CHAP 17

Assessment of reserving results

• Reserving result analyses are crucial for assessing the reasonableness of reserving exercise results
• Factors to consider when assessing the reasonableness of results include the methodology and assumptions used in the reserving exercise
• Diagnostics are used to assess the reasonableness of reserving exercise results
• Typical diagnostics commonly used to assess the reasonableness of results include analysis of experience over time.
• Analysis of experience in the context of a reserving exercise factors are critical.
• Alternative results, and professional issues in resolving them, should be addressed in reserving exercises.
• Analyzing the results of a reserving exercise is vital to determine if the figures are justifiable.
  • The methodology and assumptions used must be appropriate.

Introduction

• Assessing the reasonableness of the reserving calculation outcome is crucial for actuaries
• Actuaries must examine the reserving calculation methodology and assumptions to ensure they are justifiable.
• Diagnostics and experience analysis can aid in this process
• The results of reserving exercises should be analyzed to determine their reasonableness.

Assessing the results of a reserving exercise

• Assessing the reasonableness of reserving exercise results is a critical stage in the reserving process.
• Two key questions to ask in this assessment are:
  • Are the results reasonable?
  • Are the results supported by analysis of emerging experience?
• Actuaries often conduct an "Actual versus Expected" exercise to compare current experience with expected experience since the last review.
  • This helps assess the assumptions appropriateness.
• Anchoring error (over-reliance on previous methods and assumptions) must be avoided.
• Changes in methodology or assumptions, their reasons, and impact on results are critical.

Approaches to analysing the results

• Applying diagnostic tests to check reasonableness is essential.
• Analysis of emerging experience is another vital approach.
• Balancing thorough analysis with time-efficiency is crucial
• Assessing the entirety of the exercise for reasonableness is important
• An actuary’s experience provides an implied benchmark for expected results.
• Factors like the expected frequency of specific events, or range for a particular diagnostic, should be considered.
• Actuary experience may be limited and not typical of the entire population, careful consideration is needed accordingly.
• Peer review from more experienced actuaries can be helpful to examine methodology and assumptions.

Diagnostics

• Diagnostics are used to test and verify underlying methodologies and assumptions.
• Diagnostics often check results or data to determine if experience is consistent with assumptions.
• Diagnostics provide high-level reasonableness checks or pinpoint areas for further examination.
• These diagnostics include IBNR divided by premium, and ultimate loss ratios to aid in analysis of results.
• Analyzing unusual diagnostic features requires investigation into underlying reasons and appropriate action.

Loss ratios

• Review the potential changes in loss ratios (paid, outstanding, IBNR, incurred, and ultimate).
• Changes in premium rating strength should be noted
• Examine sources of uncertainty, inconsistencies in modeling assumptions, or errors in reserving process.
• Compare loss ratios to premiums and claims indices or benchmarks (e.g., paid and incurred ratios or external benchmarks) for reasonableness. Early detection of trends is important.

Paid to incurred/case estimates to incurred ratios

• This diagnostic evaluates case estimate strength.
• An increasing trend (paid/incurred or case/incurred) might indicate weakened case estimates.
• The converse (increasing case/incurred) could suggest improved case estimates
• Changes in business (underlying trends, settlement patterns) and large loss settlements should be considered when interpreting these ratios.

Ratio of IBNR to case estimates

• This is helpful for more mature claims cohorts where IBNR is primarily about IBNER.
• It provides insights into outstanding claims and uncertainty.

Survival ratios

• Survival ratios indicate how long an IBNR or reserve estimate is likely to last.
• Analysis should be conducted using multiple averaging periods (e.g., 1-year, 3-year) of paid or incurred claims data to avoid periods of exceptionally high or low claims activity.
• Comparing survival ratios to similar portfolios or market benchmarks can be insightful.

Claim frequency and average cost per claim diagnostics

• This is used when claim count information is available.
• Trends in frequency and severity can be highlighted using these diagnostics.
• Unusual features (high average claims cost or unusual claims count changes) should be investigated
• Comparing average claim size to a maximum claim value can be helpful.

Development pattern diagnostics

• Comparing the assumed future development with past development patterns is important.
• Splitting data by origin cohort and development period can reveal valuable insights.
• Cumulative development is useful.
• Using triangles can highlight issues with development factors and trends over time.
• Unusually high or low figures in rows or columns of figures should be investigated.
• Unusually low figures on a diagonal can signal underlying issues with claims settlement.
• Changes over time in development trends should be identified and understood.

Stability of development pattern

• Assess if development patterns are reliable.
• Evaluate if appropriate alternative approaches are necessary for unpredictable patterns.
• Understand the potential for adverse development exceeding favourable development.
• Investigate whether sufficient tail potential beyond historical development is incorporated.

Claim development versus premium development

• Comparing claim development with premium development helps assess potentially inconsistent trends or errors.
• Earned premiums should typically develop faster than incurred claims.
• Comparing to benchmarks (e.g., industry, similar portfolios) can offer valuable context.
• Assess the quality and relevance of any residual data for patterns or anomalies.

Analysis of emerging experience

• Monitoring emerging experience is crucial for effective actuarial control.
• Evaluate how well past decisions handled changing environments and risks.
• Reasons for revisiting previous decisions (e.g. new information, changed circumstances) should be identified.
• Report any deviations from expectations to key stakeholders.
• Decisions regarding estimates and assumptions, and the reasons behind why differences exist, should be communicated.

The underwriting cycle

• Understanding the underwriting cycle (hard/soft market cycles) can be useful.
• Identifying trends and potential issues in claims development is important.
• Allowing explicitly for the underwriting cycle in reserving methods is vital.
• Assessing whether the model sufficiently accounts for the underwriting cycle effects.

Comparison to other estimates

• Comparing reserving estimates from different sources (actuaries or internal stakeholders) is essential to identify potential issues.
• Understanding rationale behind alternative estimates helps in identifying and resolving discrepancies.
• Comparing to benchmarks or market data can aid in identifying factors that may lead to differing results.
• Different data sources or methodologies may lead to differing reserves and comparisons.