In order to work through the GMPE project, a scheme will need to be able to first establish the affected members and then their relevant pension figures. The affected members will include any member who had GMP accrued between 17 May 1990 and 05 April 1997 (when GMP accrual ceased), including GMP transferred in, in respect of this period), who have a liability in the scheme. Dependants whose pensions arose from a member who had any such GMP are also affected members. Clearly for any active members there’ll be no pension amounts to split, and data would be expected to be readily available when required. However, if there are any data matters requiring extra work for active members (e.g. pension splits on divorce), this should be included in the data gathering project at the same time as the non-actives to ensure it’s available when required. We haven’t explicitly covered ‘no liability members’ – these are usually members who have transferred out, died (without any further dependant pension due) or taken a full commutation of benefits. It’s possible some of these members (and/or their dependants) have lost out due to unequal benefits having been paid in the past. These members may or may not be in scope, depending on a number of factors and advice should be sought. For any which are in scope, data may be particularly difficult to obtain and this should be considered in the data gathering planning process.

For each affected member, the pension will need to be split into its constituent parts for the period 17 May 1990 to 05 April 1997 (and as far back as 06 April 1978, or possibly from earlier date of joining) for a conversion if considering all pension (e.g. if converting all GMP including any transferred-in GMP). The range of data requirements is covered in Appendix 1.

In an ideal world, all the data you may need to accurately determine whether a GMPE adjustment is due and to calculate the GMPE adjustment and accumulation of arrears would be readily available for every affected member. In practice, this is unlikely. As a result, each scheme will need to consider:

  • What data is readily available
  • What other data might be required and how it can be obtained, including the potential cost of doing so
  • What data no longer exists or might be impossible or impracticable to obtain
  • Whether the data obtained needs to vary by sub-group of members. For example the additional data requirements for those with anti-franking considerations
  • The risks of not obtaining full data
  • The validity and accuracy of any such data e.g. there may be existing issues or concerns

Much of this Guidance focuses on practical approaches schemes can adopt to data gathering including work-arounds and assumptions.


July 2020 – GMPE Data Guidance


Akash Rooprai – Chair (ITM)

Fiona McCartney (Aviva)

Dan Holmes (SAUL)

John Doherty (XPS Admin)

Brendan Doherty (Intellica)

Andy Campbell (Willis Towers Watson)

Sharon Bill (ITM)

Ian Dack (Mercer)

Megan Green (Capita)

Steve Nicholson (Equiniti)

Lucy Cresswell (Barnett Waddingham)