Managing credit risk in a COVID-19 environment—a challenge and an opportunity

The COVID-19 crisis is expected to strongly affect bank asset quality in the medium term.



Jean Philippe Peters - {Sponsoring} Partner - Risk Advisory - Deloitte

Elena Petrova - Director - Financial Advisory - Deloitte

Arnaud Duchesne - Director - Risk Advisory - Deloitte

Published on 4 February 2021

Share this article


The COVID-19 crisis is expected to strongly affect bank asset quality in the medium term. Even on the assumption that a major second wave of the virus could be avoided, Moody’s Investors Service predicts a global marked rise in the proportion of nonperforming loans (NPLs) in bank loan portfolios from 2019 to 2022, ranging between 1 and 3 percent of total loan exposure per year in France, Germany, Italy, Spain and the United Kingdom (UK).

At the start of the crisis, most countries quickly implemented a string of measures to shore up liquidity. Governments allowed late payments of various taxes (income and social security taxes, VAT, etc.), banks provided payment moratoria to struggling borrowers, and governments and commercial banks collaborated to provide state-guaranteed emergency loans.

These instruments, combined with the coordinated actions of central banks, were successful in avoiding a vicious circle of uncontrolled bankruptcies. However, they will not prevent banks from facing a higher level of NPLs in the medium term,

as the effects of higher unemployment rates and the adverse gross domestic product (GDP) shock will eventually take their toll.

To date, while no significant increases in actual NPL levels have been observed, loan loss provisions have already skyrocketed. For example, the Bank for International Settlements reports that loan loss provisions of 15 major syndicated loan-arranging banks globally have almost doubled from the fourth quarter of 2019 to the second quarter of 2020. This places additional pressure on the profitability of the banking system, which has already been suffering from low-interest rates and a growing regulatory burden.

Loan loss provisions by 15 major syndicated loan-arranging banks globally, where Q1 2019 is indexed to 1

Source: BIS

The current pandemic also follows a period when corporate loan markets, buoyed by a growing economy and low-interest rates, increasingly adopted more aggressive structures where higher leverage was paired with reduced lender protection. This will undoubtedly pile even more pressure on expected recoveries and amplify potential loan losses in the future.

So, how can banks manage medium-term credit risk in this volatile COVID-19 environment?

First, banks should ensure that they have a clear understanding of the key underlying drivers of credit risk in their loan portfolios. They should be able to identify material exposure segments (e.g., by country, industry, internal credit rating and collateral available—and ideally a combination of these) and the main external drivers that would lead to a credit quality decrease for these segments (e.g., commodity and real estate prices, GDP declines, declines in consumer spending or higher unemployment).

It is important to note that a risk monitoring toolkit like this could also become the founding pillar of a credit early warning system, such as the one detailed in the European Banking Authority’s “Guidelines on Loan Origination and Monitoring”. Some banks will find this an analytical challenge, due to certain limitations of the data directly available and a lack of historical statistics and comparisons. However, we believe banks should invest in upgrading, refining and streamlining their internal risk monitoring and processes as soon as possible, saving costs as a result.

To put this into practice, banks could devise a monitoring system at the individual borrower level where both the macroeconomic drivers (top-down approach) and borrower-specific variables (bottom-up approach) would be considered. For example, the deviation of financial results from initially set budgets of a borrower would raise a red flag when looking from the bottom up. Banks could also run simulations to estimate the impact of higher unemployment rates on existing credit portfolios (top-down approach).

While this would be a powerful tool, it would require the development of standardized and semi-automated systems and templates to efficiently support this analysis, as the current volatile environment requires these simulations to be more frequently reviewed and adjusted. Cross-checking the results with market-implied data on expected defaults (e.g., by using the Merton model) may also provide additional data to help banks validate their assumptions.

This ongoing monitoring of early warning indicators, compounded with an appropriate governance and escalation structure, should automatically drive banks to review cases of increased credit risk and implement concrete actions at the appropriate level. These actions can be performed at the segment or the individual borrower level.

At the segment level, the bank may choose to hedge risk away or dispose of parts of their credit portfolios to manage their overall exposure to a given industry (e.g., restaurants and hotels in the current pandemic). The growing private debt and securitization markets are key to these disposals. According to Preqin, total assets under management (AUM) for private debt players rose by 168 percent in 2019—from USD315 billion in 2010 to USD845 billion. Preqin also expects this growth to continue, with a 73 percent increase in AUM to USD1.46 trillion by 2025. While disposals of NPL portfolios have become much more prevalent and standardized since the last crisis, the bank’s segmentation approach would still be key to selecting the relevant portfolios for disposal.

If risk is managed at the individual borrower level, the bank may begin early discussions with certain borrowers who are experiencing decreasing credit quality to discuss management plans, liquidity options or similar arrangements. These discussions could result in a wider range of mutually acceptable options, such as the provision of additional collateral and temporary covenant waivers, allowing the borrower to continue investing in the business and remain a going concern.

All of which leads to another particular challenge in the current environment. As credit availability is a key driver of economic growth, there is a need to balance governments’ desire for banks to continue financing the real economy while also remaining disciplined about the risk they take on their balance sheet. Therefore, any action taken by banks will need to be assessed not only from an economic angle but also from a social and reputational one.


  • No level of preparedness and proactive measures can completely prevent banks from experiencing increasing NPLs in the current economic situation.
  • However, compared to the last economic crisis, we believe that banks today have more options at their disposal.
  • To access these options, banks should invest in particular risk monitoring tools.
  • The banks that invest early will reap multiple benefits in the short and long term, including lower running costs, improved earnings predictability and risk management, and better regulatory compliance.

Share #DeloitteInsideNow


Credit Risk Services – Deloitte Assistance

In a constantly changing regulatory environment, it is imperative for Financial Institutions to have a clear understanding of how economic and regulations environment affect their Credit Risk profile in a context of a robust risk management framework.

© 2021. See Terms of Use for more information. Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee (“DTTL”), its network of member firms, and their related entities. DTTL and each of its member firms are legally separate and independent entities. DTTL (also referred to as “Deloitte Global”) does not provide services to clients. Please see to learn more about our global network of member firms. The Luxembourg member firm of Deloitte Touche Tohmatsu Limited Privacy Statement notice may be found at