Analysis of movement Realistic Balance Sheet

[bensondros]

In the chapter on analysis of surplus, almost everything is compared between actual and expected experience to understand the change in surplus / working capital from the start of period to the end of period.

However, for the economic variance in the analysis of movement of Peak 2 realistic balance sheet for WP funds, the economic variance is calculated by rolling the model forward to the new valuation date using actual investment returns for the inter-valuation period and economic scenarios calibrated at the new valuation date thereafter.

Why is it not comparing actual vs expected?
Shouldn't it be a 2-step process of first rolling the model forward on expected investment returns and calculating the value of assets and liabilities based on expected future economic scenarios as determined at the old valuation date, and then changing the expected assumptions into actual values?

 

[Mark Willder]

In the chapter on analysis of surplus, almost everything is compared between actual and expected experience to understand the change in surplus / working capital from the start of period to the end of period.

However, for the economic variance in the analysis of movement of Peak 2 realistic balance sheet for WP funds, the economic variance is calculated by rolling the model forward to the new valuation date using actual investment returns for the inter-valuation period and economic scenarios calibrated at the new valuation date thereafter.

Why is it not comparing actual vs expected?
Shouldn't it be a 2-step process of first rolling the model forward on expected investment returns and calculating the value of assets and liabilities based on expected future economic scenarios as determined at the old valuation date, and then changing the expected assumptions into actual values?

Good question.

To answer this, we need to think about how the reserves are calculated.

Most of this chapter looks at a traditional analysis of surplus where reserves are calculated using Pillar 1 Peak 1. These reserves are calculated prospectively using assumptions of future experience. We then compare the actual experience over the year to these expectations to analyse the surplus.

However, when we analyse the realistic balance sheet, we are looking at Pillar 1 Peak 2 reserves. These are broadly the asset share plus the cost of derivatives to ensure guarantees are met. We don't need an assumption for future investment return here - asset shares are based on past returns and the cost of guarantees is calibrated to observable market prices. So we don't have an expected return that we can use to roll everything forward.

I hope this makes sense. Good luck for the exam.

Mark

 

[bensondros]

Would never have thought of it that way. Thanks Mark!

 

[ALEX_AK]

Retrospective vs prospective reserve

Hi as mentioned here, the pillar 1 peak 2 reserves are taken as the asset share, ie calculated on retrospective approach. It seems to me that pillar 1 peak 2 reserves are always calculated on retrospective approach. I have seen this in several other cases as well where the answer to ASET only consider the pillar 1 peak 2 reserves as AS. May I ask if this should always be our approach when answering questions? Ie we will only always consider and talk about retrospective reserve (AS) when talking about pillar 1 peak 2 reserve?

 

[Mark Willder]

Hi as mentioned here, the pillar 1 peak 2 reserves are taken as the asset share, ie calculated on retrospective approach. It seems to me that pillar 1 peak 2 reserves are always calculated on retrospective approach. I have seen this in several other cases as well where the answer to ASET only consider the pillar 1 peak 2 reserves as AS. May I ask if this should always be our approach when answering questions? Ie we will only always consider and talk about retrospective reserve (AS) when talking about pillar 1 peak 2 reserve?

Using asset shares to calculate the Peak 2 reserve is nearly always the natural way to value with-profits business because asset share is nearly always what we're aiming to payout.

The Core Reading in Chapter 13 page 10 says:

"A prospective method may be used for policy classes where bonus rates are not directly determined by asset share methods (eg whole life policies) or where asset shares are determined for specimen policies only and are not calculated in aggregate."

So, if the question is about an insurer that does not sell whole life, then the majority (if not all) of the marks should be for the asset share approach. If the insurer sells a mix of business (that could include whole life) then its worth giving the prospective approach a mention.

Best wishes

Mark