Solvency II Questions

[Joe Warren]

Hi,

I have a string of Solvency II questions that I would appreciate if I could get some help answering:

1) Ancillary Own Funds: Chapter 4, page 23 states 'The intention is that the MCR is calibrated to the Value-at-Risk of the basic own funds' and page 24 states 'The SCR is calibrated to the Value-at-Risk of the basic own funds'. Do ancillary own funds contribute to the assets in the SII balance sheet, or is it a value that must be attained via basic own funds only?

2) Tiering basic own funds (Chapter 4, page 17):

a) Free from incentives to redeem: Are the incentives to redeem criteria for tier 2 and tier 3 relevant to the insurer, or to other parties? For example, for tier 2 does another party have the option to redeem repayments from the undertaking but only from a minimum of 10 years from issuance?

b) Sufficient duration: Does the tier 3 requirement mean that the undertaking cannot issue debt with maturity less than 5 years to be an admissible asset under SII?

3) MCR Calculation: Chapter 4, page 23 states that the MCR is calculated by taking the sum of 'a factor applied to technical provisions (not including the risk margin) for each line of business, net of reinsurance, subject to a minimum of zero, and a factor applied to written premiums in each line of business over the last 12 month period, net of reinsurance, subject to a minimum of zero'. Why is a factor applied to the written premiums as well as the technical provisions, wouldn't the technical provisions already contain allowance for unexpired risk (i.e., premium risk)? Is the factor that is applied to the written premiums to represent risks not yet incepted over the one-year time horizon?

4) SCR - Intangible Risk: What comprises the 'Intangible' risk? There is no notes on it within the chapter, how does the standard formula calculate a capital charge for this risk, and how is it accounted for if an internal model is used?

5) Risk Margin - Cost of Capital - Definition: The risk margin is an allowance for the cost of capital, but what is the 'cost of capital' in this context? Is this to ensure that the insurer holds enough to cover the cost of holding capital (i.e., ensuring say a shareholder makes their required return on their capital), or is it to ensure that the possible erosion of capital is accounted for and is covered for as part of the technical provisions, and as a result the actual free reserve holding isn't eroded as a result of experience being worse than the best estimate?

6) Determining the Risk Margin under SII: Chapter 4, page 20 details how the risk margin is calculated by forecasting SCRs for a project, multipying each by 6%, discounting to the present and summating. If SII is determining provisions over a one-year time horizon, why do we need to determine the cost of capital over an entire project's lifetime?

7) Premium Risk Factors: Chapter 4, page 27 states that the 99.5% VaR factors for premium risk are applied to NEP, but on the following line say premium risk factors are based upon claims data that is gross of reinsurance? Is the premium risk factor different to the VaR factor applied to the premium data? And if so, why are premium risk factors applied to gross data as I thought the capital charge for Underwriting risk was based upon net data?

8) Counterparty risk - Type 2 Exposures: Chapter 4, page 29 details that the capital requirements for Type II Exposures are 'based on an immediate shock, assuming losses of 90% of receivables which have been due for more than three months and 15% on other receivables'. How does a firm forecast what receivables won't be due for more than 3 months, and which won't be, over a one-year time horizon?

Many Thanks,
Joe

 

[Ppan13]

Hi,

I have a string of Solvency II questions that I would appreciate if I could get some help answering:

1) Ancillary Own Funds: Chapter 4, page 23 states 'The intention is that the MCR is calibrated to the Value-at-Risk of the basic own funds' and page 24 states 'The SCR is calibrated to the Value-at-Risk of the basic own funds'. Do ancillary own funds contribute to the assets in the SII balance sheet, or is it a value that must be attained via basic own funds only?

I'd say the answer to the first part of this question is 'no', i.e. in the SII balance sheet, the value of assets after netting off technical provisions is only BASIC own funds.

However, the ancillary own funds may be called upon as additional sources of capital in certain circumstances, making them still eligible for demonstrating solvency (subject to complicated classification into tiers and PRA approval).

Source: " Ancillary own funds are contingent in that they have not been paid in, and are not recognised on the balance sheet. The need for supervisory approval of such items recognises this contingent nature." paragraph 1.4 https://www.bankofengland.co.uk/-/media/boe/files/paper/2020/december/gl-ancillary-own-funds.pdf

 

[Ppan13]

Hi,
...

2) Tiering basic own funds (Chapter 4, page 17):

a) Free from incentives to redeem: Are the incentives to redeem criteria for tier 2 and tier 3 relevant to the insurer, or to other parties? For example, for tier 2 does another party have the option to redeem repayments from the undertaking but only from a minimum of 10 years from issuance?
...
Joe

Tiers 2 & 3 may include limited incentives to redeem at the option of the [re]insurer only, which would increase the likelihood of a redemption by an insurer - e.g. step-ups linked to a call option.

All conversions , repayments or exchanges would require the approval of the supervisor

For Tier 2 there is an additional condition that such incentives cannot apply in the first 10 years. (10 years refers to the period after which the insurer may be incentivised to redeem, not a flip between who chooses to redeem)

 

[Ppan13]

Hi,
...
2b) Sufficient duration: Does the tier 3 requirement mean that the undertaking cannot issue debt with maturity less than 5 years to be an admissible asset under SII?

...
Many Thanks,
Joe

Yes.

"The basic own-fund item ... is undated or has an original maturity of at least 5 years, where the maturity date is the first contractual opportunity to repay or redeem the basic own-fund item" Article 77 paragraph 1(c) of https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32015R0035&from=EN

 

[Ppan13]

4) SCR - Intangible Risk: What comprises the 'Intangible' risk? There is no notes on it within the chapter, how does the standard formula calculate a capital charge for this risk, and how is it accounted for if an internal model is used?

Intangible asset risk could include things like cost of brand infringement, damage to good will, and intellectual property litigation. However, under Solvency II, intangible assets are given an economic value only when they can be sold separately and where there exist quoted prices in an active market for the assets. If this criteria is not met then they are valued at nil for SII even if they have a value under GAAP, e.g. for goodwill.

Any kind of reduction in market value, or unexpected lack of liquidity of the relevant market for the intangible (impeding a transaction) could contribute to intangible asset risk, as well as any risks inherent in the nature of the asset itself , e.g. deteriorating of public image, risks linked to no longer benefiting in the future from the asset.

I’m not sure how a capital charge would be modelled. If I had to guess, it would be scenario analysis, perhaps relying on any similar external events to overcome lack of relevant internal loss data and applying ‘expert judgement’ for calibration of the scenarios.

But since it's unlikely many intangible assets of insurers can be assigned a 'fair value' (by virtue of being traded on markets) in the first place, I'm not sure how often in practice this capital charge has to actually be considered (besides 'nil')

 

[Ppan13]

8) Counterparty risk - Type 2 Exposures: Chapter 4, page 29 details that the capital requirements for Type II Exposures are 'based on an immediate shock, assuming losses of 90% of receivables which have been due for more than three months and 15% on other receivables'. How does a firm forecast what receivables won't be due for more than 3 months, and which won't be, over a one-year time horizon?

I think most companies have a ‘credit control’ function (within the Finance department) which is monitoring aged debts/ potential bad debts and can provided the actuarial team the data on the receivables grouped by # months overdue.

 

[Ppan13]

7) Premium Risk Factors: Chapter 4, page 27 states that the 99.5% VaR factors for premium risk are applied to NEP, but on the following line say premium risk factors are based upon claims data that is gross of reinsurance? Is the premium risk factor different to the VaR factor applied to the premium data? And if so, why are premium risk factors applied to gross data as I thought the capital charge for Underwriting risk was based upon net data?

i think of the premium risk factor as being the result of multiplying the NEP by the VaR factor, which is itself derived from gross data but can further be multiplied (depending on LOB) by an XoL reinsurance factor (so everything is still on a net basis)

 

[Joe Warren]

Thanks again Ppan13 - all very clear and very helpful.

If anyone seeing this thread has an answer for Q3, Q5 & Q6 it would be much appreciated.

 

[Ppan13]

Thanks again Ppan13 - all very clear and very helpful.

If anyone seeing this thread has an answer for Q3, Q5 & Q6 it would be much appreciated.

You're welcome, Joe.

I have some ideas for the other 3 questions but I'm wary of giving you wishy-washy (or worse- misleading!) answers so would prefer for people strong in SII to also comment.

Re: Q5 & 6

Although we may have the best estimate of the liabilities, there are still risks linked to the future liability cash flows over their whole time horizon and uncertainty in their valuation, and the risk margin covers this (see Appendix B p59 https://www.cambridge.org/core/serv...of_the_risk_margin_solvency_ii_and_beyond.pdf) . The sum of the b.e liabilities and risk margin together make up the Technical Provision, which is (apparently) what another insurance undertaking would expect to receive to take over the liabilities and meet the obligations in full, or for the insurer itself to run-off the liabilities (which probably would take longer than a year). Therefore future time periods are also considered, not just t=1. It seems there's more than one way to estimate the risk margin, but for SII the route we use is the cost of capital method, as explained in the course notes. Under this method, the risk margin is calculated as and equated to the allowance for the cost of holding future SCRs; i.e. its the present value of the opportunity cost of holding solvency capital throughout the period to run off. (I understand 'cost of capital' to be in the sense of the opportunity cost of holding the capital, therefore.)

In your Q6 above you referred to "SCRs for a project".... were you referring to the diagram saying "Step 1: Project SCR"? If so, I interpret 'project' here as a verb, not a noun. It's saying "project SCR" to mean "get the run-off pattern for the SCRs". A single SCR is calculated at t = 0 and then a run-off pattern is being applied to it. If you look at SA3 paper Apr 2016 Q1(v)-(vii), this involves projecting the SCR run-off by first assuming the reserve run-off pattern (provided in the question) is a suitable proxy pattern.

The SCR itself is defined as the 99.5% VaR of the basic own funds over a one-year time horizon; given how each of SCR and risk margin are defined, I don't think there's a particular reason the time horizon for the SCR as selected/ defined by the regulator has to match the timing of full run-off of the liabilities as with the risk margin, since they are different things (though related). You could calculate the 99.5% VaR factors for other time horizons, e.g. in SA3 paper April 2013 Q1(iv) they quote the risk charges at ultimate as well as for a one-year time horizon.

I don't know if I've really answered your questions but perhaps if you try to do those 2 past year paper questions I've mentioned above it will give you a better feel for how everything links together.... enough to answer the exam questions, anyway!

 

[Busy_Bee4422]

Hi,

I have a string of Solvency II questions that I would appreciate if I could get some help answering:

3) MCR Calculation: Chapter 4, page 23 states that the MCR is calculated by taking the sum of 'a factor applied to technical provisions (not including the risk margin) for each line of business, net of reinsurance, subject to a minimum of zero, and a factor applied to written premiums in each line of business over the last 12 month period, net of reinsurance, subject to a minimum of zero'. Why is a factor applied to the written premiums as well as the technical provisions, wouldn't the technical provisions already contain an allowance for unexpired risk (i.e., premium risk)? Is the factor that is applied to the written premiums to represent risks not yet incepted over the one-year time horizon?

Many Thanks,
Joe

My take is something like this.
You need capital to deal with the possibility of adverse developments. This uncertainty arises because of uncertainties in exposure and uncertainty in your estimate of the liability for incurred claims (reserve risk).
If you look at the full non- life BSCR uncertainty in exposure is broken down further into catastrophe, premium reserve and lapses. The MCR measures exposure risk by using a factor of the premium. The reserve risk under your BSCR is similar to the factor applied to your technical provisions under your MCR.
In principle, the MCR is using a simpler method than the SCR to measure the minimum capital required to cover somewhat similar risks but at a more global level with a factor method.

 

[Ppan13]

MCR measures exposure risk by using a factor of the premium. The reserve risk under your BSCR is similar to the factor applied to your technical provisions under your MCR.

Hi Busy_Bee4422, thanks for your explanation. What confuses me is that the MCR takes into account the Net Written Prem of previous 12 months (for the premium factor) but also adds in the TP factor (alpha) multiplied by b.e. TP, which itself also accounts for premium provisions (on top of claims provisions). So in respect of the premium provisions (future exposure) component of the TP in the TP factor of the MCR, isn't there some degree of double-counting/ overlap with the MCR premium factor?

(In effect, I have the same query as Joe regarding this).

Thanks

 

[Busy_Bee4422]

Hi PPan
I see it as more of setting capital requirements for 2 different risks. In the case of the reserve risk, it is about premium reserves being inappropriate whilst in the exposure case, it is about premiums being inappropriate more or less underwriting risk.

 

[Ppan13]

Hi PPan
I see it as more of setting capital requirements for 2 different risks. In the case of the reserve risk, it is about premium reserves being inappropriate whilst in the exposure case, it is about premiums being inappropriate more or less underwriting risk.

Thanks again Busy_Bee4422.

The statement you made above on its own makes sense to me, but then when I try to relate it back to the original question, I still get puzzled. When modelling the SCR using the SF, is there an equivalent to the 2nd part you mentioned above? ("exposure case, it is about premiums being inappropriate more or less underwriting risk")?

In the notes for SF for SCR, they refer to non-life underwriting risk as "comprising premium and reserve risk, cat risk and lapse risk". There are VaR factors for premium risk and Var factors for reserve risk.... do the 'factors for premium risk' then correspond 'premium reserves being inappropriate' or 'premiums being inappropriate' (the distinction you made above)? If the former case (premium reserves), where does the latter (exposure side) come into the SF, and if the latter are referred to by the premium risk factors, isn't this then already feeding into the MCR calc when the TP is multiplied there (i.e. exposure side risk has already been accounted for in the TP)?

I have had a look at the full SF template spreadsheet to try to get to the bottom of this and I'm trying to track back through the components, but I'm not sure it will answer my question conceptually, so any further thoughts are most welcome.

 

[Busy_Bee4422]

Hi PPan

Maybe you can have a look at the template that I am familiar with https://www.lloyds.com/resources-an...serving/capital-guidance/standard-formula-scr. If you look in the tab Non-Life and NSLT Health P&R if you look at the rows from 59 onwards they have the premium volumes. The rows 31 to 56 have technical reserves. They are combined calcs in rows 10 to 30 that are used to calculate the SCR in cell D5.

 

[Ppan13]

Hi PPan

Maybe you can have a look at the template that I am familiar with https://www.lloyds.com/resources-an...serving/capital-guidance/standard-formula-scr. If you look in the tab Non-Life and NSLT Health P&R if you look at the rows from 59 onwards they have the premium volumes. The rows 31 to 56 have technical reserves. They are combined calcs in rows 10 to 30 that are used to calculate the SCR in cell D5.

Thanks. Yes, that's the same SF template spreadsheet I was referring to.

 

[Busy_Bee4422]

Hi PPan

I omitted stating that Solvency ii is a prospective capital regime. The premium risk relates to the exposure to be written over the next year or longer. You can measure exposure as either it's going to be the same as the prior 12 months' premiums or we may project them. The MCR uses the past 12 months, the SCR has either past or next 12 months for the future business but also includes a longer horizon too.

I think once you get that the premiums reserve risk relates to prior periods and the premium reserve relates to future periods it kinda eliminates the double counting neatly.

 

[Ppan13]

Hi PPan

I omitted stating that Solvency ii is a prospective capital regime. The premium risk relates to the exposure to be written over the next year or longer. You can measure exposure as either it's going to be the same as the prior 12 months' premiums or we may project them. The MCR uses the past 12 months, the SCR has either past or next 12 months for the future business but also includes a longer horizon too.

I think once you get that the premiums reserve risk relates to prior periods and the premium reserve relates to future periods it kinda eliminates the double counting neatly.

Many thanks, Busy_Bee4422. I read your message last night and decided to sleep on it, and this morning it's clearer to me. I realise I was actually muddling up the calculation of SRC with the calculation of TP in my mind, too, hence getting into a knot. I'll still sit down and go through it all again to satisfy myself, but your latest message above definitely put me back on the right track.

 

[Busy_Bee4422]

Many thanks, Busy_Bee4422. I read your message last night and decided to sleep on it, and this morning it's clearer to me. I realise I was actually muddling up the calculation of SRC with the calculation of TP in my mind, too, hence getting into a knot. I'll still sit down and go through it all again to satisfy myself, but your latest message above definitely put me back on the right track.

You're welcome.