Stress testing

What is stress testing?

We need to make sure banks and insurance companies are strong enough to withstand another financial crisis. So we set them ‘stress tests’ to find out if they are prepared for the worst.

Stress testing: banks and building societies

Banking stress tests assess how banks can cope with severe economic scenarios. We look at banks’ resilience, making sure they have enough capital to withstand extreme shocks and are able to support the economy.

Stress testing of banks: an introduction

Types of banking stress test

There are three types of banking stress test:

1. We run an annual concurrent stress test of the largest UK banks and building societies. This informs the setting of capital buffers by our Financial Policy Committee and our Prudential Regulation Authority (PRA). 
2. Firms that are not part of this annual stress test must carry out their own stress testing. The PRA publishes a scenario every six months to serve as a guide for banks and building societies designing their own scenarios.
3. Every other year, we run an additional scenario intended to probe the resilience of the banking system to risks that may not be neatly linked to the financial cycle – the biennial exploratory scenario.
   
    

Annual concurrent stress test

We published the results of the 2019 stress test of the UK banking system on 16 December 2019.

• Stress testing the UK banking system: 2019 results (PDF) Opens in a new window
• Bank-specific results (PDF) Opens in a new window
• Effectiveness of stress-testing frameworks (PDF) Opens in a new window
• Key elements of the 2019 stress test (PDF) Opens in a new window
• Variable paths for the 2019 stress test (XLSX)
• Traded risk scenario for the 2019 stress test (XLSX)
• 2019 stress guidance for banks and building societies (PDF) Opens in a new window

We cancelled the 2020 annual stress test to help lenders focus on meeting the needs of UK households and businesses via the continuing provision of credit.

Instead, in May 2020 we published the results of a desktop stress test Opens in a new window of major UK banks and building societies using the economic scenario outlined in the May 2020 Monetary Policy Report.

In August 2020 we carried out a ‘reverse stress test Opens in a new window’. This analysed how much worse than the central projection the economic outcome would need to be in order to deplete regulatory capital buffers by as much as in the 2019 stress test.

Data templates for the 2019 Concurrent Stress Testing exercise 

We publish Concurrent Stress Testing exercise data requirements to participating firms for submission in Excel and in XBRL.

These Excel templates will be used for data submissions by firms subject to the 2019 Concurrent Stress Testing exercise. They have been provided directly to participating firms.

XBRL taxonomy for the 2019 Concurrent Stress Testing exercise

The link below contains the final version of the XBRL taxonomy, annotated templates and dictionary for the 2019 Concurrent Stress Testing exercise.

• 2019 XBRL taxonomy (17MB) (ZIP)
• 2019 Data dictionary (8MB) (ZIP)

Data templates and dictionary for the 2020 Concurrent Stress Testing exercise

We publish Concurrent Stress Testing exercise data requirements to participating firms for submission in Excel and in XBRL.

These Excel templates and dictionary are for data submissions by firms subject to the 2020 Concurrent Stress Testing exercise.

• 2020 Data templates (3MB) (zip)
• 2020 STDF dictionary (7MB) (zip)

A summary of the Bank of England 2020 Concurrent Stress Testing data request, an overview of the main changes since the previous version, and the operating model for the reporting of stress test data by participating firms can be found below:

• Bank of England Stress Test - index of artefacts and associated submission timeframes 202002 (xlsx)
• STDF change summary 202002 (xlsx)
• Operating model for the reporting of stress test data 202001 (docx)

The asset liability management template below is not submitted as part of the concurrent stress test, but feeds into our stress test analysis.

• 2020 asset liability management template (xlsx)

XBRL taxonomy for the 2020 Concurrent Stress Testing exercise

The link below contains the final version of the XBRL taxonomy, annotated templates and dictionary for the 2020 Concurrent Stress Testing exercise.

• 2020 XBRL taxonomy updated February 2020 (15MB) (ZIP)
• 2020 Data dictionary (4MB) (ZIP)

How to use our stress scenario

You should consider the relevance of our stress test scenarios in the context of your business and its own risks. 

You should use our scenario as a starting point to design adequately severe scenarios for your firm under Pillar 2. We know any single scenario which is designed for firms with different business models and risks, has its limits. We expect you to choose a scenario that provides a strong challenge for your business. 

You are responsible for developing your own scenarios to test your firm's resilience. Large banks and building societies should use the annual cyclical scenario.  

Firms who are subject to IFRS 9 should consider the following clarifications covering the expected approach to IFRS 9 within ICAAPs.

Historical guidance

• Clarification on IFRS 9 for 2017 ICAAP stress testing and capital planning
• Clarification on IFRS 9 for 2018 ICAAP stress testing and capital planning

Scenarios for banks and building societies not part of concurrent stress testing

We have taken a range of supervisory and policy measures to address the challenges of Covid-19, including the cancellation of the annual stress test (ACS) for systemic firms and increased emphasis on buffer usability.

Following the cancellation of the ACS, the PRA is not publishing the related stress scenarios for non-systemic firms.

In line with Supervisory Statement 31/15 ‘The Internal Capital Adequacy Assessment Process (ICAAP) and the Supervisory Review and Evaluation Process (SREP)’, ICAAPs should be updated and stress testing undertaken at least annually.

The PRA expects firms’ to consider and develop appropriate internal stress scenarios for their own use of stress testing, and more widely as a risk management tool.

Firms may wish to consider the May 2020 scenario published in the Monetary Policy Report on 7 May 2020, particularly in terms of its shape, but should consider more severe variations of this or alternative scenarios.

Firms should also continue to refer to the PRA’s most recently published 2019 annual stress scenario for non-systemic firms as a benchmark for severity. Given the current economic outlook, firms may consider adjusting the change in the economic variables of the 2019 scenario, in line with our October 2015 stated approach to stress testing Opens in a new window.

This guidance is aimed at helping firms develop scenarios for risk management purposes, rather than solely for internal buffer assessments.

Given the current economic climate, stress testing remains an important supervisory tool to inform our understanding of the resilience of the sector.

Therefore our capital SREP assessments will have an additional emphasis on firms’ use of stress testing to understand their business model vulnerabilities and suite of mitigating actions. This will include an increased focus on reverse stress testing.

Firms should carefully consider the economic and idiosyncratic variables, and levels of stress severity that would lead to a breach of Total Capital Requirements. This should include consideration of a combination of variables within an overall coherent scenario.

We published our most recent scenarios on 28 March 2019. They are here:

ICAAP scenario

The results of stress tests are an important consideration when we decide how to set capital requirements for banks and building societies. 

Read our guidance on the role of stress testing in our Supervisory Statement on the internal capital adequacy assessment process and the supervisory review and evaluation process.

Biennial exploratory scenario

Since 2017, we have also run a second concurrent stress test. Conducted every other year, this is known as the ‘biennial exploratory scenario’ (BES). Its focus changes from exercise to exercise, and is designed to explore risks facing banks not covered by the annual test. Previous tests have covered risks from persistently low interest rates and liquidity risks.

We had been due to publish the results of the 2019 BES on liquidity in mid-2020. However, to alleviate demands on core treasury staff at participating banks during the Covid-19 epidemic, we have paused this exercise until further notice. 

The 2021 BES is due to cover risks to the UK financial sector from change. ln December 2019 we published a discussion paper outline its intentions for the exercise. In May 2020 the Financial Policy Committee, together with the Prudential Regulation Committee, agreed to postpone the launch of the 2021 Climate BES until mid-2021 at the earliest. The exercise will cover both major UK banks and insurers.

Stress testing: insurers

Insurers should develop, implement and action an effective stress testing programme. Stress testing should assess their ability to meet capital and liquidity requirements in stressed conditions, as a key component of effective risk management. All firms should undertake relevant analysis, equal to the nature, scale and complexity of their business.

The PRA also runs its own stress tests on a periodic basis for a number of insurance firms. It does this regularly for specific high-impact firms and others as the need arises. This assesses their ability to meet minimum stated capital levels throughout a stress period. 

System-wide stress testing will be used by firms using a common scenario for financial stability purposes. To support its framework, the PRA sets policy for firms' stress testing requirements and stress scenarios, monitoring test results.

On Wednesday 17 June 2020 the PRA published a letter to participating general and life insurers on feedback for general and life insurers from the 2019 Insurance Stress Test.

Reverse stress tests

The PRA also expects insurance firms to apply reverse stress testing as part of their own risk and solvency assessment (ORSA) process, to continuously assess their overall solvency needs for their insurance specific risk profile.

Reverse stress tests are stress tests that require a firm to assess scenarios and circumstances that would make its business model unworkable, identifying potential business vulnerabilities. 

This differs from typical stress and scenario testing, which tests for outcomes arising from changes in circumstances. A firm's business model is described as being unrealistic at the point when forming risks cause the market to lose confidence in the firm.

Reverse stress testing is designed to be a risk management tool, encouraging firms to explore the vulnerabilities and faults in its business model, including tail risks. Tail-risk is the risk, or probability, of rare insured events occurring in the future. 

The PRA works with others in the EU and internationally on approaches to stress testing.

Stress testing updates