Defined Benefit Pension Scheme: triennial valuation

9 September 2013 Performance and Resources Board 13 To consider Defined Benefit Pension Scheme: triennial valuation Issue 1. An update of the actua...
Author: Oswin Simmons
6 downloads 0 Views 953KB Size
9 September 2013 Performance and Resources Board

13

To consider

Defined Benefit Pension Scheme: triennial valuation Issue 1. An update of the actuarial advice we have received on the assumptions used to produce the draft valuation results, proposals to report these to Council for consideration, and an initial assessment of the implications for our long term pension provision. Recommendations 2.

The Performance and Resources Board is asked to: a. Agree to share the actuarial advice which we have received with the Trustees of the Defined Benefit Pension Scheme. b.

Note the proposals for reporting the valuation to Council.

c. Note the implications of the valuation and changes to national insurance for long term pensions costs.

Issue 3. At its meeting on 25 June 2013 the Performance and Resources Board received a report on the preliminary triennial valuation of our defined benefit pension scheme. This was based on the initial actuarial assumptions used by scheme Trustees. The initial valuation shows a past service deficit of £9m and a total future service contribution rate of 33.8% (this includes an employee contribution of 5%). The current total contribution rate is 24.4% 4. The assumptions used need to be agreed by both the Trustees and the employer, usually following a dialogue between the parties. Trustees may wish to revisit their initial assumptions, but early sight of their draft assumptions provides an opportunity for us to form an initial view. 5. The advice received from Aon Hewitt on these initial assumptions is set out at Annex A. In discussion with Aon Hewitt we have identified two assumptions where we believe it might be reasonable to adopt a different approach and would therefore ask Trustees to consider these as part of the valuation process. These are: a. The long-term gap between the Retail Price Index and the Consumer Price Index being 0.6% pa rather than 0.5% per annum. b. The discount rate using an approach consistent with the previous valuation (2% per annum above gilts for non-pensioners and in line with gilts for pensioners) rather than a time-dependent addition approach used in the initial valuation. 6. We propose that this initial advice is shared with scheme Trustees to allow them to consider it as part of their on-going work on the valuation. The Board of Pension Trustees next meet on 19th September 2013. 7. We plan to report to Council at its meeting on 25 September 2013 and set out the valuation process and relevant background. This would be ahead of a formal report asking them to agree the actuarial assumptions. 8. Once this has been completed there would be a further report to Council to seek approval for a recovery plan and any further decisions on pension’s strategy that were required. In addition to the formal sign off of the valuation process we would expect to report to Council on the longer term implications of the valuation outcome and our pension costs. 9. If Trustees agreed to the adjustments to the actuarial assumption set out above it would have the following impact: a. The Technical Provisions fall from £110.8M to £106.5M, so reducing the deficit from £9.0M to £4.7M.

2

b. The joint future service cost falls from 33.8% to 27.8%, so based on member contributions of 5% the Employer future service cost falls from 28.8% to 22.8%. 10. While these figures may change as a result of the dialogue between the employer and scheme Trustees it is reasonable to assume that there will be a deficit at the end of the process and an increase in the future employer contribution rate from 19.4%. 11. Further cost pressures will arise in 2016 if the government ends contracting out. This will increase the employer national insurance contribution by approximately 3% for members of our defined benefit pension scheme. 12. Employees in the scheme will also see their national insurance contributions rise. An employee earning £25,000 would face an annual increase of £270. In addition it is clear that the unusual prevailing economic circumstances are driving up the costs of defined benefit schemes and the required contribution rates may not be regarded as sustainable by either the employer or the employee. 13. While there will be an increase in costs related to the Defined Benefit pension scheme our new Defined Contribution pension scheme (known as the Aviva Group Personal Pension Scheme) will cover an increasing proportion of our staff and in the longer term our pension costs will reduce. However, in the short term, there is significant upward pressure on our pension costs following on from valuation and through planned national insurance increases. 14. If we choose not to meet these additional costs then a range of options would be available ranging from a move to a universal Defined Contribution pension scheme, to further reforms of the Defined Benefit pension scheme. These could include options such moving towards a career average scheme for future accrual and introducing higher employee contributions for higher paid staff. 15. Our initial assessment suggests that measures of this sort, combined with a growing Defined Contribution pension scheme would achieve total pension costs below 20% of payroll.

3

Supporting Information How this issue relates to the corporate strategy and business plan 16. The triennial valuation has significant financial consequences as it determines future employer contributions and the scale of any recovery plan. The process, and any potential changes to pension provision, therefore relates to strategic aim 7: continuing to use our resources efficiently and effectively. What engagement approach has been used to inform the work (and what further communication and engagement is needed) 17. Changes to pension arrangements for existing staff would usually require a consultation process to meet our statutory obligations and involve significant change to employment contracts. There will be options available to employers to make changes to their pension provision to offset increased national insurance costs that will not require statutory consultation, but still require a significant engagement process. We have a well established set of arrangements for communicating with staff on pensions. What equality and diversity considerations relate to this issue 18. Any changes to pension provision would be subject to an equality assessment. If you have any questions about this paper please contact: Andrew Bratt, Assistant Director – HR, [email protected], 0161 923 6215

4

13 – Defined Benefit Pension Scheme: triennial valuation

Annex A

GMC Staff Superannuation Scheme - Initial results of 31/12/12 valuation

A1

GMC Staff Superannuation Scheme Initial results of 31/12/12 valuation

Prepared for: Neil Roberts Prepared by: John Coulthard FIA and Ian Graham FIA 30 August 2013

Executive summary For the 31 December 2012 valuations: ƒ Valuation approach needs to be agreed by Council and Trustees, and Scheme Actuary must certify that resulting contributions are no lower than had he set them ƒ Initial deficit is £9.0M (set out in James Miller’s paper dated 25 June 2013) – in our view a deficit in the range c. £2-9M could be viewed as a credible/prudent ƒ Initial company future service rate is 21.6% to 28.8% - in our view these are credible (and below 21.6% is difficult to justify with current benefits) ƒ While a number of assumption changes are plausible, we have particular comments on the proposed changes to the discount rate methodology and CPI assumption ƒ We have prepared some alternative scenarios for your consideration, to make discussions more concrete To address long term risks to Council: ƒ Merit in articulating a broad long term plan ƒ Merit in considering potential changes to member benefits as part of valuation – CARE for future service would reduce rate by c.4.3%

1

Background TPR has published its 2013 annual statement on scheme funding: ƒ Trustees may need to make greater use of the flexibilities available than needed for their preceding valuations ƒ Trustees can use the flexibility available in setting the discount rates. The assumptions made for the relative returns of different asset classes may rise or fall from preceding valuations ƒ As a starting point, trustees should consider whether the current level of contributions can be maintained ƒ Trustees should allow for an appropriate level of risk to be taken that is neither overly prudent nor overly optimistic ƒ We are moving away from setting triggers focused on individual items – such as technical provisions – and continue to evolve a suite of risk indicators Source: TPR cover e-mail sent 9 May 2013 Trustee

Sponsor

Neutral

ƒ TPR has also been given a draft new objective “To minimise any adverse impact on the sustainable growth of an employer.”

2

What is likely impact of statement? ƒ Positives from corporate perspective – Eminently quotable words from Michael O'Higgins, Chair of TPR: “…does not require [Trustees] to be overly prudent” – Some encouragement to revisit discount rate – in contrast to 2012 statement – TPR acknowledgement that recovery periods may need to be longer than in the past, though care over reducing contributions

ƒ Potential flip slide – On flip side, Michael O'Higgins, Chair of TPR also says “I want to see pension trustees agree long-term strategies with employers” – Strong steer not to reduce currently agreed contributions: “As a starting point, trustees should consider whether the current level of contributions can be maintained” – Integrated risk management encouraged – integrated approach to funding, covenant, and investment (previously referred to as having a “financial management plan”)

Flexibility given by trustees in exchange for long term strategy?

3

Initial results (1 / 2) ƒ Past service and future service results (see Appendix for assumptions used): Assumption

Previous valuation results (£M)

Initial 2012 valuation (£M)

Neutral estimate (£M)

Solvency (£M)

Value of liabilities

68.2

110.8

62.9

142.2

Value of assets

67.4

101.8

101.8

101.8

Past service surplus (deficit)

(0.8)

(9.0)

38.9

(40.4)

Funding ratio

99%

92%

162%

72%

Assumption

Previous valuation (% of Pensionable Salaries)

Initial 2012 valuation (% of Pensionable Salaries)

Initial 2012 valuation with 1.1% p.a. allowance for outperformance (% of Pensionable Salaries)

Initial 2012 valuation with 1.7% p.a. allowance for outperformance (% of Pensionable Salaries)

Value of benefits accruing

24.3%

33.8%

29.1%

26.6%

Less member contributions*

3.1%

5.0%

5.0%

5.0%

Net cost of future benefit accrual

21.2%

28.8%

24.1%

21.6%

4

Initial results (2 / 2) ƒ Recovery Plans considered by Trustees: Illustrative Recovery Plan/Deficit contributions required including allowance for 1.7% pa additional investment returns and Employer paying 21.6% towards future benefit accrual

Illustrative Recovery Plan/Deficit contributions required including allowance for 1.1% pa additional investment returns and Employer paying 24.1% towards future benefit accrual

Recovery Plan end date

Additional lump sum deficit contributions required £M pa

Recovery Plan end date

Additional lump sum deficit contributions required £M pa

31 December 2016

2.84

31 December 2016

3.19

31 December 2018

1.43

31 December 2018

1.75

ƒ All assume that deficit contributions commence on 1 January 2014, so actual recovery period is 3 or 5 years ƒ Relatively short recovery period vs UK average of around 8 years – and noting market (and TPR) are expecting recovery plans to be higher for this round of valuations

5

Comments on Technical Provisions assumptions (1 / 3) Assumption

Used in initial results

Comment

Discount rate

UK Gilt yield curve plus 1.5% p.a. until 31 December 2019 (i.e. 7 years)

Implicit “long term target” is currently Gilts+0%, when all members are pensioners

declining to 0% p.a. by 31 December 2038 (i.e. 26 years)

We have the following comments:

The proposed new approach is time-based, and focuses thinking on achieving a “long term target” of Gilts+0% by 26 years using relatively prudent assumptions (so expect to arrive quicker using best estimate assumptions) ƒ As part of this methodology change, the value of the liabilities increases by £2.2M, and the future service rate increases by about 4.4% ƒ It is more usual to have an separate explicit long term target, and we believe this could be a better approach TPR is suggesting that some Trustees may be prepared to increase discount rates in light of historic low gilt yields. If assumption was to be increased by 0.25% across the curve, this would reduce the Technical Provisions by £5M

RPI

RPI Gilt Curve

Propose no change

CPI

RPI less 0.2% p.a. at valuation date, then RPI less 0.5% thereafter

We believe the method used to “tweak” the assumption at the valuation date is not unreasonable Aon Hewitt’s best estimate view of gap between RPI/CPI has increased by 0.2% since previous valuation, and not unreasonable for this to be incorporated (although it is not yet possible to “hedge” CPI using market instruments so this would arguably increase the risk exposure, and may not be preferred) If the margin was to be increased by 0.2% p.a. this would reduce the Technical Provisions by around £4M

6

Comments on Technical Provisions assumptions (2 / 3) Assumption

Used in initial results

Comment

Investment outperformance in Recovery Plan

1.1/1.7% p.a.

This is 1/3 and 1/2 respectively of the expected outperformance over the next 3-5 years, with no allowance for gilt yield reversion

Salary increases

RPI + 1.5% p.a.

Salary increases are under company control (to a large extent) and so could be challenged. Reducing assumption by 0.25% would reduce the Technical Provisions by around £1M, and future service by about 0.7% p.a.

Mortality – base table

The standard SAPS tables, S1N[M/F]A based on member's year of birth with individual AHLM scaling factors

Based on experience view - propose no change

Mortality – future improvements

In line with CMI projections, with long-term rates of improvement of 1.75% p.a. (all members)

This increases Technical Provisions by around £1.6M. Use of 1.75% p.a. becoming more common in marketplace though at prudent end of market practice

Tax-free cash assumption

Members assumed to take the middle, 3/80th lump sum option

Seems reasonable – scope to consider experience

7

Margins proposed fairly typical of market practice though could push for higher allowance

Note that neutral estimate, which can be seen as the Scheme Actuary’s “best estimate” of future improvements, allows for 1.5% p.a. long-term improvements

Comments on Technical Provisions assumptions (3 / 3) Assumption

Used in initial results

Comment

Allowance for early retirement

‘Medium’ allowance made

Impact fairly cost-neutral (£1.2M gain in combination with allowance for early retirement) and improves accuracy of cashflows produced. Gain likely to be offset when factor review carried out following valuation. Propose no change

Withdrawal

Lighter allowance from 2035 and heavier allowance from 45-55

Impact fairly cost-neutral (£1.2M gain in combination with allowance for early retirement). Propose no change

Proportion married

90% assumed to be married at retirement (also includes implicit allowance for children’s pensions)

A typical assumption

Age differences

Husbands assumed to be three years older than wives

Propose no change – industry standard assumption and generally makes little difference to overall assumption

8

Using an 80% is also plausible, and would reduce value place on liabilities by around £1M

Future service ƒ Extent for quantum of liabilities to grow considerably: – About 47% of liabilities relate to actives – Average age is 37.4, i.e. reasonably young ƒ Variety of potential actions including: – Do nothing – Increase employee contributions – CARE for future service – CARE for future and past service – Reduce accrual rate – Close to existing entrants ƒ Trustee agreement, and member consultation required – not a process to be entered into lightly ƒ As indicative figures for moving to CARE: – Future service rate would reduce by about 4.1% – If salary link broken for past service, past service liabilities would reduce by about £6.7M (the size of this potential “escape valve” would likely reduce in future)

9

Source: 2012 NAPF survey

Potential Council position for discussion purposes While there are a number of potential areas to discuss in the valuation basis, we propose: ƒ Focusing on a small number of changes ƒ Being aware that under pension legislation any change to assumptions needs to be justified by a change in legal, economic or demographic circumstances. ƒ Noting that Trustee agreement is needed to the assumptions, and that the Scheme Actuary must certify that the resulting contributions are no higher than had he set them himself (this is a requirement of the Rules, and is relatively unusual) With this in mind, a credible and reasonable approach in our view would be to suggest: ƒ A gap between RPI and CPI of 0.6% p.a. rather than 0.5% p.a. ƒ To revert back to pre-/post-retirement discount rate approach This would have the following implications: ƒ The past service deficit would reduce from £9.0M to £4.7M ƒ The headline future service rate would reduce from 28.8% (before allowing or asset outperformance) of Pensionable Salaries to 22.7% of Pensionable Salaries A more radical proposal would be to amend the RPI / CPI gap to 0.7% p.a. ƒ This would result in a past service deficit of around £2M, and a headline future service rate of around 21.7%. We believe this would be more difficult to achieve Trustee agreement and Scheme Actuary sign-off.

10

“I want to see pension trustees agree long-term strategies” TPR Chair, Michael O’Higgins

Source: Aon Hewitt Global Pension Risk Survey 2013

Regulator’s change in focus mirrors (but extends?) existing industry trends 11

11

What does self-sufficiency actually mean?

Source: Aon Hewitt internal survey 2012

12

12

Sample long term strategy Core plan: Current position ƒ Assets: 70% growth, 30% matching ƒ Target: Technical provisions ƒ VaR of about £30M

Responding to change – Investment ƒ Reduce growth allocation in 5% bands automatically, when affordable ƒ Manual triggers to consider re-risking (if funding falls two bands below) ƒ Daily tracking, trade within 5 days Responding to change – Managing liabilities ƒ Close to new entrants ƒ Introduce CARE for existing staff ƒ Actively consider other options to smooth/accelerate plan

Core plan: Long term target ƒ Assets: 10% growth, 90% matching ƒ Target: Gilts + 0.0% p.a. ƒ VaR of about £5M ƒ 70% chance of reaching within 15 years 13

Responding to change – Covenant ƒ Annual review of key performance indicators ƒ If material detriment is believed to have occurred, long term plan (as well as short term actions) is revisited

Appendix

Appendix – Assumptions used for initial results ƒ Key assumptions used in valuing liabilities on all four bases are set out below Assumption

2009 valuation

Initial 2012 valuation

Neutral estimate

Solvency

6.4% pa pre-retirement 4.4% pa post-retirement

UK Gilt yield curve plus 1.5% pa until 31 December 2019 declining to 0% pa by 31 December 2038

Best estimate until 2019 then addition over UK Gilt yield curve reducing linearly to zero by 2038

UK Gilt yield curve less 0.25% pa (Future pensioners) UK Gilt yield curve (Current pensioners)

RPI inflation

3.9% pa

RPI Gilt Curve

RPI Gilt Curve

RPI Gilt Curve

CPI inflation

3.4% pa

RPI inflation less 0.2% pa

RPI inflation less 0.5% pa

RPI Gilt Curve

Salary inflation

5.4% pa

RPI inflation plus 1.5% pa

RPI inflation plus 1.5% pa

N/A

Post-retirement mortality assumption – base table

The standard SAPS tables, S1N[M/F]A based on member's year of birth with individual AHLM scaling factors

The standard SAPS tables, S1N[M/F]A based on member's year of birth with individual AHLM scaling factors

The standard SAPS tables, S1N[M/F]A based on member's year of birth with individual AHLM scaling factors

The standard SAPS tables, S1N[M/F]A based on member's year of birth with individual AHLM scaling factors

Post-retirement mortality assumption – future improvements

CMI 2009 core projections with long-term improvement rate of 1.5% pa for men and women

CMI 2012 core projections with long-term improvement rate of 1.75% pa for men and women

CMI 2012 core projections with long-term improvement rate of 1.5% pa for men and women

CMI 2012 core projections with long-term improvement rate of 1.5% pa for men and women

Discount rate

ƒ Neutral assumptions help to give idea of margins for prudence in initial 2012 valuation (although could argue some margins left in minor assumptions)

15

Technical Actuarial Standard R: Reporting Actuarial Information’ (‘TAS R’), ‘Technical Actuarial Standard D: Data’ (‘TAS D’), ‘Technical Actuarial Standard M: Modelling’ (‘TAS M’) and the ‘Pensions Technical Actuarial Standard’ (‘Pensions TAS’) apply to this document and the work relating to it, and have been complied with. The compliance is on the basis that General Medical Council is the addressee and the only user of this report and that it is only to be used in assisting the company in making decisions in relation to the actuarial method and assumptions for determining the technical provisions and recovery plans as part of the actuarial valuation at 31 December 2012 and development of longer term funding and investment strategies (‘flight plans’). Copyright © 2013 Aon Hewitt Limited. All rights reserved. Aon Hewitt Limited, 8 Devonshire Square London EC2M 4PL Registered in England & Wales No. 4396810 To protect the confidential and proprietary information included in this material, it may not be disclosed or provided to any third parties without the prior written consent of Aon Hewitt Limited. Aon Hewitt Limited does not accept or assume any responsibility for any consequences arising from any person, other than the intended recipient, using or relying on this material. Authorised and regulated by the Financial Conduct Authority.

Suggest Documents