April 2023 Ques 1 part (i)

[Srishti Goyal]

Hi,

I have a confusion regarding my interpretation for Ques 1 of April 2023.
"QUES: The company has carried out an analysis of its supervisory valuation surplus. Statutory liabilities are assessed on a best estimate basis and required capital is assessed on a risk-based approach that is the same as Solvency II. Initial output from this analysis identified a significant positive non-economic variance.
(i) Discuss possible reasons for this result."

My Interpretation:

Ques is asking about the Analysis of Surplus and has mentioned Positive Non-economic Variance.

By Variance it means: Actual less the Expected.

Non-Economic Variance means: Mortality, Lapses, Expenses and mix of business(optional).

Positive means: Actuals are higher than expected

Based on this interpretation, I will draft the answer focusing on higher mortality, higher lapses and higher expenses.
However, in the exam report, they have mentioned about lower lapses and lower expenses.

Can anyone please help me understand what the correct approach is to interpret the answer here.

Thanks in Advance.

 

[Lindsay Smitherman]

A 'positive variance' within an analysis of surplus is one that generates a positive contribution to the surplus arising (or 'profit') over the period.

Because expenses are a deduction from profit (or surplus arising), a 'positive variance' relating to expenses must mean lower than expected expenses.

For mortality, higher than expected mortality would generate a positive variance for annuity business, but lower than expected mortality would generate a positive variance for term assurances (say) - since claims are also a deduction from profit.

For lapses, it is slightly less clear whether higher or lower than expected lapses would give a positive variance (although probably lower) - so alternative outcomes are covered in the solution to the question you mention.

 

[Srishti Goyal]

Thank you for the clarification. It helped me understand the approach that IFOA has opted in exam.

However, I still want to understand how we should approach a question. Should it be broken on word-by-word basis or overall meaning. Here, if we split on word-by-word basis the interpretation is different. In course material also, it is mentioned that the Variance is the difference between Actuals and expected. Basis that, Positive Variance can be concluded as higher actuals than expected.

If we mention our interpretation in the exam clearly and comment based on that, will we be marked accordingly?

 

[Lindsay Smitherman]

We aren't involved with the examining so can't say whether or not that would have gained any credit. If something is ambiguous then generally marks would be given for different interpretations. However, I am not convinced that the examiners would have considered this to be ambiguous. The phrase 'significant positive non-economic variance' means that non-economic experience has made a positive contribution to surplus arising, ie the company is now in a better position. [You need to look at the overall meaning of a phrase.]

Also bear in mind that this 'non-economic variance' will be a single item in the analysis of surplus, comprising the collective impact on the surplus arising (or 'profit') of all non-economic items of experience differing from what was expected.

In other words, for this to be a positive variance, we must have:
{premiums - expenses - claims + investment earnings - increase in liabilities} based on actual non-economic experience
>
{premiums - expenses - claims + investment earnings - increase in liabilities} based on expected non-economic experience

It doesn't make sense to break this down into saying that therefore we must have actual expenses > expected expenses, because in fact the formula shows that the above inequality is more likely to hold if actual expenses < expected expenses.

 

[Srishti Goyal]

Thank you so much for the detailed explanation.

 

[1495_sc]

Hi Lindsay,

I want to understand why the solution doesn't consider any impact on assets. Only the impact on BEL and SCR is considered. Why so? In real world, if mortality is lower than expected for annuity, assets will be higher than expected at the end of the year due to fewer annuity payments made during the year.

For expenses, the impact is only on assets although it is not mentioned explicitly. No impact on BEL as BEL considers only expected expenses. So, if we mention that profit will be higher due to lower expenses charged from assets, we should be good?

Please help!

 

[Em Francis]

Hi

The solution includes both impact on assets and liabilities, for example:

There may have been more actual deaths than expected ...

… for example, due to a pandemic …

… and this would lead to a positive non-economic experience variance for the immediate annuity portfolio …

and
If the death benefit were a return of fund, as would typically be the case

… then fewer actual deaths than expected could lead to a positive variance, due to retaining value from future profits

---

---

 

[Em Francis]

Hi Lindsay,

I want to understand why the solution doesn't consider any impact on assets. Only the impact on BEL and SCR is considered. Why so? In real world, if mortality is lower than expected for annuity, assets will be higher than expected at the end of the year due to fewer annuity payments made during the year.

For expenses, the impact is only on assets although it is not mentioned explicitly. No impact on BEL as BEL considers only expected expenses. So, if we mention that profit will be higher due to lower expenses charged from assets, we should be good?

Yes, assets at the end of the period will be higher than expected due to paying out less in expenses.

Please help!

 

[1495_sc]

Hi

The solution includes both impact on assets and liabilities, for example:

There may have been more actual deaths than expected ...

… for example, due to a pandemic …

… and this would lead to a positive non-economic experience variance for the immediate annuity portfolio …

and
If the death benefit were a return of fund, as would typically be the case

… then fewer actual deaths than expected could lead to a positive variance, due to retaining value from future profits

---

---

For mortality experience variance in annuity- the examiner's report mentions that

"There have been more deaths than expected.
For example due to a bad winter/hot summer/data cleanse/poor underwriting
(any sensible example) [½]
This would lead to a positive non-economic experience variance for the annuity book [1]
as reserves [½]
and regulatory capital held for annuitants who have died are released to surplus."

Here, there is no mention of assets. Why so? If this type of question comes up again, what is the approach to answer?
Similarly, for lapse experience variance, the solution mentions that -
"Lower levels of lapses in the pension book will increase in force at the end of the
year
The reserves are likely to be the value of expenses less charges [½]
which are likely to be negative for a profitable product [½]
thus reserves will be lower at the end of the year. [½]
The required capital for the pension book is likely to be higher as the volume of
business is higher [½]
but this is not likely to be sufficient to offset the reduction in reserves… [½]
leading to a positive variance."

Again, no mention about impact on assets. Am I missing something here? I completely agree with the impact on BEL and SCR. However, not sure why we should not comment about assets.

 

[Lindsay Smitherman]

You are effectively saying: 'If more annuitants die than expected, the assets at the year end may be higher than expected because any annuity payments due between when they died and the year end will not have been made, and the reserves and capital requirements held at the year end will be lower than expected because there is no need to hold them for those annuitants anymore - so overall there is a positive experience variance.'

The Examiners' Report is effectively saying 'If more annuitants die than expected, the reserves and capital requirements for those annuitants can immediately be released into surplus (ie at the point of death) - so there is a positive experience variance.'

They are saying the same thing, it's just a different way of expressing the same basic idea. By thinking about the immediate impact (and hence automatically including both any remaining benefit payments that would be due over the rest of the year and what would have been set up at the year end to cover benefits beyond that), the ER has explained it more concisely.

For the persistency variance, bear in mind that this is unit-linked pension business. The solution focuses on the impact on future profits (and their capitalisation within the non-unit reserve) as that is what is going to drive the experience variance impact. It is likely that the unit fund value will be paid out on surrender, so surrender payments out will offset against the change in unit reserve. Hence that element (change in assets and equivalent change in unit reserve) is neutral and so isn't going to explain the experience variance.

 

[1495_sc]

Thank you!

 

[CAKABOGU23]

For the persistency variance, bear in mind that this is unit-linked pension business. The solution focuses on the impact on future profits (and their capitalisation within the non-unit reserve) as that is what is going to drive the experience variance impact. It is likely that the unit fund value will be paid out on surrender, so surrender payments out will offset against the change in unit reserve. Hence that element (change in assets and equivalent change in unit reserve) is neutral and so isn't going to explain the experience vavariance.

The ER mentions reserves are likely to be negative for a profitable product...I thought supervisory reserves had a lower bound of zero?

 

[Em Francis]

The ER mentions reserves are likely to be negative for a profitable product...I thought supervisory reserves had a lower bound of zero?

This will depend on jurisdiction. Some jurisdictions allow negative reserves, under certain conditions. This is covered in SP2.