IOPS Toolkit for Risk-based Supervision

1
IOPS Toolkit for Risk-based Supervision

• Module 4 Risk Mitigation and Scoring

2
RBS Process
3
(No Transcript)
4
Risk Mitigants

• Main mitigating factors to be considered
• Quality of governing board / trustees
• Management controls
• Compliance culture
• Effectiveness of operational management
• Adequacy of risk management systems
• Adequacy of independent review
• Role of administrator
• Sponsor
• Financial support

5
Risk Mitigants APRA
6
Risk Matrix OFSI Canada
7
Risk Weightings

• The following should be considered when
  establishing a methodology for weighting risk,
• Nature of the pension system
• The type of pension fund (i.e. DC, DB, Hybrid)
• Risk factors with measureable financial
  consequences
• Nature, scale and complexity of the supervised
  entity
• External environmental and market (systemic)
  factors
• Weightings changing over time
• Sensitivity tests or back testing to ensure
  accuracy and consistency

8
Risk Weightings APRA
9
Probability

• Conditional probability characteristics
  historically known to correlate with the
  occurrence of an event. Probability expressed as
  a function of the characteristics in a particular
  fund.
• Most RBS probability models are either additive
  or multiplicative, suggesting that risk
  characteristics are positively correlated.
• Some supervisors combine probability and impact
  of risk into a single score i.e. probability of
  the risk leading to a significant to high impact.
  Assumption of high interdependence

10
Probability APRA
11
Impact

• Impact measures assist in determining the
  supervisory oversight a fund will receive.
• Most authorities use size of the entity to
  capture the damage that would be inflicted if an
  adverse event occurred.
• Number of factors to determine size number of
  active or retired members, total assets, etc.
• Higher impact assigned to systemically important
  funds, although systemic importance requires
  definition.
• Impact thresholds determined by level of
  protection in the system (e.g. guarantee schemes,
  sponsor backup, etc.)

12
Impact Measures - Example
13
Consistency of Scores

• Risk scores need to be checked for accuracy and
  consistency.
• Central vs. Individual Judgement.
• Pre-populating scores useful in centre
  structuring the judgement of supervisors. Also
  captures external or systemic risks.
• Checking mechanisms central checking to ensure
  consistency across a large number of supervisors,
  internal comparisons and validations, training,
  etc.
• Separate unit for the design and maintenance of
  the risk assessment system.
• Retrospective testing of risk assessment models
  to validate risk scores given.

14
Consistency of Scores
15
Thank You

• Presentations of practical examples to follow