CMI Working Papers 34 and 35

CMI Working Paper 35 contains the final graduations of the CMI SAPS experience for the period 2000 to 2006 based on data collected by 30 June 2007

A description of the methodology used to produce the graduations is outlined and the key statistics for each of the graduations are summarised. It also provides a summary of points made in feedback during the consultation on the draft tables published in CMI Working Paper 32 and the Committees response to this feedback.

It is the responsibility of any actuary or other person using a published mortality table to ensure that it is appropriate for the particular purpose to which it is put.

These tables have been approved by the Management Board for adoption by the IFoA. It should be noted that adoption implies that the IFoA is satisfied that the tables have been well-constructed and subjected to rigorous peer review, such that the IFoA is happy for the tables to be published in its name. It does not carry any implication that the tables are appropriate as a standard for any particular purpose. 

The mortality rates are also available in a spreadsheet and in version 3.3 of the CMI Tables Program (STP).

CMI Working Paper 34 sets out the methodology and assumptions underlying the dataset that has been used for the graduations contained in Working Paper 35

It also documents recent changes to the methodology and assumptions and the reasoning behind these changes. Working Paper 34 refers to Comparison of Approaches for Calculating Initial Exposure

Comment on the graduated tables added on 31 December 2008

The CMI SAPS Mortality Committee wishes to draw to the attention of users of the ‘S1’ series of mortality tables that female mortality rates are higher than male mortality rates at ages 55 and 56 in the ‘all pensioner’ series S1PFA and S1PMA tables. This is a genuine feature of the data at these ages and also at some younger ages. As there was little or no data at these younger ages, these tables were extended to younger ages in an arbitrary manner (see section 6 of CMI Working Paper 34); as a result this feature persists down to age 28 for qx and 29 for μx. This feature does not appear in the other graduations.

The Committee regrets that this feature of the ‘all pensioner’ graduations was not drawn to readers' attention in CMI Working Paper 35. Users of the tables are reminded that they should be aware of how they have been constructed and should only use them when it is deemed appropriate by the user to do so.

Summary data: all classes of pensioner, males, split by pension amount

• Males £0-£1500
• Males £1500-£3000
• Males £3000-£4500
• Males £4500-£8500
• Males £8500-£13000
• Males £13000-£25000
• Males £25000+

Summary data: all classes of pensioner, females, split by pension amount

• Females £0-£750
• Females £750-£1500
• Females £1500-£3000
• Females £3000-£4750
• Females £4750-£8000

Summary data: all classes of pensioners, males and females, not split by pension amount

• All data

Please note that central exposed to risk is shown in these spreadsheets.