Stress Testing - Guideline impact analysis statement (2009)

I. Background

Stress testing is an integral component of the risk identification and risk management processes of financial institutions including the Internal Capital Adequacy Assessment Process (ICAAP) for deposit-taking institutions and Dynamic Capital Adequacy Testing (DCAT) processes for insurers. Events in financial markets during the last two years have highlighted the importance of rigorous and actionable stress testing programmes. Experience has shown that stress testing did not always include events considered to be unlikely or extreme, or events for which no historical data exists.

II. Problem Identification

OSFI’s experience and that of other regulators indicates that many financial institutions, to varying degrees, need to improve their stress testing programmes with respect to timeliness, identifying the interaction of risks and severity of stresses chosen.

III. Objectives

OSFI would like to establish a standard against which stress testing programs can be evaluated and to ensure the principles apply equally to all regulated financial institutions. OSFI would also like to address the need for improvements in the DCAT and stress testing in insurance companies generally. At the same time, OSFI desires to have guidance that is consistent with and embodies the principles included in the Basel Committee’s May 2009 guidance, Principles for sound stress testing practices and supervision.

IV. Identification and Assessment of Options

Option 1 – Rely solely on existing international guidance

The advantage of this option is that OSFI would not have to devote resources to developing its own guidance on stress testing.

However, the most recent stress testing guidance, which was issued by the Basel Committee on Banking Supervision, has a banking focus that does not address insurance risks. Further, international guidance cannot always reflect the Canadian marketplace and regulatory environment.

Option 2 – Develop common guidance applicable to the deposit taking and insurance sectors

The advantage of this option is that it establishes common principles and standards for both deposit taking and insurance institutions, placing all institutions on an equal footing.

The challenge with this option is the need to develop principles and standards that appropriately reflect the particularities of the two sectors. However, the high level principles apply to both sectors, and sector specific material can be included in the discussion of the principles, for example discussions of DCAT–related issues.

Option 3 – Develop separate guidance for the insurance and banking sectors

This option would allow for guidance to be specifically tailored to each of the insurance and deposit taking sectors. However, this would result in a duplication of effort because the same underlying principles apply to the insurance and banking sectors. Further, this option diverges from OSFI’s desire to apply similar guidance, where appropriate, to the different financial sectors.

V. Recommendation

Given that broadly the same principles and standards for stress testing apply to both the deposit taking and insurance sectors, and OSFI’s desire to have common guidance for all sectors where appropriate, OSFI is of the view that the creation of common guidance is the most appropriate option for ensuring sufficiently robust guidance is available to financial institutions. Further, sole reliance on international guidance does not necessarily take into consideration factors that may be unique to the Canadian marketplace or regulatory environment.

VI. Consultations

OSFI issued the draft version of the guideline for comment on August 26, 2009. The main focus of comments received was on the timing of DCAT and when full compliance with the guideline would be expected.

The draft guideline set out an expectation that an insurer’s annual DCAT would be available to the board of directors no later than six months after year end. Commenters pointed out that requiring DCAT to be completed within six months of year end does not mesh well with insurers’ planning cycles, and that uncoupling DCAT from the planning cycle would reduce its effectiveness as a management tool. As a result, we removed this requirement and replaced it with a requirement that DCAT be integrated with the business planning process.

The GIAS accompanying the draft guideline stated that it was OSFI’s intent to implement the guideline by year end 2009. Many commenters interpreted this to mean that OSFI would expect full compliance with the guideline by year end 2009. The cover letter accompanying the final guideline clarifies that we expect institutions to assess their stress testing programs against the principles set out in the guideline and develop a plan to achieve compliance within a reasonable timeframe.

A number of minor changes were also made to the guideline in response to comments requesting clarification.

VII. Implementation & Evaluation

OSFI will continue to assess institutions’ stress testing programmes as part of the supervisory review process, including its reviews of the ICAAP and DCAT. These assessments will take into consideration whether the institutions are incorporating the principles outlined in Guideline E-18 into their risk identification and risk management processes.