Actuarial assumptions for New Zealand superannuation scheme valuations
Post on 22-Feb-2016
36 Views
Preview:
DESCRIPTION
Transcript
Actuarial assumptions for New Zealand superannuation scheme valuations
Andrea Gluyas and Christine OrmrodNovember 2010
PricewaterhouseCoopers
Why this paper?
Pensioner mortality assumptions
Other valuation assumptions
Actuarial assumptions for New Zealand superannuation scheme valuations
PricewaterhouseCoopers
Why this paper?
PricewaterhouseCoopers
“If information to be used as audit evidence has been prepared using the work of a management’s expert [e.g. an actuary], the auditor shall, to the extent necessary, having regard to the significance of that expert’s work for the auditor’s purposes,:
Evaluate the competence, capabilities and objectivity of that expert;
Obtain an understanding of the work of that expert; and
Evaluate the appropriateness of that expert’s work as audit evidence for the relevant assertion.”
“An understanding of the work of the management’s expert includes … determination of whether the auditor has the expertise to evaluate the work of the management’s expert, or whether the auditor needs an auditor’s expert for this purpose.”
The role of the actuary assisting in audit
PricewaterhouseCoopers
“Aspects of the management’s expert’s field relevant to the auditor’s understanding may include:
Whether that expert’s field has areas of specialty within it that are relevant to the audit.
Whether any professional or other standards and regulatory or legal requirements apply.
What assumptions and methods are used by the management’s expert, and whether they are generally accepted within that expert’s field and appropriate for financial reporting purposes
The nature of internal and external data or information the auditor’s expert uses.”
“Considerations when evaluating the appropriateness of the management’s expert’s work as audit evidence for the relevant assertion may include:
The relevance and reasonableness of that expert’s findings or conclusions, their consistency with other audit evidence, and whether they have been appropriately reflected in the financial statements;
If that expert’s work involves use of significant assumptions and methods, the relevance and reasonableness of those assumptions and methods; and
If that expert’s work involves significant use of source data the relevance, completeness, and accuracy of that source data.”
The role of the actuary assisting in audit -cont
PricewaterhouseCoopers
Superannuation Schemes Act 1989 No guidance
NZSA Professional Standard No.2 Mostly relates to disclosure
NZ IAS 19 valuations Discount rate: risk –free Other assumptions: Entity’s best estimate
Superannuation scheme valuations in New Zealand
PricewaterhouseCoopers
Pensioner Mortality Assumptions
PricewaterhouseCoopers
NZLT 2005-2007 less 2 years NZLT 2005-2007 less 1 year and less 3
years NZLT 2000-2002 less 2 years PA(90) less 3 years and less 4 years NZLT 2005-2007 less 1 year, with mortality
improvements
Is the mortality assumption reasonable?
What are we seeing?
PricewaterhouseCoopers
Consistent evidence of mortality improvement
Mortality differentials
DBPA and non DBPA show statistically different mortality
Males and females are significantly different percentages of population mortality (DBPA: 80% and 100%)
No evidence of selection against the Schemes
Consistency in the “shape” of the curve over the years and different shapes for differ Schemes
Percentage rather than an age deduction gives a better fit to the experience
Findings from the NPF Schemes
PricewaterhouseCoopers
65-69 70-74 75-79 80-84 85-89 90-94 95-9980%
90%
100%
110%
120%
130%
140%
150%
160%
170%
180%
Actual mortality as a percentage of NZLT2005-07 - Non-DBPA males
1996-1999 1999-2002 2002-2005 2005-2008 NZLT2005-2007
Age
Non-DBPA male pensioner mortality experience
PricewaterhouseCoopers
65-69 70-74 75-79 80-84 85-89 90-94 95-9980%
90%
100%
110%
120%
130%
140%
150%
160%
170%
180%
Actual mortality as a percentage of NZLT2005-07 - DBPA males
1996-1999 1999-2002 2002-2005 2005-2008 NZLT2005-2007
Age
DBPA male pensioner mortality experience
PricewaterhouseCoopers
65-69 70-74 75-79 80-84 85-89 90-94 95-9960%
70%
80%
90%
100%
110%
120%
130%
140%
150%
160%
Actual mortality as a percentage of NZLT2005-07 - Non-DBPA females
1996-1999 1999-2002 2002-2005 2005-2008 NZLT2005-2007
Age
Non-DBPA female pensioner mortality experience
PricewaterhouseCoopers
65-69 70-74 75-79 80-84 85-89 90-94 95-9960%
70%
80%
90%
100%
110%
120%
130%
140%
150%
160%
Actual mortality as a percentage of NZLT2005-07 - DBPA Females
1996-1999 1999-2002 2002-2005 2005-2008 NZLT2005-2007
Age
DBPA female pensioner mortality experience
PricewaterhouseCoopers
Population mortality improvements
6061626364656667686970717273747576777879808182838485868788899091929394959697989995%
100%
105%
110%
115%
120%
Three years of mortality improvement from population statistics
Females Males
PricewaterhouseCoopers
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99100%
105%
110%
115%
120%
125%
Mortality improvement expected using mortality improvement formula
2002-2005 1999-2002 1996-1999
NPF mortality improvement formula
PricewaterhouseCoopers
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99100%
105%
110%
115%
120%
125%
130%
135%
140%
145%
150%
Total mortality improvement implied by an age adjustment
-1 age adjustment -2 age adjustment -3 age adjustment
Age adjustment mortality improvement allowance
PricewaterhouseCoopers
35 45 55 65 75 85-15.0%
-10.0%
-5.0%
0.0%
5.0%
10.0%
Value of annuity of $1,000pa - Male single lifePercentage difference from value using NZ2005-07 -2 years
Non-DBPA DBPA NZLT0507 mort impNZLT0507-2 NZLT0507
How much difference does it make?
PricewaterhouseCoopers
35 45 55 65 75 85-15.0%
-10.0%
-5.0%
0.0%
5.0%
10.0%
Value of annuity of $1,000pa - Female single lifePercentage difference from value using NZ2005-07 -2 years
Non-DBPA DBPA NZLT0507 mort impNZLT0507-2 NZLT0507
Age
How much difference does it make?
PricewaterhouseCoopers
Given:
We find quite different mortality rates between different schemes Irrefutable evidence of mortality improvement Pensioner mortality is generally one of the more significant
assumptions Pensioner liabilities are often increasing as a proportion of a
scheme’s total liabilities, and That there is a relatively straightforward actuarial formula for
mortality improvement.
Should we be telling our auditing colleagues that New Zealand life tables with a two year deduction is reasonable for both current pensioner mortality and future improvements in pensioner mortality?
What mortality assumption is reasonable?
PricewaterhouseCoopers
Sliding scale against population mortality
Effect of removing impaired lives
And may be?
Other potential approaches to mortality asssumptions
PricewaterhouseCoopers
Other valuation assumptions
PricewaterhouseCoopers
PS2 asks for:
“an explanation of how the values for these assumptions were derived. This explanation shall include at least:
If the investment earnings assumption is one of the most financially significant assumptions, an explanation of the relationship between the investment earnings assumption and the current investment strategy the scheme, any changes assumed in the future to the investments strategy and the allowances made for each of future investment expenses, administration expenses and taxation”
Discount rates – Funding valuations
PricewaterhouseCoopers
Methodology for Risk-free Discount Rates and CPI Assumptions for Accounting Valuation Purposes NZ IAS 19 valuations, issued by the Treasury
Long term risk free discount rates Considers all available data Adopted a stable approach to extrapolation Issued at 30 June, 31 October, 31 December
and 28 February Clear methodology
Discount rates – NZ IAS 19
PricewaterhouseCoopers
Other issues discussed in the paper:
Use of a term structure for discount rate, to automatically match the duration of liabilities
Risk premium Scarcity discount Adjustments to reflect the liquidity of liabilities The differences between Government Stock and bank
swap rates The approach for durations longer than the longest
traded government stock Common errors such as not annualising half yearly rates
Treasury paper
PricewaterhouseCoopers
Pension increases
Administration expenses
Retirement ages
Commutation option
Other assumptions
top related