The Bank of England has published details of its 2015 stress testing exercise for the largest UK banks and building societies, its second concurrent exercise (the first was in 2014).

This year's exercise is focused on assessing UK banks' vulnerability to a synchronised global downturn affecting Asia and the Euro area, coupled with amplifying global disinflationary pressures, and the resultant impact on the UK. The scenario also incorporates the assessment of the impact of a reduction in global risk appetite, particularly in indebted economies, as well as a number of counterparty defaults.

Given the focus of the scenario on stress in emerging economies and the new methodology created for traded risk, the 2015 exercise is likely to better test the vulnerabilities of global, non-retail focussed banks. The exercise appears to be more onerous on firms overall, whilst assessing different vulnerabilities.

Key differences compared to the 2014 exercise

  • The scope has been reduced to seven firms - HSBC, Barclays, Lloyds Banking Group, RBS, Nationwide, Standard Chartered and Santander. The Co-operative Bank will not be part of the exercise this year.
  • Firms are required to forecast projections over five years rather than the three-year time horizon used in last year's exercise.
  • A new market risk methodology has been implemented, which better aligns shocks applied to firms' trading book positions with their corresponding liquidity horizon. Firms have to apply instantaneous shocks to their positions, the size of which depends on the liquidity of the position. For anything greater than two weeks, or for positions that the bank would have to hold for franchise reasons, additional stresses need to be applied.
  • A leverage ratio hurdle rate of 3% has been introduced (of which 25% can be made up of Additional Tier 1 capital). The CET1 hurdle rate remains at 4.5% of RWAs, as last year.
  • Firms have been provided with more specific details on foreign exchange rate movements under stress. Currencies depreciate against dollar due to reduction in global risk appetite (Euro 25%, Brazilian Real 40%, Chinese Renminbi 10%, South African Rand 35%, Emerging Market average 25%). The scenario looks to assess the impact of FX movements on $-denominated bonds issued by other countries.
  • Chinese residential property falls 35% below end-2014 levels with larger falls for commercial property. Real GDP growth drops to 1.7% at its trough in Q4 2015.
  • Aggregate euro GDP growth troughs at -2.1% in 2016 Q1 whilst inflation doesn't turn positive until Q4 2018.
  • Oil price troughs at $38 a barrel, continually putting downward pressure on global inflation.
  • The Bank of England provides aggregate lending profiles in the stress. Overall, a firm's market share of the stock of lending in each year of the stress scenario should be at least as large as the corresponding share in the baseline scenario. It cannot contract its lending under stress at a quicker rate than it contracts under its corporate plan, unless this is represented as a management action.

Key challenges for firms

  • The Bank of England has outlined an approach for counterparty credit risk which is quite onerous and leaves an element of interpretation for firms.Firms have to estimate losses from defaults of specific 'vulnerable' counterparties. For uncollateralised exposures, firms have to identify their two most 'vulnerable' from their ten largest Asian exposures, and their most 'vulnerable' from their ten largest European exposures, under stress. For collateralised, they also have to identify their two most 'vulnerable' counterparties out of their 20 largest global exposures.
  • Firms should include the effects of regulatory, legal or accounting changes in their projections where final requirements and implementation or effective dates have been announced or publically endorsed on or before 30 March 2015. Firms are likely to face challenges in undertaking and modelling this aspect of exercise. This is likely to bring IFRS 9 and ring-fencing within scope.
  • A more defined approach for conduct risk (now labelled 'misconduct' risk) is outlined. Given legal issues surrounding the quantification of potential conduct losses, and the link with provisions, firms are likely to have difficulty agreeing their approach to calculating the 'prudential estimate' for such losses.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.