators are agreed that Asset Quality is the most likely transmission method of the pandemic to bank solvency. The interruption of economic activity in Q2 caused recessions in many parts of the world that were off the scale of living memory, and in some cases centuries. Although Q3 often resulted in a healthy “bounce”, the potential remains for economic distress to impose credit losses in a manner and on a scale not experienced previously, particularly where renewed and prolonged restrictions damage economic capacity permanently.
Starting with consumer debt, government support for laid-off workers and forbearance measures have so far cushioned the recession. This cannot continue ad infinitum though, and as these schemes are withdrawn from Q1 2021, increases in defaults are inevitable. The impact may be highly segmented though. Residential property prices are at record highs in many countries, and this may help soften the blow for lenders. Unsecured and lower-quality secured lending may well be the source of most losses in this sector.
For corporate lending, as ever in a downturn, the risks will lie in portfolio concentration and correlation. Banks heavily exposed to airlines, leisure, retail, and tourism may suffer disproportionally higher losses in their corporate lending than capital models predict. Commercial property looks even sicker than usual in a downturn, as the pandemic may have accelerated the decentralised offices and online retail rendering many prime city centre buildings obsolete. The level of government intervention in particular industries and the economies in general may be a critical driver of eventual credit losses in the banking sector.
One important factor currently directly influencing capital levels is the recognition of loss provisions. This can be highly variable across jurisdictions and individual firms. In Germany under IFRS 9 Deutsche Bank moved away from so-called “point in time” loss models to a more “through the cycle” view. This should avoid extreme volatility in its allowances and capital, but Deutsche still expects provisioning to be €2.0 billion in 2020 for a portfolio which performed strongly during the last recession.
In the USA, where pending accounting standards will require full lifetime provisioning, the sector provided $73 billion of losses in HY1 2020, although Q3 saw much lower levels of $5.4 billion. Ultimately, whether prudent, early recognition of losses or a more phased approach is appropriate in these unique circumstances depends upon whether Covid-19 is viewed as a temporary or more lasting impact upon economic output.
One of the challenges posed by the pandemic for Management has been the lack of availability of traditional channels of customer interaction such as branches or even call centres. One likely permanent outcome for the sector is that firms which had strategically embraced technology earlier,and were able to provide better service with staff working from home, are likely to win a lasting competitive advantage. This logic can also be applied to other technology-led initiatives in the sector such as cloud data storage, artificial intelligence, fraud, cyber-security, and operational resilience.
Another factor likely to lend greater stability to the banking system than the is the greater emphasis placed on stable funding and liquidity. The Basel III liquidity coverage ratio now requires all banks to hold sufficient high-quality liquid assets to withstand a 30-day stress including a complete lock-out of wholesale and currency funding. Whilst this has resulted in much greater levels of liquid assets, it has, along with the net stable funding ratio, dis-incentivised the use of short-term wholesale funding sources that were such a means of systemic transmission of the 2007-8 crisis. In place of the short-term wholesale funding banks have largely funded themselves with either more stable deposits or long-term wholesale sources, taking advantage of flatter yield curves.
Another reduction in systemic risk has occurred through the extensive re-regulation of over the counter (OTC) derivatives. Although central counterparties (CCP’s) for derivatives existed in 2008, the requirement for plain vanilla contracts across interest and credit index trades means that their ability to “fire-break” counterparty defaults is now much more significant. For OTC contracts which are not centrally cleared, regulation of margin levels is similarly designed to protect bilateral counterparties much more effectively.
Despite these improvements in the structure of balance sheets, risk management and financial system architecture, concerns do remain about firm’s earnings potential. Revenues thus far have held up relatively well, although provisioning levels have risen as described above. The largest long term-threat appears to be market risk arising from depressed levels of interest rates. US Treasury yields have halved from pre-pandemic levels, and Eurozone bond yield curves are negative yielding beyond the 10-year benchmark. The average exposure to a 1% fall in rates for large global banks is estimated to be less than 1% of CET1, but this could be as much as 10-15% on an economic value of equity basis where rates shift permanently lower.
However, the recent rise in UK Gilt yields in response to what appears to be a permanent post-pandemic structural deficit, provides a reminder of the risks of rapidly rising rates. Bank’s exposure to rising rates are typically greater than falling, and are transmitted through FVOCI liquidity portfolios and cashflow hedging reserves. Defending a falling currency in the face of rising inflation could become very costly for a country’s banking sector and would also feed into rising credit risk in the form of payment shock to retail and corporate borrowers. A falling currency could itself pose significant capital adequacy stresses where banks have significant overseas operations.
Covid19’s effect on the financial sector is as yet unknown, and ultimately may be dependent upon the size of assistance that governments are willing (and able) to provide to their economies. Many uncertainties exist, not least the ability of banks to earn revenues and recapitalise in an environment where loss provisioning is driven upwards, and yield curves may become a long-term flatline.
However, the measures that we have outlined above to increase capital adequacy and liquidity, reduce systemic risk and improve resolvability can be argued to at least have “mended the roof whilst the sun shone”, and leaves the sector as well placed as could be expected to face the uncertainties of 2020 and beyond.