2nd Non-life underwriting risk data request

Qualitative Questionnaire

Insurer Name:

<QSP

QSP>

Insurer number[1]:

<QSR

</QSR

Natural catastropherisk

  1. Is the business captured correctly allocated to the appropriate zones based on risk address?

N1a

<N1a

  1. If business is not captured by risk address, what steps are being taken to improve this before SAM Implementation?

N1b

<N1b

  1. What difficulties do you foresee with this process?

N1c

<N1c

  1. Can you give an approximate indication of how your total business is split between South Africa and other territories?

<N2>

<N2>

  1. Please indicate which of the following reasons, if any, are a cause for incorrect allocation of business amongst zones:

a)Group Scheme business – brokers capture incorrectly or don’t capture relevant exposure information

b)Large risks – only the primary location or head office address is captured

c)Other

N3

<N3

  1. If you responded (a) or (b) to Question 3, please can you elaborate on your plans to remedy this problem going forward?

N4

<N4

  1. If you responded (c) to Question 3, please can you provide detail on the problem in question?

N5

<N5

  1. Please indicate for your sources of information an approximate percentage split between Internal/Broker/UMA in terms of where the information is captured?

N6

<N6

  1. Please can you provide us with your definition of corporate business, should you write this class?

<N7

<N7

Premium and reserve risk

  1. Segmentation
  2. Was segment allocation done on a policy/contract level, on a policy section level or on a claim type/peril level? If a mixture was used dependent on the product or line of business, please describe your approach. Considering Liability as an example, a company’s approach might have been to allocate any Liability cover provided under an Engineering policy to Engineering, whilst allocating any Liability cover provided under a Commercial Property policy to Commercial Liability if shown as a separate policy section. In the latter case, it may have been decided to allocate Liability cover to Commercial Property if a separate Liability policy section is not shown.

PR1a

<PR1a

  1. Please describe the quality of your data with regard to the ability to allocate premiums and claims to the correct segments (regardless of accessibility issues asked about in questions x). Please refer to historical information as well as to data currently being captured.

PR1b

<PR1b

  1. Are any steps being taken to improve the quality of historical data with regard to segmentation?

<PR1c

<PR1c

  1. Are any steps being taken/will steps be taken to improve the future quality of data with regard to segmentation? What are your main challenges in achieving this?

<PR1d

<PR1d

  1. Data
  2. Please describe any distorting effects in your data, e.g. resulting from large events or large single claims. Which segments, origin years and development years do those relate to? Have any of those effects been allowed for in the data submitted? Please explain.

PR2a

<PR2a

  1. Please describe the quality of your claims data with regard to the availability of data on claim payments, claim expenses, salvages and recoveries on an individual claim and transaction level. Are any steps being taken to improve the quality of data?

PR2b

<PR2b

  1. Please describe the quality of your claims data with regard to the availability of full movement data on reported outstanding claim estimates, as opposed to storing only initial and current estimates. Are any steps being taken to improve the quality of data?

<PR2c

<PR2c

  1. Please describe any issues regarding the accessibility of your claims data?

<PR2d

<PR2d

  1. Please indicate for your sources of information an approximate percentage split between Internal/Broker/UMA in terms of where the information is captured?

<PR2e

<PR2e

  1. Please describe the quality of your claims data with regard to the ability to extract/remove data relating to an event, such as a natural catastrophe.

<PR2f

<PR2f

  1. Please describe the quality of your claims data with regard to the ability to extract/remove portions of claims relating to specific reinsurance contracts.

<PR2g

<PR2g

  1. Calibration methods
  2. Do you have comments on the appropriateness of the proposed premium risk calibration methods, in general and/or for your business? Please describe.

PR3a

PR3a

  1. Do you have comments on the appropriateness of the proposed reserve risk calibration methods, in general and/or for your business? Please describe.

PR3b

PR3b

  1. Please describe any alternative premium risk calibration methods that might be used.

PR3c

PR3c

  1. Please describe any alternative reserve risk calibration methods that might be used.

PR3d

PR3d

  1. Please comment on the potential for “double counting”under this and other components of the non-life underwriting risk module of the standard formula.

PR3e

PR3e

  1. Confidentiality of data
  2. There is a proposal to publish industry triangles per segment based on an aggregation of data submitted. Do you have any objections against and/or concerns with this, bearing in mind that summary claims data is currently available from the statutory returns?

PR4a

PR4a

  1. In terms of confidentiality, do you have any concerns with the FSB providing the data to the NLUR Working Group as received? If so, please explain why. The current intention, should confidentiality be deemed an issue by the industry, is to have the data scaled and made anonymous by the FSB before providing information to the Working Group.

PR4b

PR4b

  1. If you indicated in b) that you have data confidentiality concerns with the FSB providing data to the NLUR Working Group, an alternative approach suggested is that the Working Group has access to the data while in the offices of the FSB for calibration purposes. Would this adequately address your concerns?

PR4c

PR4c

  1. Would published industry triangles be beneficial to your company? How would you potentially make use of this information?

PR4d

PR4d

  1. Are your company’s run-off triangles used internally? Please list current uses thereof? Are the triangles the same as those submitted? If not, how are they different? E.g. different segmentation, removal of distorting factors, different claims definitions?

PR4e

PR4e

  1. Do you have any other comments relevant to the calibration?

<PR5

<PR5

[1]Please use the Ixxx number which is part of the reference number 10/10/Ixxx/8 in Statement A1 of the annual statutory return.