Fall 2017 #15

For IRIS 4 in this problem, is the reason we do (Commission %) * (QS% * UEPR) instead of (Commission %) * UEPR because we first need to get the UEPR for NonAffiliates? Meaning the UEPR given (ie 220-120) are the total UEPR and not the Unaffiliated UEPR?

Comments

  • Yes, that's correct.

  • Got it, thanks!

  • For part d in this problem, if the reinsurer were to go insolvent, one of the answers state "IRIS 11 would be 10/40". Does that mean if a reinsurer goes insolvent the prior year reserves and the incurred in schedule P part 2 gets restated?

  • Not quite. This was actually a bit tricky. The denominator of 40 (PHS) doesn't change from what was calculated in part (c). But the reason it's reasonable to use 10 in the numerator of IRIS 11, instead of 3 as in in part (c), is that if the reinsurer goes insolvent, the reinsurance contract is no longer in effect so that:

    • AY 2015 One-Year Direct Reserve Development
    • = AY 2015 One-Year Net Reserve Development

    In part (c), these quantities were different due to the quota share percentage of 70%, but now the quota-share percentage is 0%.

  • Wouldn't the reserve development be calculated using Schedule P though? If so, doesn't that mean the prior year's net reserve would need to be restated to be direct?

  • I guess I am thinking of this as if this is a commutation with $0 settlement, and thinking the triangles would be distorted but we wouldn't get 10?

  • I think that's what was done. If you do the part (c) calculations in a slightly different order and apply the quota-share percentage first, you get:

    • net AY 2015 losses at yr-end 2015 = 80 x (1-70%) = 24
    • net AY 2015 losses at yr-end 2016 = 90 x (1-70%) = 27

    So the 1-yr reserve development is 27-24=3.

    But if you restate both these net values assuming no reinsurance recoverables you get:

    • restated net AY 2015 losses at yr-end 2015 = 80 x (1-0%) = 80
    • restated net AY 2015 losses at yr-end 2016 = 90 x (1-0%) = 90

    So the 1-yr reserve development is 90-80=10.

    Note however, that the source text doesn't specifically address the scenario of an reinsurer going insolvent, so it isn't clear that the prior year reserves need to be restated. I think you could make a case for adjusting only the yr-end 2016 value (possibly writing off the yr-end 2015 recoverables in bad debt or some other balance sheet account) so the restated reserve development would be 90-24=66.

  • In part (c), we're given: "Total losses unpaid as of December 31, 2015" = 80. Are we meant to assume that this represents incurred, since there is no amount given for paid in 2015? It seems confusing, since 90 - 80 ought to represent (AY2015 Incurred @2016) - (AY2015 Incurred @2015), right?

    Am I missing something?

  • I don't think you have that quite right. The 90-80 is intended to capture how the original $80 reserve developed over 2016. It would include changes to case and IBNR amounts, but also changes due to claims being paid and closed for amounts different than their original reserves.

    Here's how I think about it:

    I believe the unpaid value of 80 represents case + IBNR (not incurred which generally means paid + case.) We don't really need to know what was paid in 2015 on AY 2015. For CY 2015, all we need is the total reserve (unpaid) at the end of 2015. (Recall the company started operations in 2015 which simplifies things.)

    Then at the end of 2016, we want to know how that value of 80 has developed. The solution calculates this development as (150-75)+(35-20). Let's consider these separately:

    • (150-75)
    • = (total unpaid for all years) - (unpaid for AY 2016)
    • = unpaid for AY 2015 at the end of 2016
    • = 75

    But this unpaid amount of 75 doesn't tell the whole story. If you kept going and did just this part of the calculation for 2017, 2018, etc, you would eventually end up with 0 when all the claims had been paid and closed. This would not tell how the original reserves developed over that time. IRIS 11 is basically a measure of accuracy of original reserve amounts.

    What happens along the way is that claims are continually being paid and closed. If they are always paid and closed for the exact amount of their reserve then that means the original unpaid of 80 at the end of 2015 was exactly correct. But sometimes (often) claims are paid for more than their reserves, so you have to take into account these payments to truly understand reserve development.

    So every time you make a payment and reduce the reserves, you have to add that payment back to the numerator of IRIS 11. If the $15 payments in 2016 for AY 2015 had exactly equaled the corresponding reserve amounts, then the reserve development would have been:

    • 80-80=0

    But in this case, what must have happened is that payments in 2016 for AY 2015 were actually $10 greater than the original reserve amounts, and this is what you want to capture when calculating reserve development. That's why you have to include the second piece that captures amounts paid in 2016 on AY 2015:

    • (35-20)
    • = (total paid during 2016 on all years) - (paid for AY 2016)
    • = 15

    That's kind of a long explanation. Hope it helps.

  • Very helpful, thank you!

  • for part C the examiner's report lists one common mistake as "using the adjusted PHS instead of the original". why is it not appropriate to remove surplus aid the following year like we did for the 2015 ratios?

  • Well, my unsatisfying answer is that the IRIS ratios aren't defined that way. If 4 is unusual for year X you recalc that year's 1,2,7,10,13 but there's nothing about checking the prior year's ratio 4 and using it to adjust ratio 11 (or the second prior year's 4 to adjust 12).

    I don't think it's an inappropriate thing to consider...but they just asked for the ratio, so the correct answer is to give them the ratio.

  • That would be my answer too. Even though IRIS 11 uses surplus (for the prior year), there is no specific requirement to recalculate IRIS 11.

    (I suppose you could do a deep dive and investigate NAIC's reasoning for not requiring IRIS 11 to be recalculated but that isn't discussed in the source text and would be beyond the scope of what you need to know for the exam.)

  • For part a, the formula I am used to for surplus aid is: ceded commissions / ceded premiums * unearned premium non-affiliates, such as in 2018.Spring.16.

    Here, does the ceded percent of 30% = ceded commissions / ceded premiums?

    Then, to restate the first question in this thread to make sure I understand, we apply the 70% quota share percent to the total unearned premium number to get the uep for the non-affiliates?

  • edited October 2021

    Here's how I solved the problem using the formulas given in the first web-based problem of quiz 1 in the NAIC.IRIS wiki article. I think this will answer your questions. If not, let me know.

    First, here's the link to the quiz:

    Using the formulas for surplus aid given in the web template:

    • CWP_total (Ceded WP total)
    • = 70% x (WP for 2015)
    • = 70% 220
    • = 154

    And:

    • CC_total (Ceded Commissions total)
    • = 30% x CWP_total
    • = 30% x 154
    • = 46.2

    Now, to get CUEPR, we first need this formula for UEPR during 2015

    • EP = WP - (change in UEPR)
    • 120 = 220 - (change in UEPR)

    Therefore:

    • (change in UEPR) = 100

    And:

    • CUEPR
    • = 70% x (change in UEPR)
    • = 70% x 100
    • = 70

    Putting this all together, you can now calculate surplus aid:

    • surplus aid
    • = CUEPR_non-affiliates x CC_total / CWP_total
    • = 70 x 46.2 / 154
    • = 21

    And finally you can calculate IRIS 4:

    • IRIS 4
    • = (surplus aid) / surplus
    • = 21 / 40
    • = 52.5%

    This is greater than 15% so it's an unusual value.

  • Thanks for your long explanation on Part C. I understood how the original reserves developed over that time. BUT when I checked on Fall 2018 # 17, 1 year loss reserve development is calculated by the difference between incurred loss in Schedule Part 2. Could you please explain the difference?

  • In 2017.F.15, incurreds are not given, so we have to work with paids and reserves separately. In 2018.F.17, incurreds are given, which makes the job easier.

    For the same set of accident years:

    incurred1 - incurred0 = paid1 - paid0 + reserve1 - reserve0
    = reserve1 - [reserve0 - (paid1 - paid0)]

    Calendar year paid reduces the original reserve. The difference of this reduced original reserve and the ending reserve is considered the reserve development.

  • edited April 2022

    Gotcha! Making sense for me that reserve0 grows to reserve1 + (paid1-paid0) after 1 year. Then the adverse development is the difference between these two parts. Thanks very much.

  • Sure, good luck.

  • JJJJJJ
    edited March 2023

    If incurred1 - incurred0 = paid1 - paid0 + reserve1 - reserve0, why'd we not need to know what was paid in 2015 on AY 2015? How's is related to the company only starting the business in 2015?

  • Because (paid1 - paid0) is loss paid in 2016 on AY2015. You don't need the paid in 2015 on AY 2015.

    One-year development is on all prior years. Because the company started business in 2015, you need to focus only on AY2015 to get one-year development.

  • Got it. Thanks!

  • Sure, good luck.

  • edited February 24

    for part a, a key point is reinsurance ceded commission = % of reinsurance ceded premium. Surplus aid = WP x 0.7 x 0.3/WP x 0.7 x UEP of non- aff = 0.3 x (220-120) x 0.7

  • I don't follow the second part of your equation, but the rest of your comment is correct.

Sign In or Register to comment.