On 2 November, the European Banking Authority (EBA) published the results of stress tests designed to measure how well the EU's banks could cope with an adverse shock. Mario Quagliariello, the EBA’s director of economic analysis and statistics, explains in more detail how the stress tests work and why they are important.

What exactly is a stress test?

A stress test is a diagnostic tool for banks. Supervisors use it for assessing how sensitive banks’ portfolios are to a hypothetical – extreme but plausible – adverse scenario. If they find potential vulnerabilities, supervisors can demand that banks take countermeasures, for example improving risk management, disposing of some of the riskier assets or putting aside more capital.

How, in practical terms, do they work? What are the sort of criteria banks need to fulfil? Was there anything new this year?

Our stress test is a bottom-up exercise. This means that banks themselves run the exercise using internal models but following a series of assumptions and constraints included in the common methodology developed by the EBA. Basically, what banks need to assess is the impact of a common adverse scenario on their profitability and capital positions.

Since 2016, the EU-wide stress test is not a pass/fail exercise. Banks are not assessed based on their capacity to meet a specific capital threshold, but the outcome of the stress test feeds into the supervisory review and evaluation process.

If I have to pick just one key new aspect, I would definitely mention the introduction of International Financial Reporting Standards (IFRS) 9. IFRS 9 is the new accounting standard based on the recognition of expected credit losses. Exposures are classified as stage 1, stage 2 or stage 3 depending on their credit quality and move from one stage to another as credit quality deteriorates. This reasoning is reflected – albeit with some adjustments – in the stress test methodology. Another important addition is a more severe treatment for Level 2 and Level 3 assets, which reflects the fact that they are less liquid and their pricing is subject to model risks (i.e. the risk of a model being poorly specified, incorrectly implemented or used in a manner for which it is inappropriate).

Very briefly, what were the main results of the stress test this year? Are European banks stable?

The impact of the stress test is assessed as the decline of the CET1 (Common Equity Tier) capital ratio – the key metric in our stress test – in the adverse scenario compared to the starting point. The average fully loaded CET1 ratio was 14.2% in 2017, the starting point for this year’s exercise. The negative impact of IFRS 9 first implementation was 20 basis points (bps) for the 48 banks in our sample. At the end of the 2020 adverse scenario, the reduction amounts to 395 bps. The main drivers vary considerably from bank to bank, but on average, credit risk losses are the main contributor. For some banks, low profitability is also a key driver – and many banks that performed relatively poorly in the stress test are the ones with low profit generation.

The outcome of the stress test shows that banks' efforts to build up their capital base in recent years have helped improve their resilience and capacity to withstand severe shocks. However, although the aggregate results show resilience, when looking at banks’ results it is crucial to bear in mind their individual balance sheet structure, business models and their particular vulnerabilities. This is what the supervisors will do as part of their wider review and evaluation of banks’ capital positions.

What is the origin of stress tests? When did they begin?

Stress tests have been used in the banking industry since the early 1990s. At the global level, the 1996 amendment to the Basel 1 Capital Accord stipulated that banks using internal models for market risk must develop stress test methodologies. This has become a general requirement for all risks under the Basel 2 Accord. In Europe, under the EU Capital Requirements Directive, banks are expected to use stress testing to assess their ability to absorb losses and maintain sufficient capital levels, including in stress situations.

At the EBA, when we started in 2011 in the midst of the crisis, the main objective was to address market uncertainty on banks’ exposures and potential capital shortfalls. We also decided to be fully transparent with the bank-by-bank outcomes and to disclose detailed bank data.

In 2014, our stress test was instrumental to the setting-up of the Single Supervisory Mechanism. It also contributed to the comprehensive assessment the European Central Bank carried out before taking over supervisory responsibility for the banks in the euro area. This was our last exercise with an explicit objective of quantifying capital shortfalls and recapitalizing banks. In 2016, the focus shifted to the identification of potential vulnerabilities, rather than immediate capital problems. Without a pass/fail capital hurdle rate, the stress test became a component of the wider supervisory assessment.

What role do stress tests play? Why are they important?

Stress tests are an important element of the supervisory toolkit. They have a forward-looking dimension that allows banks and supervisors to think outside of the box. It allows them to assess the impact of risks that, while perhaps not current, could still materialise in the future. And it lets them formulate contingency plans for dealing with hypothetical, but not impossible, adverse market dynamics. However, we are not talking about a crystal ball! Stress tests cannot cover all possible risks and do not give banks a ‘clean bill of health’. This is why I don’t like it when banks claim to have ‘passed’ the stress test. This is not the point. The point is to be more aware of possible risks and therefore better prepared to deal with them.

Read more about EU-wide stress testing