Hi there, In the notes it says "It may prove difficult to obtain original terms reinsurane on with-profits business as the reinsurer would be obliged to follow the cedant's bonus rates". Similarly, I've seen in a past paper "It may prove difficult to obtain original terms reinsurane on unit-linked business as the reinsurer would be obliged to pay a proportion of benefit and so match the unit fund performance" I'm struggling to follow this, would someone be able to clarify why original terms is difficult to obtain and why risk premium doesn't also have this problem? Thank you in advance, Clare
Hi Clare Yes that's right. Original terms (OT) doesn't work well with either with-profits or unit-linked. Under OT the reinsurer takes a percentage of both the claims and the premiums. This sounds fair enough with a term assurance. The reinsurer gets30% of the premium say, and so pays 30% of the death claims (the reinsurer may pay some reinsurance commission to cover some share of expenses). Note there are no maturity or surrender claims to pay. So the insurer has only reinsured the mortality risk, which is what it wanted to do. If OT was used for unit-linked, then the reinsurer would need to pay 30% of all the claims, so that would include surrenders and maturities too. This means the reinsurer would have to set up a unit fund that exactly matched the insurer's unit fund. This is hard to do, hence why OT is rare for unit-linked. There is a solution to this problem - the reinsurer could deposit back its reserves with the insurer so that its investments behaved exactly the same as the insurer - but this sounds like a lot of extra work. It's much easier to just use risk premium reinsurance and just reinsure the mortality risk, charging a premium of q_x * (sum assured - unit fund). Similarly, OT wouldn't work well with with-profits because the reinsurer would have to pay 30% of the maturity an surrender claims as well as the death claims. This means that again, the reinsurer would have to match the investments o the insurer, and again a deposit back arrangement might help a bit. But the situation is worse here because the maturity and death benefits depend on how big the bonuses are that will depend on the smoothing policy of the insurer. The reinsurer won't want the risk that the insurer smooths upwards and ends up paying out more than the asset share. Again risk premium makes more sense, maybe charge q_x * (sum assured plus declared bonuses - reserve). I hope this helps to explain why OT works for some policies, but not others. Best wishes Mark
Hi Mark, thanks for your explaining. I have some further questions: 1- for TA, an insurer faces more mortality but still some investment risk. When the insurer only transfer mortality risk under the OT reinsurance? 2 - Can OT apply to immediate annuity business? If so, it seems this can transfer all three risks, i.e. longevity, investment and expense risks via the reinsurance payments? 3 - Your reply explains under risk premium reinsurance charges a premium of q_x * (sum assured - unit fund) for UL business. Does this mean it effectively saves non-unit reserves? Hence, it transfers mortality risk if insurer chooses a matched investment strategy. Can we use reinsurance to transfer other risks under NUR, e.g. maturity or surrender risks? Other questions relating to Fin Re: 4 - what is the logic behind that a higher initial reinsurance commission is only regarded as an asset but not as a liability? 5 - course notes saying "Financial reinsurance is not effective under accounting or supervisory regimes where credit can already be taken for the future profits and/or where a realistic liability must be held in respect of the loan repayments." How to understand this? Does this mean, under prudent basis, a smaller initial profit but higher expected amount at later years due to surplus arising? While, under realistic basis, we expect a level profit over the years. Hence, we are not expecting further profit emerging. But is the release of solvency capital regarded as surplus arising? Many thanks
Hi Damien Thanks for these questions. I'll answer them in turn below. 1. Under original terms (OT) the reinsurer could take half the premiums and so pay half the claims for example. They will need to invest the reinsurance premiums so will be exposed to some investment risk. However, the reserves for term assurance policies are small (compared to endowments where the reserves need to grow to meet the maturity payout) and so the investment risk is much smaller than the mortality risk. 2. In theory OT could be used for annuity business. The reinsurer would again get half of the premiums let's say, which they would need to invest, and would have to pay half of the claims. So the reinsurer would be taking on half of the insurer's investment and longevity risk. The reinsurer would not be paying any of the insurer's expenses though, so the insurer would retain all its expense risk. The reinsurer would have to pay its own expenses and so would carry its own expense risk. 3. Yes, the insurer should be able to reduce its non-unit reserves if it uses risk premium reinsurance on its unit-linked business. Matching is not relevant to this business. The policyholder chooses which unit funds to invest in. The insurer will choose the investments in the non-unit fund, but this will not affect the reinsurance transaction in any way. Original terms reinsurance would transfer part of the maturity risk as the reinsurer would be paying a fraction of all the claims, not just the death claims. Surrender risk is primarily due to the accumulated charges not covering the expenses so far if the policy surrenders too early - as the reinsurer does not pay a proportion of expenses then this risk has not been transferred to the reinsurer. As explained in my previous post, OT doesn't work very well with unit-linked and so wouldn't usually be used. If the insurer wanted to cover investment risk it would be better to use derivatives rather than reinsurance. For example, swaps could be used to hedge a guaranteed annuity rate, put options would give some protection against a guaranteed maturity value. 4. You'r right to query the capital effectiveness of higher initial reinsurance commission. It won't work under many regulations in a similar way to 5. below. The insurer may have a choice between low reinsurance commission now and low reinsurance premiums in the future, or high reinsurance commission now and higher reinsurance premiums in the future. The two possibilities will have been priced to have similar present values, so you're right that any increase in assets (high reinsurance commission) could be offset by an increase in liabilities (high future reinsurance premiums). If the solvency regulations work this way then there is little to be gained from having higher reinsurance commission (except maybe for improved liquidity). However, some regulations may only account for the reinsurance premiums in the year in which they are paid. In this case the reinsurance commission will increase the assets at outset and the higher reinsurance premiums will lead to corresponding losses in later years. 5. The discussion in the notes here is ignoring the impact on solvency capital. Yes, you're right about the two valuation bases. Using a prudent valuation basis we often get an initial loss (due to setting up the big prudent reserves) followed by later profits (as the reserves are released). Using a best estimate basis we should get an initial profit followed by zero profits in later years (as the reserves are exactly the right amount to cover the cashflows if experience turns out the same as our best estimate). Financial reinsurance effectively brings forward the later profits from the release of the margins in prudent reserves. There are no such margins to bring forward if we use best estimate reserves, hence there's no point using financial reinsurance for this reason. I hope this helps, but do let me know if you have further questions. Best wishes Mark
Hi Mark, Thanks for this explanation. Please could I clarify something. Your comment that risk premium reinsurance makes things simpler b/c the reinsurer only reinsures mortality risk, I understand it for the UL example, b/c we're dealing with a fixed death SA. But in the WP example, in order for the reinsurer to strictly assume no investment risk then would it not need to reinsurer only the initial SA, rather the initial SA + attaching bonuses? Thanks, Matt
Hi Matt For the WP risk premium, the risk premiums would be determined annually (or even monthly). So, I think a charge of q_x * (sum assured plus declared bonuses - reserve) would be ok and not result in the reinsurer taking investment risk. (If investment returns are good and result in higher bonuses being attached, this is reflected next year/month when the sum at risk being insured reduces accordingly, as does the risk premium.) Hope that helps Lynn
