Model Validation: are you sure your internal risk parameter estimates are adequate, robust and reliable?
Control your model risk with the right tool to develop and validate models, bridge the talent gap and reduce implementation risks
How confident are you that your risk models show what you need to see? Would your regulator agree?
Written by Guillermo Monje, Consultant
and Can Soypak, Principal Consultant
The effects of the COVID-19 pandemic and the following lockdown on the real economy are expected to be significantly worse than those of the last financial crisis. According to OECD projections for 20201, the Eurozone’s GDP is expected to decline by 9.1%, if there is no second wave, and by 11.5% if there is. In this context, it is very important to analyse what conditions are the banks likely to face.
To begin with, their profits are likely to decrease due to lower general economic activity. Furthermore, because their clients’ risk increases, they must adjust their expected losses estimates and, consequently, have an incentive to decrease credit supply. Also, due to their systemic importance, they face constant regulatory and policy changes, which present further challenges to their operational continuity.
This article analyses the effects of the COVID-19 from a bank’s credit risk point of view. First, it assesses how COVID-19 hurts the economy, the way the risk profiles of stock and potential clients deteriorate due to the pandemic’s negative effects on the economy, and it shows how this lowers credit supply. Then it elaborates on the regulations and policies – which basically target the amount of risk that banks can take on - aimed at preventing credit shortages and the results they had. Then, it explains the challenges that banks’ internal credit risk frameworks consequently face, and details valuable partners to help tackle them.
Covid-19 is a new virus, more contagious and deadlier than normal flu. Due to its rapid global infection propagation, it is considered a pandemic by WHO. Given that it is a new virus, there is no cure and the probability of requiring intensive medical care or dying after contracting it is considerable. Since most people do not want to risk their lives and governments want to avoid a healthcare system collapse, voluntary and mandatory lockdowns have been put into place to reduce the rate of spread of the infection. As expected, this greatly hurts the economy; the industrial production considerably diminishes - Eurostat reports that, until June, the total cumulative decrease in industrial production in the EU since February amounts to 18.8%2. This creates a negative shock to aggregate demand because firms’ lower earnings and investment lead to unemployment, in turn leading to less spending. The International Labour Organisation estimates that 12 million full-time jobs will be lost by the end of the year in Europe3. The service industry, especially tourism, will be worst affected. It is estimated that the EU’s tourism sector will lose 1 billion € per month and that global international tourist arrivals could decline by 20% to 30% this year4.
Banks, constantly monitor risks in order to conduct their activities, which are correlated with economic activity. Because covid-19 strongly damages the economy, it is important to think about the ways in which this pandemic might challenge the banks risk management activities and framework. Whereas a full journal paper could be written about this topic, this article mainly deals with the deterioration in the risk profiles of stock and potential clients.
Regarding the stock of clients that have loans with a given bank, the pandemic and consecutive restriction measures automatically increase their overall level of risk. More specifically, all the components of the expected loss they were assigned when underwriting their loans instantaneously become larger. Many clients that were given loans at a moment when there was no pandemic will face financial stress and their probability of default will be higher than initially estimated. Furthermore, because their cash flows will face pressure, the drawing amount of their available credit lines will most likely increase, rendering CCF estimations insufficient and, thus, their exposure at default higher. Additionally, as they will be struggling to maintain the amount of funds they need on a regular basis, chances are that, if a default occurs, their loss given default will be higher. With regards to potential clients, the scenario is quite similar, although from an ex-ante point of view. Because economic conditions have worsened, banks now know that their available estimates of the probability of default, exposure at default and loss given default do not match the current context and need to be increased. Hence, banks have incentives to constrain credit supply.
Banks have a strong procyclical behaviour, which means that, when the economy is doing good they increase lending and, during recessions, restrict the amount of credit they are willing to grant5. One of the reasons behind this is that banks constantly monitor the level of risk of their obligors, which they consider lower during economic upturns and vice-versa6. Credit supply is a key variable for the economy, as it directly impacts the financial cycle, which is measured as the ratio of domestic credit to GDP. In this context, it plays a significant role in the generation or prevention of financial crises7. Recessions that arise when the financial cycle is contracting are usually more severe, as during such phases the decrease of GDP can be expected to be higher by a factor of 50%, relative to financial cycle expansion phases. Finally, it is also demonstrated that access to credit positively affects GDP growth8. Consequently, to ensure that credit supply does not drop is a key to fighting off the negative economic consequences of the pandemic and to prevent them from being even more severe. In a nutshell, making sure that the availability of credit is not negatively affected cushions the decline of GDP and helps to maintain financial stability, in a context where the pandemic increases volatility in the stock markets9.
Aware of how important credit supply is during crises, the governments and regulators have set up a regulatory framework, aimed at maintaining it during economic distress. Some of them have an ex-ante approach, whereas others were implemented once the pandemic was already active. As of June of 2020, it can be stated that so far, they have been successful.
Among the most important ex-ante regulations are the capital requirements regulation and stress testing guidelines, mainly based on Basel III. Under these rules, the banks are required to keep ratios of capital relative to risk-weighted assets (RWA) to cover future losses. The first buffer mandates banks to have a capital to RWA ratio equal to 4.5%. On top of that, two extra buffers are included. A capital conservation buffer that requires an extra 2.5% as a proportion of RWAs and a discretionary counter-cyclical buffer which can require as much as an extra 2.5% of capital. The European Systemic Risk Board also mandates an additional buffer for systemically important institutions. Stress testing requires simulations to be carried out to check what losses banks would experience in extreme situations, thus helping them to prepare for such events.
However, capital requirements may also have a procyclical effect since they restrict credit availability when it is most needed; the riskier the clients, the more capital is required and less funds are thus available for lending, worsening a recession. Authorities also responded to this by announcing many – so far 151 – flexibility measures regarding different capital and reporting requirements, as well as guaranteeing loans, following the onset of the pandemic. For example, on the 12th March, the ECB announced that it would lower capital required for systemic institutions. Furthermore, on the 20th March, it announced that banks would benefit from public guarantees for NPLs and from flexibility to determine which clients are classified as “unlikely to pay”. On the 19th June, the ECB indicated it would ease rules for banks to lend to households10. In this context, both the European Commission and ECB concluded that banks face this crisis robust and well capitalised11 12. Furthermore, it can be concluded that the regulatory flexibility measures had helped to partially offset the crisis, although uncertainty remains, and some sectors are facing a decrease in credit supply13. The July 2020 Bank Lending Survey states that, so far, the Covid-19 crisis has had much smaller negative effects than the last financial crisis. In fact, the net percentage of banks indicating that they expect to reduce credit to the enterprise sector equalled only 1% in 2020 Q2 and 4% in 2020 Q1. Nonetheless, net percentages of banks reporting credit tightening for housing purchase or consumer lending was considerable 22% and 26%, respectively.
It should be noted, however, that banks’ internal credit risk frameworks face various challenges, some of which may require some external help.
Even though the banks came to this crisis well prepared and were so far able to withstand the storm, it is key for them to be able to estimate the expected loss parameters (PD, EAD, LGD) as accurately as possible. If expected losses are overestimated, a bank might keep too much capital which might be effective in preventing future losses, but it will decrease its profitability, particularly considering the negative ECB rates. Alternatively, an underestimation of risks might lead to good profits when there is no crisis but endanger the bank during crises.
Additionally, adjusting the expected loss’ parameters to the new and uncertain scenario is key to managing the stock portfolio and the potential credit requests. Both approval and pricing policies must be adapted. Integrating relevant macroeconomic forecasts and uncertainty with the bank’s fundamentals to develop a meaningful and flexible policy framework is invaluable. In this sense, It is important to emphasize that current credit state guarantees can generate a problem of moral hazard and, thus, making use of different perspectives regarding filtering “good” from “bad” clients may be expedient.
Similarly, developing adequate stress tests, both ex-ante and ex-post the pandemic became one of the hot topics in the banking sector. These tests give a good approximation of how the bank will operate under extreme conditions and, thus, allow for adequate preparation. It is very important for banks to carefully integrate their fundamentals with negative macroeconomic indicators and uncertainty, to determine realistic impacts for them, and to avoid excess or lack of capitalisation.
Moreover, the current regulatory environment can be difficult to track, due to the frequent policy introductions and regulatory flexibility measures implemented. Tracking all the different regulatory changes as effectively and efficiently as possible is of utmost importance, both to take as much advantage as possible to secure continuity of operations, as well as to avoid costly fines or take measures that would lead to a comparative disadvantage. Quickly changing regulatory landscape may, in the short term, be the worst challenge that the (often understaffed) departments of various banking institutions face.
This article assessed the impacts that banks face due to the economy’s deterioration because of the COVID-19 pandemic, from a credit risk point of view. Specifically, it elaborated on how their stock and potential portfolio risk increases, inducing them to constrain credit supply, and how authorities’ ex-ante and ex-post regulations addressed this. Finally, it showed challenges that banks are likely to experience in this context. Banks should put in as much effort as possible to tackle them. Experience and state of the art practices to estimate losses and design stress tests, meaningful pricing frameworks development and efficient tracking of regulatory updates should be a priority to develop, both internally, as well as with the help of competent partners.
1 www.oecd.org/economic-outlook/june2020/ 2 ec.europa.eu/eurostat/statistics-explained/index.php crisis_on_industrial_production 3 www.euronews.com/2020/04/22/theeu-s-100-billion-euro-scheme-to-tackle-unemployment-caused-by-covid-19 4 www.europarl.europa.eu/RegData/etudes/ATAG/2020/649368/EPRS_ ATA(2020)649368_EN.pdf 5 Becker & Ivashina (2014) 6 Bunn & Redwood (2004) 7 Borio (2014) 8 Rajan & Zingales (1998) 9 Onali (2020) 10 www.europarl.europa.eu/news/en/ press-room/20200615IPR81234/covid-19-easing-rules-to-encourage-banks-to-lend-tocompanies-and-households 11 ec.europa.eu/commission/presscorner/ detail/en/qanda_20_757 12 https://www.ecb.europa.eu/stats/ecb_surveys/bank_lending_survey/html/ecb.blssurvey2020q2~d8de5b89f0.en.html#toc3 13 www.ecb.europa.eu/stats/ecb_surveys/bank_lending_survey/html/ecb.blssurvey2020q2~d8de5b89f0.en.html