Authors: Andre Fifer Andre Clarke
We are supportive of the additional provisions within TAS 310 v1.1 as these should help ensure high quality technical actuarial work is carried out by those advising parties involved in setting up and operating CMP schemes.
Key points from our response
- A recommendation that communications to users include only material cross-subsidies
- Noting the inconsistency in expectations from TPR in the draft Code of Practice with regards to not planning and reserving for continuity option 3 at scheme inception, but a requirement in TAS 310 v1.1 P5.1d to consider the impact of closure on the uncertainty of benefit adjustments
- A request for confirmation that actuarially certified member communications are not expected to fall within the scope of TAS 310 v1.1 (either intentionally or unintentionally)
Appendix 1 - response to consultation
Actuarial equivalence
1. What are your views on the proposed provisions P7.1 and P7.5? Are there any specific rating factors (for example, age) that you think should or shouldn't be used in determining actuarial equivalence?
We believe it is reasonable for practitioners to consider all material rating factors, though we note that this is a very broad phrase, and would lead to a large number of factors needing to be considered.
As set out in Unconnected Multiple Employer Schemes (UMES) regulations actuarial equivalence is intended to avoid excessive cross subsidies. We therefore believe it may be appropriate to amend P7.5b to relate to "the expected material cross-subsidies". For completeness, we do not expect this re-wording to be applicable to P7.1b as we still expect practitioners to consider all cross-subsidies.
Age, sex, salaries, postcode and industry are likely to have the biggest impact on actuarial equivalence. Naturally no rating factors for protected characteristics should be used. We note if machine learning models are used with significant data points, this may inadvertently allow for protected characteristics. We believe practitioners should be cautious in these instances.
We however note that excessive underwriting undermines the CMP model, for example early deaths subsidising older retirees is a fundamental part of CMP schemes. It is expected that some material cross-subsidies will need to remain in order for schemes to successfully function.
2. What are your views on provisions P7.2 and P7.6? Are there any other issues relating to the choice of method that should be communicated to intended users?
It is reasonable to require practitioners to consider the extent of cross-subsidies from the choice of method of actuarial equivalence. Especially as TPR's draft code requires trustees and scheme proprietors to demonstrate the risk of unintended cross-subsidies in their demonstration of scheme design soundness, so this is essential for them to be confident in meeting their authorisation duties.
The requirements in P7.6 are reasonable, though we would expect practitioners to also demonstrate the sensitivity of these cross-subsidies under certain scenarios, such as large demographic shifts due to high employee turnover or employer entry / withdrawal.
3. What are your views on provisions P7.3 and P7.7? What other considerations are there in the choice of relevant period?
We consider the additional provisions are reasonable.
We would suggest P7.3c and P7.7c refer to the "circumstances under which the accrual rate of benefits may need to be reassessed" rather than "changed" during the relevant period. Consideration will also need to be given to how any change is communicated to members, given TPR requires that an actuary certify member communications.
4. What are your views on provisions P7.4 and P7.8? Are there any other areas where you expect a difference between the assumptions for actuarial equivalence and the actuarial valuation?
We believe the provisions are reasonable.
5. Do you agree with the proposal to extend the requirements of P3.1 to accrual rates? Please provide reasons for your answer and alternative approaches where relevant.
Yes, as far as the requirement is only applicable to UMES.
Actuarial factors
6. What are your views on the proposed new provisions P8.3 and P8.7? Do you believe the proposed changes create any additional requirements in relation to single employer CMP schemes? Please explain your rationale.
These requirements are reasonable.
Our expectation is that this should not create additional burden on single employer CMP schemes, given accrual must be uniform across members.
We note the inconsistency in that P8.7 requires communication on any "material cross-subsidies… as a result of these actuarial factors," but no provisions in P8.1 to P8.5 for practitioners to consider these cross-subsidies.
Viability assessments
7. What are your views on the proposed changes to provisions in relation to viability assessments? Are there any other areas an actuary should consider in relation to soundness of a CDC scheme?
We agree with removing the word "trustees" from P5.1, given the scheme proprietor of an UMES must approve the viability report.
P5.1d is also reasonable, though we note this impact will depend on the point in time at which closure to members or accrual happens. Does the FRC have an expectation of a particular point in time at which the impact of scheme closure needs to be assessed?
Additionally, there is currently no expectation in TPR's draft code of practice that trustees will plan or reserve for continuity option 3 when first applying for authorisation. We believe this presents an inconsistency in expectations, in that the FRC expects practitioners to consider and communicate the impact of closure, but trustees do not need to plan for this.
Separately, we note the typographical error in P5.1 "…soundness of a scheme design in in relation to…"
The proposed changes to P5.2b are also reasonable.
Further, assuming the consultation document refers to retaining P5.5c (as P5.6c does not exist) in section 5.13, then we are happy with this decision.
We expect UMES are more likely to experience large changes in demographics due to entry and exit of employers, and consideration will need to be given on the impact of this on the soundness of scheme design.
Finally, P5.1a relates only to the assessment of intergenerational cross-subsidies. In light of UMESs potential to cause other forms of cross-subsidy, does the FRC intend to amend P5.1a to require practitioners to consider the risks of other material forms of cross-subsidy when assessing the soundness of the scheme?
Other considerations
8. Are there any areas where additional guidance would be helpful? If so, please set out the specific areas and/or provisions where guidance may be helpful.
As noted elsewhere in the response, guidance on whether the FRC expects member communications to fall under the scope of TAS 310 for UMESs, given the need for actuarial certification of these communications.
9. Are there any further aspects of technical actuarial work you expect to be impacted by the introduction of UMESs regulations which are not adequately covered by the proposed changes to TAS 310? If so, please explain what they are.
We do not believe there are any other areas of the draft UMES regulations that require further changes to TAS 310. However, this is on the assumption that the FRC does not expect member communications to fall under the scope of TAS 310, despite TPR's requirement that actuaries are expected to certify the accuracy of member communications in the viability assessment.
Timing and implementation
10. What are your views on the proposal that the standard would be effective from 31 July 2026? Please set out any practical difficulties which you believe this might cause.
We support this target, to ensure this aligns with the UMES regulations being enacted.
We expect this will only impact providers who are close to finalising their authorisation submissions, as the new requirements may cause additional work.
Impact assessment
11. Do you agree with our impact assessment? Please give reasons for your response.
The impact assessment is reasonable, and we note that this impact is predominately due to the underlying regulations, rather than the standards set out in TAS 310 Version 1.1.
