1. The Statute of the Pension Reserve Fund (PRF) of the Council of Europe, Article 6.2, states that every three years, the Management Board shall carry out a thorough review of the Fund’s investment policies, goals, guidelines and performance, its actuarial assumptions, rate of contributions and its capitalisation target, and shall report its findings, decisions and recommendations to the Committee of Ministers. The current Third Three-year Review covers the period 2013-2016.

2. The document first presents an Executive Summary with the Management Board’s proposals for approval by the Committee of Ministers and a Statement of the conclusions drawn throughout the review process. A detailed description of how the Board carried out the Review is presented in APPENDIX I.

PROPOSALS FOR APPROVAL

3. The Management Board of the PRF invites the Committee of Ministers, in accordance with the provisions of Resolution Res(2006)1 on the Statute of the Pension Reserve Fund:

· To take note of the information presented in this document;

· To approve the following changes to the Statute:

- Article 4.2, of which the last sentence should read as follows:

“The Board should decide when special circumstances apply and shall inform the Committee of Ministers should a deviation from the SAA occur.”;

- Article 5.3 should read as follows:

“The members of the Management Board, with the exception of those belonging to the Council of Europe's staff, shall receive a remuneration.[2]”;

- Article 6.2, of which the last sentence should read as follows:

“Every four years in parallel with the actuarial study provided for in Article 3.b, the Management Board shall carry out a thorough review of the Fund’s investment policies, goals, guidelines and performance, its actuarial assumptions, rate of contributions and its capitalisation target, and report its findings, decisions and recommendations to the Committee of Ministers. These reviews shall be complemented, at least once in two years, with a limited intermediate follow-up study to assess the evolution of the Fund.”;

· To approve a Strategic Asset Allocation aiming to obtain a net real target return of 3.4% p.a. in the long term with an associated volatility of 7.8%, as detailed below:

ASSET CLASS	STRATEGIC ASSET ALLOCATION
World Equity	45%
Emerging Markets Equity	10%
TOTAL EQUITIES	55%
World Government Bonds (EUR hedged)	23%
Euro Area Corporate Bonds	12%
TOTAL FIXED INCOME	35%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%

· To approve the Socially Responsible Investments (SRI) strategy below, to make it relevant to the asset classes included in the SAA proposal. The changes as regards the previous strategy are underlined.

o The current SRI Policy should apply to at least 90% of the investment portfolio.

o Exclusion, when possible, of companies with more than 5% of their activities involving alcohol, tobacco, weapons, pornography and gambling. The selection of companies should preferably be made with a 'best in class' approach. For those asset classes for which this is not possible, the strategy shall rely on an SRI Engagement Overlay approach. This consists of running a portfolio with traditional financial criteria while companies in the portfolio are approached by an external SRI expert with the aim of carrying out a process of dialogue and negotiation to convince them to improve their corporate behaviour in terms of Environmental, Social, and Governance criteria (as expressed in the SRI policy of the PRF).

o In the evaluation of companies, the asset manager should use his own environmental, social and governance criteria, which will be examined by the Management Board to assess their compliance with COE's SRI principles.

o Government bonds: as most issuers of government bonds from global developed countries and all EMU countries are members or observers of the COE and thus respect the values of the Organisation, no specific SRI criteria will be applied.

o Alternative investments (real estate and infrastructure): investments in real estate and infrastructure may be placed in traditional (non-SRI) funds as long as combined they represent less than 10% of the portfolio.

STATEMENT OF THE BOARD’S CONCLUSIONS

1. The Second Three-Year Review of the Pension Reserve Fund (PRF) of the Council of Europe covers the period of 2014 – 2016 and examines all phases and aspects of the investment process, i.e.:

A. Governance

B. Investment policy

C. Investment strategy

D. Investment structure

E. Administration

A. GOVERNANCE

2. The Governance defines the responsibilities and tasks of the different bodies involved in the management and administration of the PRF.

3. In general, the Governance has worked well during the period under review. The Management Board would like to propose the following improvements for approval to the Committee of Ministers:

· A change to the article 4.2 of the Statute of which the last sentence should read as follows:

“The Board should decide when special circumstances apply and shall inform the Committee of Ministers should a deviation from the SAA occur.”;

· A remuneration scheme for Board members. Thus, article 5.3 should read as follows:

“The members of the Management Board, with the exception of those belonging to the Council of Europe's staff, shall receive a remuneration.[3]”;

· A change of the Statute to reflect the parallel execution of the Global Contribution Rate Study and Review of the Fund, and to allow an intermediate assessment of the evolution of the Fund. Thus, the last sentence of article 6.2 should read as follows:

B. INVESTMENT POLICY

4. The Investment policy describes the objectives of the PRF. It has two aspects: financial and extra-financial [Socially Responsible Investments (SRI)].

5. The financial policy establishes the investment target return of the Fund given the objective of the Organisation for the Fund, the resources in terms of assets and net income and the time horizon the PRF disposes of to achieve it. The financial policy is based on the PRF’s objective as defined in the Statute, its expected life cycle and the value of the Fund at the time of the Review.

6. The Board concludes that sustainability - i.e. the pension contributions plus the return of the Fund investments meet the pension benefits as they come due - is the appropriate objective of the Organisation for the PRF, as it is in line with the Statute and more favourable, or less costly, in financial terms.

7. The Board notes that - given the parameters impacting the life cycle of the Fund in 2017, including the reasonable target return assumption of 3.4% - the objective of sustainability can only be achieved with an increase of the GCR from 30.1% to 37.85%. The Board acknowledges the decision of the Committee of Ministers to gradually increase contributions in this respect over a four-year period, and that a mid-term review of the required level of funding will be presented in the last quarter of 2019.

8. The Board considers the present SRI policy of the Fund to correctly translate the values of the COE into its investments. A full description of the principles of the COE applicable to the SRI criteria is presented in ANNEX I of APPENDIX I.

C. INVESTMENT STRATEGY

9. The Investment strategy designs the most adequate portfolio deemed to meet the objectives of the investment policy, i.e. the target return. It has two aspects: financial and extra-financial (SRI).

10. The financial investment strategy defines the optimal asset allocation expected to produce a target return considered reasonable and able to meet the objectives of the investment policy.

11. In the Board's view, the reasonable target return taking into account portfolio diversification should be 3.4%. Likewise, given its current cash flow profile and the planned Reviews (Adjustments) the Fund shall undergo, the PRF is able to bear the volatility associated with an investment strategy aiming at 3.4%, both in the long term and short term. The Committee of Ministers should note that with 3.4% the COE would still be partially financing pensions with investment risk.

12. The Board would like to propose the following SAA, deemed necessary to achieve a 3.4% real return in the long term, for approval to the Committee of Ministers:

Table I - Proposal for SAA - Target Return: 3.4%

ASSET CLASS	STRATEGIC ASSET ALLOCATION
World Equity	45%
Emerging Markets Equity	10%
TOTAL EQUITIES	55%
World Government Bonds (EUR hedged)	23%
Euro Area Corporate Bonds	12%
TOTAL FIXED INCOME	35%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
*Volatility*	7.8%

13. Given the long-term horizon for which the investment strategy has been created, the extreme difficulty of generating higher returns when reacting to changing market conditions, and the higher volatility and costs of a dynamic implementation, the asset allocation will continue to be implemented in a static way. This means the Board would faithfully implement the strategy over time, without consideration for short-term market expectations or fluctuations. Only in periods of exceptional circumstances may the Board decide to diverge from the Strategic Asset Allocation. This SAA remains valid until the next review (the next intermediary assessment or the Fourth Four-Year Review).

14. The extra-financial (SRI) investment strategy describes the optimal way to implement the SRI principles of the Council of Europe in the portfolio.

15. The Board proposes to the Committee of Ministers for approval to extend the current SRI strategy of the PRF to the other asset classes included in the SAA proposal. The updated SRI strategy would read as follows:

16. The current SRI Policy should apply to at least 90% of the investment portfolio. The changes as regards the previous strategy are underlined.

o Alternative investments (real estate and infrastructure): investments in real estate and infrastructure may be placed in traditional (non-SRI) mutual funds as long as combined they represent less than 10% of the portfolio.

D. INVESTMENT STRUCTURE

17. The present investment structure of the PRF is as follows:

Table II - Present Investment Structure of the PRF

ASSET CLASS

(share and rebalancing limits)

BENCHMARK

MANAGEMENT STYLE

INVESTMENT VEHICLE

ASSET MANAGER

World equity

(35% +/-5%)

MSCI World

(Net Return)

Index-linked (traditional)
+ SRI Engagement Overlay

Mutual fund:

iShares Developed World Index Fund

Engagement Overlay Provider: BMO

BlackRock

Emerging markets equity

(15% +/-3%)

MSCI Emerging markets

(Net Return)

Index-linked (traditional)
+ SRI Engagement Overlay

Mutual fund:

Vanguard Emerging Markets Equity Stock Index Fund

Engagement Overlay Provider: BMO

Vanguard

Euro area equity

(10% +/-2%)

MSCI EMU

(Net Return)

Active (SRI)

+/- 50% Mutual fund:

Amundi Actions Euro ISR

Amundi Asset Management

+/- 50% Mutual fund:

Allianz Valeurs Durables

Allianz Global Investors

Euro area government bonds

(24% +/-5%)

Citigroup EMU Government Bonds Index

(Total Return)

Index-linked

(traditional)

Mutual fund:

SSgA EMU Government Bond Index Fund

State Street Global Advisors

Euro corporate bonds

(16% +/-5%)

Markit iBoxx Euro Corporates

(Total Return)

Active (SRI)

Mutual fund:

BNP Parvest Sustainable Bond Euro Corporate

BNP Paribas Asset Management

18. The Board has reviewed and assessed the benchmarks, the management style, the investment vehicles and asset managers, and concluded the following:

19. As regards the management style, the Board considers that, given the current investment strategy of the PRF, each asset class is managed with the most appropriate management style. Only a change in the SRI-policy of the PRF, implementing an engagement overlay on an index-linked portfolio, could warrant a change in management style (from active to index-linked).

20. As regards the investment vehicle, taking into account the sound legal framework of mutual funds, their simplicity in terms of accounting and administration, their higher external control and moderate costs, the Board considers that mutual funds are best suited for the investments of the PRF in traditional asset classes. For the new alternative asset classes, other investment vehicles such as funds owned by limited partnerships and trusts will be considered.

21. On the evaluation of asset managers, the Board concludes that the assets of the PRF are well managed by the selected managers, respecting their investment objective, cost-effectively in comparison with the peer group, and with gross performances above their respective benchmarks and most of their peers. The Engagement Overlay Service provider is also producing the expected results.

22. While the benchmarks and rebalancing limits for the current SAA are considered appropriate, the benchmarks for the new SAA would be the following:

Table III - Benchmarks of the new SAA

ASSET CLASS (Strategic share +/- rebalancing limit)	BENCHMARK
World equity (45% +/- 5%)	MSCI World (Net Return)
Emerging markets equity (10% +/- 2%)	MSCI Emerging markets (Net Return)
Global Government bonds EUR hedged (23% +/- 3%)	BofA Merrill Lynch Global Government Index EUR Hedged (Total Return)
Global Government bonds EUR hedged (23% +/- 3%)
Euro corporate bonds (12% +/- 2%)	Markit iBoxx Euro Corporates (Total Return)
Global Direct Real Estate (5% +/- 2%)	To be determined, depending on the selected investment
Global Direct Infrastructure (5% +/- 2%)	To be determined, depending on the selected investment

E. ADMINISTRATION

23. The administration of the PRF covers the treasury management, the execution and administration of the investments, the accounting and the reporting.

24. A part of the PRF’s assets is held as short-term treasury, earmarked to complement the monthly transfer of staff and member States’ contributions that serve to pay pension benefits and administration costs in the month. This part is managed following the yearly reviewed Treasury Management Plan. The Board concludes that the latter has proved effective during the period under review and that short-term investments have been managed prudently while producing a return above that of the benchmark and the return of Euro short-term money market mutual funds.

25. The Board notes that on the execution and administration of investments no operational incidents during the period 2013-2016 are to be reported, meaning that the operational risk in the administration of the PRF has been successfully minimised.

26. On the execution of investments and rebalancing operations during exceptional circumstances, the Board notes that the procedure on exceptional circumstances effectively dealt with extreme market volatility, e.g. in 2011, and was not triggered since.

27. To comply with international standards on performance calculation and to better monitor the risks of the portfolio, the Board decided - from the implementation of a new Strategic Asset Allocation onwards - to report risk measures on a one, three and five years basis as well as since the beginning of the strategy's implementation. These risk measures would be calculated using weekly observations.

28. Management costs, as a percentage of assets held by the Fund, diminished from 0.40% in 2013 to 0.26% in 2016. As from the implementation of the new SAA at the beginning of 2016, the share of actively-managed funds decreased in the portfolio, resulting in a sizeable decrease in management fees (net of rebates). This effect was partially offset in 2016 by the increase in dilution fees (reverted to the mutual fund participants) and the costs related to the new Engagement Overlay strategy with BMO Global Asset Management. Simultaneously, custody fees increased between 2015 and 2016 because the transfer agent no longer used the low-cost settlement platform Euroclear for the World Equity and Emerging Market Equity funds (representing 50% of the portfolio).

29. Total costs, including administration costs, showed the same trend, diminishing from 0.56% in 2013 to 0.35% in 2016. This also reflects the growing portfolio size, and fairly steady administration costs.

APPENDIX I: THIRD THREE-YEAR REVIEW - FULL REPORT

INTRODUCTION

1. The Statute of the Pension Reserve Fund (PRF) of the Council of Europe (COE), Article 6.2, states that every three years, the Management Board shall carry out a thorough review of the Fund’s investment policies, goals, guidelines and performance, its actuarial assumptions, rate of contributions and its capitalisation target, and report its findings, decisions and recommendations to the Committee of Ministers.

2. This Third Three-Year Review covers the period 2013-2016^{^[4]} and examines each phase of the investment process:

A. GOVERNANCE – review of the well-functioning of the Fund's governing processes, leading to a proposal for a change of the Statute (for approval by the Committee of Ministers).

B. INVESTMENT POLICY – determination of the objectives of the Organisation for the Fund and how they can be translated into a target investment return on the one hand; determination of Socially Responsible Investment (SRI) principles in line with the COE's values on the other hand, resulting in the definition of the Investment Policy (for approval by the Committee of Ministers).

Two aspects are considered: financial and extra-financial or SRI.

B.1 Financial – review of the Organisations' objectives for the PRF, description of the projected life cycle and determination of the reasonable target return.

B.2 Extra-financial – examination of the Socially Responsible Investment (SRI) policy.

C. INVESTMENT STRATEGY – design of an investment portfolio that fulfils the objectives of the investment policy, i.e. that meets the target return while respecting the SRI-related goals.

Two aspects are considered: financial and extra-financial or SRI.

C.1 Financial – review of the following points:

· the universe of asset classes;

· hypotheses of expected return and risk;

· expected return of the current strategy with updated hypotheses of return and risk;

· portfolio optimisation resulting in the proposal of a new strategy (for approval by the Committee of Ministers);

· risk implications.

C.2 Extra-financial – review of the SRI strategy in place, its relevance in view of the COE’s principles, of the present development of the SRI industry and of the proposed SAA, leading to a proposal for an update of the current SRI strategy (for approval by the Committee of Ministers);

D. INVESTMENT STRUCTURE – review of the investment vehicles, management style and asset managers. For the latter, a peer comparison of fees, costs and performance is made.

E. ADMINISTRATION – review of activities related to the day-to-day management of the Fund:

E.1 Treasury management.

E.2 Accounting and back office.

E.3 Reporting.

3. To undertake this Review, the ISRP analysed the present situation for each item, and made an assessment of possible improvements. Conclusions on such assessments are presented throughout the document for approval by the Committee of Ministers.

A. GOVERNANCE OF THE FUND

Current situation

4. The Governance of the PRF is established in its Statute. A description of the Governance of the PRF, including an overview of the developments during the past three years and challenges ahead is given below.

FIGURE I - Governance structure of the PRF

5. TheCommittee of Ministers approves the general investment guidelines covering both the investment policy and the investment strategy; further, it approves the operational budget and the annual financial statements. Moreover, it follows up on the situation and performance of the Fund.

6. The Management Board is nominated by the Committee of Ministers (four members on the proposal of member countries, one member on the proposal of the Secretary-General and one member on the proposal of the Staff Committee). The Board is in charge of presenting proposals to the Committee of Ministers relating to the investment policy and strategy, and of ensuring their implementation. The latter involves drafting the investment procedures for the execution of investments, definition of the investment structure, selection of asset managers, supervision of the execution of operations and monitoring of results. Management Board members meet a minimum of four times a year; occasionally meetings are held electronically and/or decisions are taken by written procedure. The Board reports twice a year to the Committee of Ministers on its activities and on the situation and performance of the Fund.

7. The table below shows the composition of the Management Board since its creation:

Table IV - Composition of the Management Board (2006 - 2017)

*MANAGEMENT BOARD MEMBER*	*APPOINTED ON THE PROPOSAL OF*	*DATE OF 1^ST APPOINTMENT*	*DURATION OF APPOINTMENT*	*RE-APPOINTED*	DURATION OF RE-APPOINTMENT
Mr PETAUTON former Chair	France	24/05/2006	4 years	No
Ms SANDRI IN FLORIDI former Chair; former Vice-Chair	Italy	24/05/2006	5 years	Yes	3 years - mandate ended 23/05/2014
Mr SCHÖN former Vice-Chair	Germany	24/05/2006	3 years	No
Mr HUZAREK	Poland	22/11/2006	5 years	Mandate was ceased
Mr MERGOS former Chair	Greece	24/05/2009	3 years	No
Mr WEINBERG former Vice-Chair	Germany	25/11/2009	2 years	Yes	3 years - mandate ended 21/11/2014
Mr MARTINS	Secretary-General	24/05/2006	3 years	Yes	3 years - mandate ended 24/05/2012
Mr BAECHEL - Ongoing	Staff Committee	24/05/2006	4 years	Yes (2x)	3 years – mandate renewed for another 3 years starting 24/05/2016
Mr KESIK	Turkey	24/05/2010	3 years	No
Ms DAHREMÖLLER	Secretary-General	24/05/2012	3 years	Yes	3 years - left the Organisation on 01/10/2016; replaced by Ms Prinz first ad interim (18/01/2016), then officially (01/02/2017)
Mr GRIGORYAN - Ongoing Vice-Chair	Armenia	12/09/2012	3 years	Yes	3 years
Mr COŞKUN	Turkey	24/05/2013	3 years	Yes	3 years
Mr DOISY - Ongoing Chair	France	24/05/2014	3 years	Yes	3 years - mandate renewed starting 31/05/2017
Mr EKELI - Ongoing	Norway	22/11/2014	3 years	Yes	3 years - mandate renewed starting 22/11/2017
Mr ÖCAL - Ongoing	Turkey	24/05/2016	3 years
Ms PRINZ	Secretary-General	01/02/2017	remaining term of office of Ms Dahremoeller until 23/05/2018		Left the Organisation as per 01/10/2017
Ms SIDEBOTTOM - Ongoing	Secretary-General	01/10/2017	remaining term of office of Ms Prinz until 23/05/2018

8. The International Service for Remunerations and Pensions (ISRP), appointed by the Secretary-General on a recommendation of the Management Board, carries out the following tasks:

· Secretariat of the Management Board: preparation of meetings, reporting documents and comprehensive assistance to the Board in decisions regarding the financial and administrative management of the Fund.

· Implementation of the investment strategy: call for tenders for financial services providers, negotiation of contracts in collaboration with the Financial and Legal departments of the COE.

· Daily operations: follow-up of asset managers, treasury management, investment of new contributions and rebalancing.

· Accounting of the pension fund: booking, and reporting on both, the pension fund and the investment activity; integration in the consolidated accounting of the Organisation in collaboration with the Finance department of the COE, and preparation of the biannual and annual financial statements.

· Preparation and management of the operating budget.

· Relationship with the COE’s Auditors.

· Performance follow-up.

· Cost monitoring.

9. In 2017, the external auditors of the Council of Europe recommended to consider the establishment of a Service Level Agreement between the Organisation and the ISRP, completing the Letter of Mission currently in place. Furthermore, the auditors recommended a regular examination of the ISRP by auditors experienced in the organisation and performance of the pension fund sector, and communication of the results to the Management Board. The implementation of such recommendations is under consideration.

Assessment

10. The Board addressed four questions regarding the governance of the PRF:

a. The Board contemplated on installing a self-assessment procedure to evaluate its own performance and effectiveness.

b. The Board reflected on the principle of non-remuneration of Board members, as defined in article 5.3 of the Statute of the PRF, and wondered if this could be of nature to reduce the pool of potentially qualified candidates. The latter could be seen as somewhat contradictory with article 5.1 stating that the Board should be composed of four specialists with expertise in the management of investment funds. The Board decided to invite the Committee of Ministers to consider a remuneration scheme for Board members, with the exception of those belonging to the Council of Europe's staff, thus modifying the article 5.3 of the Statute.

c. The article 4.2 of the Statute allows the Management Board to deviate temporarily from the strategic asset allocation in periods of exceptional circumstances, such as the implementation of a new investment strategy or the occurrence of extreme uncertainty in financial markets. The Board should decide when special circumstances apply and shall seek to obtain the approval of the Committee of Ministers immediately. The Board noted that special circumstances could require immediate action from the Board. A timely reaction, therefore, could be jeopardized by the need to obtain the approval from the Committee of Ministers. The Board would like to propose to alternatively notify the Committee of Ministers should a deviation from the SAA occur.

d. The Board analysed how the pension financing decisions were taken within the COE[5] and considered that the actuarial study that determines the Global Contribution Rate (GCR), and the Three-Year Review determining the Reasonable Target Return (RTR) should be synchronised. This would allow the Committee of Ministers to analyse the interaction of both rates at the time of making the related decisions. This approach was adopted for the Third Three-Year Review and the GCR Study of 2017. The frequency of the Fund Reviews could be set at every four years, in parallel with the GCR actuarial studies. The calendar of the studies concerning the financing of the pension schemes should be as consistent as possible with the decision processes of the COE.

CONCLUSION i:

In general, the Governance has worked well for the period under review. The Management Board would like to propose the following improvements to the Committee of Ministers, for approval:

· A remuneration scheme for Board members to be added to article 5.3 of the Statute. Thus, article 5.3 would read as follows:

“The members of the Management Board, with the exception of those belonging to the Council of Europe's staff, shall receive a remuneration.[6]”

· A change of the article 4.2 of the Statute of which the last sentence should read as follows:

“The Board should decide when special circumstances apply and shall inform the Committee of Ministers should a deviation from the SAA occur.”

B. INVESTMENT POLICY

11. The investment policy of the PRF considers two aspects:

· Financial: This part of the investment policy establishes the financial target of the Fund given the resources in terms of current assets, future net income and the time horizon the PRF disposes of to achieve the financial objective of the Fund. It is based on the Organisation's objective for the Fund, as established in the Statute, the current value and the expected life (income and expenses) of the Fund. The aim is to determine the reasonable rate of return the investments can achieve and the objectives feasible with a reasonable target return.

· Extra-Financial: The objective of the COE is to invest the PRF in Socially Responsible Investments (SRI), in accordance with the Organisation’s values.

B.1. FINANCIAL INVESTMENT POLICY

Current situation

12. The PRF’s objective, stated in art. 1 of its Statute, is “to smooth, in the medium and long term, the financing of the member States’ obligations under the Organisation’s Pension Schemes”. In the context of an expected increase in budgetary expenditures for pensions, member States could have been required, in the future, to largely and abruptly increase their annual contributions in order to meet their obligations. To avoid this, the COE decided to:

· Increase the contributions to pensions now and allocate the surplus to an investment fund, such that contributions would not need to be abruptly increased in the future.

· Create an investment fund; the return on investments should generate extra income to help finance pensions.

13. In summary, by creating the PRF, the COE intended to smooth the contributions of member States over time and to create an extra source of income. This still implies that, as long as the PRF exists, member States’ contributions to pay pensions would increase steadily (as per the evolution of the salary mass), but would be lower than in a pure pay-as-you-go system. However, should the PRF stop to exist, contributions would need to be increased abruptly to replace the source of income derived from the PRF’s investments and to fully cover the future increase in the number of pensions benefits to be paid.

14. On this basis, when the Management Board last considered the investment policy, its primary objective was defined as “to expand the life of the Fund as much as possible while keeping a reasonable level of risk”. The reasonablelevel of risk is assessed as regards the following aspects:

· Cash flow profile: positive cash flows allow the Fund to bear more volatile investments as it will not need to realise eventual short-term losses to pay pensions.

· Portfolio diversification: the best practice to reduce investment risk or volatility (necessary to obtain the desired return) is to adequately diversify the portfolio. For this reason the Board seeks to create an appropriately diversified portfolio when recommending an investment strategy and associated return and volatility.

· Feasibility of the investments: the profile of the Fund can put some limits on the investment universe, the management style and therefore on the level of the achievable or reasonable return.

· Governance and administration framework: when the Organisation aims at regularly reviewing the Fund as part of the governance, the long-term risk can be better borne and more risky strategies are possible. In parallel, the governance and the administration framework may determine some features of the investment strategy.

15. Based on these parameters and a set of investment assumptions, the Board proposed in the last Review of the Fund in 2014 a long-term real reasonable target return of 5% yearly, associated with an investment risk (expressed by the annualised volatility of returns) of 8.5%. With the member States’ GCR set in 2013 at 30.1% of the salary mass and given the expected pension payments, this long-term real reasonable target return would make the Fund sustainable, i.e. the future inflows combined with the investment return would permanently be meeting the pension payments.

16. In summary, the current financial investment policy of the PRF entails the following elements:

· Objective: make the Fund sustainable, while keeping a reasonable level of risk.

· Rate of return required (target return) to meet the objective of the PRF: 5%, which was considered reasonable.

· The primary risk faced by the PRF was its potential depletion. Risk was assumed to come from the return of the investment strategy only – all other factors related to the life of the Fund were considered to remain stable, since out of the scope of the Board. It was measured as the standard deviation of annual returns: 8.5%.

Assessment

17. In the current third Three-year Review, two questions are raised concerning the financial investment policy:

1. Is the objective of the COE for the PRF still the same?

CONCLUSION ii:

The Board considers the objective of the Organisation for the PRF, derived from the Statute of the Fund - i.e. “to expand the life of the Fund as much as possible while keeping a reasonable level of risk” - appropriate to determine a rate of return/volatility to be achieved/borne by the investments.

2. Is 5% still a reasonable target rate of return?

18. The Board notes that the performance of financial markets in recent years has contributed very positively to the return of the PRF. In the Board’s view, the current low levels of interest rates, the moderate Gross Domestic Product (GDP) and productivity growth, and the current pricing of equity markets suggest that the future returns of various assets classes will be much lower in comparison to the previous Three-Year Review - done during 2014 - where 5% was kept as the RTR.

CONCLUSION iii:

The Board has conducted a thorough analysis to determine the reasonable target return using new assumptions for future expected assets returns [CM(2017)91]]. The conclusion from this analysis was, first, that in terms of portfolio diversification the long-term real reasonable target return (RTR) would be 3.4%. Likewise, the associated volatility of an investment strategy aiming at 3.4% would be 7.8%. This risk - both in the long and short term - seems in line with the risk appetite of the COE, and can be reasonably borne by the PRF given its current cash flow profile and the planned Reviews (Adjustments) of the Fund (see chapter C for the complete analysis).

3. The question that follows from this conclusion is: Can the RTR of 3.4% meet the objective of the Organisation for the PRF?

19. To respond to this question, the life cycle needs to be drawn as from 2016. For that, the parameters affecting the life cycle of the Fund need to be updated. The table below summarises the changes in assumptions/parameters since 2013 impacting the life of the Fund:

Table V - Assumptions & Parameters Impacting the Expected Life Cycle of the PRF

	VALUE IN 2013 (Source in italics)	VALUE 2016 (Source in italics)
Member States’ contributions - Global contribution rate (GCR)	30.1% (COE’s Global Contribution Rate actuarial study of 2013)	Several scenarios are being analysed: - Current GCR: 30.1% - Moderately increased: 32.08% - Required to keep the Fund sustainable with 4% TR: 35.75% - Required to keep the Fund sustainable and 3.4% RTR: 37.85% - Required to keep the Fund sustainable and 3.1% TR: 38.88% (SIRP/E(2017)16/REV1 Global Contribution Rate update as at 31/12/2016)
Staff contributions	9.0% COPS - 9.3% NPS (Contribution rate applied as from 1 January 2010)	9.5% COPS - 9.3% NPS - 9.4% TPS (Contribution rates applied as from 1 January 2015)
Fund assets	EUR 178 063 978 (at 31 December 2012)	EUR 319 428 350 (at 31 December 2016)
Investment income¹ - target return (TR)	5%	Several scenarios (geometrical adjustment): - Current: 5% - Required to keep the GCR unchanged and the Fund sustainable: 5.57% - Maximum achievable with limits on SAA: 4% - Reasonable: 3.4% - Reasonable and Conservative: 3.1% (Reasonable Target Return study of 2017 [CM(2017)91]
	Historical French inflation: 1.8%	ECB target minus the 10-year average difference between French and European inflation at 2016 YE: 1.85%
	Asset management and administration costs: 0.67%	Not considered as the Organisation will fund administration costs separately and asset management costs are assumed to be covered by the asset managers.
Level of Pension Benefits	As established in the Coordinated Pension Scheme (CoPS) and the New Pension Scheme (NPS) Introduction of the Third Pension Scheme (TPS) on 1/4/2013	Same as in 2013
Evolution of the COE’s population	2 207 actives (51% COPS, 49% NPS), 21 deferred and 714 pensioners (at 31 December 2012)	2 206 actives (44% COPS, 38% NPS and 18% TPS), 21 deferred and 846 pensioners (at 31 December 2016)

Note: all of the assumptions related to investment income are developed in Chapter C – Investment strategy.

20. Taking into account the characteristics of the Fund in 2016 as described in Table V, and assuming several TRs and several GCRs, the expected life cycle (evolution of the value of the PRF) as from 2016 are depicted in the table below:

Table VI - Return, GCR and Fund Evolution

RETURN	GCR	FUND EVOLUTION
5.57% - REQUIRED TO LEAVE THE GCR UNCHANGED AND MAKE THE FUND SUSTAINABLE ⁽¹⁾	30.1% (current)	The Fund would be sustainable but the return is not realistic (2)
5.0% - CURRENT	32.08% (moderate increase)	The Fund would be sustainable but the return is not realistic (2)
4.0% - MAXIMUM ACHIEVABLE	35.75%	Sustainable
	32.08% (moderate increase)	Depleted in 2056
	30.1% (current)	Depleted in 2051
3.4% - REASONABLE	37.85%	Sustainable
	32.08% (moderate increase)	Depleted in 2052
	30.1% (current)	Depleted in 2048
3.1% - REASONABLE AND CONSERVATIVE	38.88%	Sustainable
	32.08% (moderate increase)	Depleted in 2050
	30.1% (current)	Depleted in 2047

Source: ISRP, as per document SIRP/E(2017)16/REV1 Global Contribution Rate update as at 31/12/2016

Note (1): The return required to keep the Fund sustainable while maintaining the same GCR has increased w.r.t. the previous Review, where it was set at 5%. This is due to demographic changes, the inclusion of family allowances in the current actuarial study, and lower estimated future salaries which lower absolute contributions (see document SIRP/E(2017)16/REV1 for further reference).

Note (2): This is proved in the chapter related to the investment strategy.

21. The Board notes that, of the three schemes currently in place at the COE - the Coordinated Pension Scheme (CoPS), the New Pension Scheme (NPS) and the Third Pension Scheme (TPS) - only the CoPS presents a financing problem. Indeed, when considering the funding of the three schemes separately, both the NPS and the TPS are sustainable, i.e. the contributions plus the return of the Fund can pay pensions as they come due. They are even fully funded, i.e. the assets equal the scheme liabilities. Being fully funded implies sustainability, and the level of Fund assets is higher in a fully funded scheme. However, the CoPS presents a severe underfunding and is not even sustainable, due to the member countries' decision of not funding this scheme from its creation until 2003, while at the same time the staff contributions had been paid and used to pay the pension benefits. With a 3.4% RTR, to be sustainable the CoPS alone would need a contribution from the member States of 33% of the total salary mass from 2017 to 2047 and 24% of the total salary mass from 2048 to 2063. This reduction in the contribution rate is a result of a closed pension scheme, which does not have any new entrants and whose pensioner numbers will be decreasing once the active population affiliated to the COPS retires. It should be noted that the CoPS employees will have to keep contributing 9.5% of their salaries (current employee contribution rate to be updated in 2020 and every 5 years afterwards).

22. At the same time the other two schemes would need to receive separate contributions from both, the employees and the Organisation. The Organisation contribution rates would have to be 60% of the total contribution rate for the NPS (currently 9.3% for the employees and 13.95% of the NPS members’ salaries for the Organisation, to be updated in 2020 too) and 55% of the contribution rate of the TPS (currently 9.4% for the employees and 11.5% of the TPS members’ salaries for the Organisation, to be updated in 2020 as well).

23. By mingling the financing of the three schemes in a single Fund, part of the assets from the NPS and the TPS are being used to pay the benefits of the CoPS so that the three schemes can be sustainable with a certain combination of RTR - GCR, presently estimated to be 3.4% RTR and 37.85% of GCR.

24. If the current GCR remains unchanged (30.1%), the Fund would be depleted in 2048. Then, contributions would need to increase abruptly as from the year the PRF disappears to replace the return on PRF investments as a source of income. The estimated benefit payments are EUR 108 million for 2048 and EUR 105 million/year on average for the following 10 years (2049-2058). This would require an extra contribution of EUR 55 million in 2048, and EUR 51 million/year on average for 2049-2058 if the GCR remains at 30.1% instead of increasing to 37.85% as from January 2017. If the GCR increases to 37.85%, this would entail an average increase of EUR 12 million for the next 10 years but EUR 55 million less in 2048 and EUR 51 million/year less during 2049-2058. The required contribution would obviously be higher if increased later because it would not benefit from the return of financial markets.

CONCLUSION iv:

The Board concludes that sustainability - i.e. pension contributions plus the return of the Fund investments meet the pension benefits as they come due - is the appropriate objective of the Organisation for the PRF, as it is in line with the Statute and more favourable or less costly in financial terms.

The Board notes that - given the parameters impacting the life cycle of the Fund in 2017, including the reasonable target return assumption of 3.4% - the objective of sustainability can only be achieved with an increase of the GCR from 30.1% to 37.85%. The Board acknowledges the decision of the Committee of Ministers to gradually increase contributions in this respect over a four-year period, and that a mid-term review of the required level of funding will be presented in the last quarter of 2019.

B.2. EXTRA–FINANCIAL POLICY – SOCIALLY RESPONSIBLE INVESTMENTS (SRI)

Current situation

25. The current SRI Policy of the COE should apply to at least 90% of the investment portfolio and has two main principles:

· Exclusion (when possible) of companies belonging to the sectors of alcohol, tobacco, weapons, gambling and pornography. The exclusions are made on the grounds that those activities are considered potentially harmful for human dignity and health.

· Investments in companies are made on the basis of their current or their expected corporate social responsibilitymeasured in terms of their relation to environment, their social policy and their corporate governance. The criteria used to determine the social responsibility of companies in the three areas of analysis are based on the principles of the COE, which are mostly taken from the norms it has created or supports, and the Organisations’ activities and policies. Since the protection of human rights is one of the COE's fundamental values, priority to these criteria is given when possible.

Assessment

CONCLUSION v:

The Board considers the present SRI policy of the Fund to correctly translate the values of the COE into its investments. A full description of the principles of the COE applicable to the SRI criteria is presented in ANNEX I.

C. INVESTMENT STRATEGY

26. The investment strategy designs a portfolio of investments that fulfils the target of the investment policy in terms of risk and return. Two aspects are considered: financial and extra-financial (SRI).

C.1 FINANCIAL INVESTMENT STRATEGY

Background

27. The present Investment strategy of the PRF was approved by the Committee of Ministers, on proposal of the Management Board, in 2014[7] as one of the conclusions of the Second Three-Year Review. The Board proposed a portfolio of investments expected to generate a reasonable real return (5%) and able to fulfil the objectives of the investment policy, i.e. maximise the life of the Fund up to sustainability with a reasonable risk. Indeed at that time the reasonable target return allowed to meet the objective of sustainability with no changes in the GCR.

28. The design of the SAA or investment strategy is a process that entails three steps:

❶ Selection of the investment universe;

❷ Estimation of expected long-term return and investment risk or volatility;

❸ Composition of the optimal portfolio.

29. The current SAA was designed with the Modern Portfolio Theory (MPT), a model widely used to maximise portfolio return and minimise volatility by setting an optimal combination of several (preselected) assets. Global equity, emerging markets equity, EMU government bonds and euro corporate bonds were considered adequate for the PRF portfolio. The expected returns for assets were based on a forward-looking analysis while a combination of historical bullish (upward trend) and bearish (downward trend) periods defined the volatility.

30. The optimal SAA was set as follows: 35% of global equity, 15% of emerging markets equity, 10% of euro area equity, 24% of EMU government bonds and 16% of euro corporate bonds. This asset allocation was expected to deliver, in the long run, an average annual real return of 5%, with a volatility of 8.5%. In terms of implementation, the approach used was static, i.e. the Board aimed at respecting the strategic weights over time without modifying them to react to market conditions or expectations. Deviations to a certain extent only are permitted to allow for mechanical drift movements.

31. The strategy was implemented in January 2016. Therefore, the results (11.49% in real terms for the period February 2016 - March 2017) cannot yet be evaluated over the long term. Actually, with the updated hypotheses of long-term return and volatility the expected real return of the current strategy is 3.35% and volatility 8.68%.

Assessment

32. The Board, in its assessment of the investment strategy (and despite the relatively short implementation history of the current portfolio), wishes to reconsider the investment universe, the expected future returns of the asset classes and the optimal combination of assets that should generate a reasonable target return.

❶ Selection of the investment universe

33. Taking into account the objectives of the COE, the size of the PRF and its governance and administration framework, the investment universe considered appropriate for creating a well-diversified portfolio consists of the following asset classes^{^[8]}^,[9]:

· World equity;

· Emerging markets equity;

· Euro area equity;

· World government bonds (EUR hedged);

· Euro corporate bonds;

· Global direct real estate;

· Global direct infrastructure.

❷ Estimation of expected long-term return and volatility

34. To estimate the long-term expected returns of the selected asset classes, the Board used ex-ante forecasting financial models that were contrasted with the findings of several well-known asset managers: State Street Global Advisors (SSGA), J.P. Morgan AM, UBS Global AM, BlackRock and Goldman Sachs AM.

35. The volatility was estimated by the Board using (ex-post) historical figures.

RETURN - BOARD'S ESTIMATION

Return – Equity

36. The estimation of equity returns was done with an ex-ante approach based on the Gordon-Shapiro model, according to which the expected return of equity in a given country/region equals its dividend-yield plus the expected growth of dividends.

37. To estimate the dividend yield, historical figures were extrapolated. For the expected growth of dividends, the expected nominal GDP growth was used of the region the equity market is located at. Expected nominal GDP growth was in turn estimated with a time-weighted average of the OECD forecasts for real growth and inflation for the period 2018-2060.

Thus, the following formula gives the projected nominal equity return:

Equity return = historical dividend yield + forecast real GDP growth + forecast inflation

38. The Board decided not to apply the Gordon-Shapiro model to emerging markets, as the emerging economies (and their expected growth) are not fully reflected in the stock markets. Instead, it relied on the Board members' outlook on the expected return for this asset class.

39. The results provided by the Gordon-Shapiro formula for world and euro area equity, together with the outlook of the Board for the expected return of emerging markets equity are presented in Table VII below.

Table VII - Projected Equity Nominal Returns

		Return
World Equity		6.6%
Emerging Markets Equity		7.6%
Euro Area Equity		6.5%

Return - Fixed Income

40. The return of government bonds was projected to be equal to the long-term nominal growth of the issuers’ country or region. The application of this ex-ante approach, together with the long-term OECD forecasts for world GDP growth and inflation, and the estimation of the currency hedge impact calculated as the historical return difference between the hedged and the unhedged representative indices, gives the return assumptions for world government bonds EUR hedged.

41. Euro corporate bondsreturns were estimated by the Management Board that wished to incorporate a risk premium for corporate issuers (more risky) as regards governments. The estimations are presented in Table VIII below.

Table VIII - Projected Fixed-Income Nominal Returns

Return before EUR hedge

Impact of EUR hedge

Return after EUR hedge

World Government Bonds

(EUR hedged)

3.9%

-0.5%

3.4%

Euro Corporate Bonds

Return – Alternative Investments: Direct Real Estate and Direct Infrastructure

42. For direct real estate, the Board considered the target return (6%) of the most liquid investment vehicle found for global direct real estate (UBS (Lux) Real Estate Funds Selection - Global).

43. The expected return for global direct infrastructure was estimated to be 5%, i.e. in-between equity and fixed income.

44. The estimations of the returns for alternative investments are presented in Table IX below.

Table IX - Projected Alternative Investments Nominal Returns

		Return
Global Direct Real Estate		6%
Global Direct Infrastructure		5%

RETURN - ESTIMATIONS FROM OTHER SOURCES

45. The Board looked into the recently published long-term (10 years or longer) return assumptions from the following asset managers: SSGA, J.P. Morgan, UBS, BlackRock and Goldman Sachs. Table X below shows the projected returns for the different asset classes, taking into account the hedging impact as described above.

Table X - Projected Nominal Returns from Other Sources and Management Board

ASSET CLASS	SOURCE
ASSET CLASS	SSGA	J.P. Morgan	UBS	BlackRock	Goldman Sachs	Board
EQUITIES
World Equity	6.1%	6.1%	7.4%	6.4%	5.9%	6.6%
Emerging Markets Equity	8.6%	9.5%	8.7%	6.5%	8.2%	7.6%
Euro Area Equity	6.5%	7.4%	8.0%	7.2%	5.5%	6.5%
FIXED INCOME
World Government Bonds (EUR hedged)	0.8%	0.8%	3.7%	0.9%	1.0%	3.4%
Euro Area Corporate Bonds	1.3%	2.1%	4.2%	1.2%	1.4%	4.0%
ALTERNATIVE INVESTMENTS
Global Direct Real Estate	5.3%	5.3%	5.3%	4.3%	5.3%	6.0%
Global Direct Infrastructure	6.2%	6.2%	6.2%	6.8%	6.2%	5.0%

Note: figures in grey indicate that the asset manager did not provide a return assumption. In that case, J.P. Morgan’s projection was used instead to allow the computation of an expected return for all the portfolios.

Source: SSGA: Long-Term Asset Class Forecasts - December 2016; JP Morgan: Long-Term Capital Market Assumptions 2017; UBS: Quarterly Investment Strategy - First Quarter Outlook 2017; BlackRock: Capital Markets Assumptions, BlackRock Investment Institute, January 2017; Goldman Sachs: Global Portfolio Solutions - Strategic Long-Term Assumptions - December 2016.

46. The Board's estimations for equities are within the range of the appraisals made by other financial experts. Nonetheless, the Board's estimations are more optimistic for bonds and real estate, and more conservative for infrastructure. This difference is explained, for bonds, by the use of very long-term nominal GDP growth in developed markets to estimate the expected returns. In infrastructure and real estate, it is difficult to make comparisons between the estimates from different sources as the returns of an infrastructure or real estate investment depend more on the specific portfolio universe and manager skills than on the performance of the infrastructure or real estate considered as an asset class. In any case, these disparities were taken into account by setting limits to the SAA and by doing a sensitivity analysis of the SAA results to the different sources of assumptions.

VOLATILITY

47. Table XI and XII present the assumptions for volatility and correlations which were calculated by extrapolating the historical volatility of the benchmark indices representing the asset classes.

Table XI - Projected Volatility

*Asset Categories*	*Volatility*
World Equity	12.4%
Emerging Markets Equity	19.8%
Euro Area Equity	16.9%
World Government Bonds (EUR hedged)	2.9%
Euro Area Corporate Bonds	4.0%
Global Direct Real Estate	13.2%
Global Direct Infrastructure	15.0%

Source: ISRP

Table XII - Correlations Assumptions

Source: ISRP

❸ Composition of the optimal portfolio

OPTIMISATION WITHOUT CONSTRAINTS

48. The first exercise to create the optimal investment portfolio was done using the Markowitz modern portfolio theory, which combines pre-selected asset classes in such a way that the combined portfolio shows the lowest possible volatility for a given target return. The target returns considered are the following:

· return required to keep the Fund sustainable at the current GCR: 5.57%;

· return required to keep the Fund sustainable with a moderate increase of the GCR at 32.08%: 5% (this is also the current target return of the PRF);

· the reasonable target return: 3.4%

· reasonable and conservative return: 3.1%.

49. The Board constructed alternative optimal portfolios based upon its expectations. These portfolios are presented in Table XIII.

Table XIII - SAA - Optimisation without Limits

ASSET CLASS	Optimisation without limits - SAA by Target Return (Board's expectations)
ASSET CLASS	5.57%	5.00%	3.40%	3.10%
World Equity	30%	30%	40%	33%
Emerging Markets Equity	44%	44%	4%	3%
Euro Area Equity	0%	0%	0%	0%
TOTAL EQUITIES	74%	74%	44%	36%
World Government Bonds (EUR hedged)	0%	0%	29%	42%
Euro Area Corporate Bonds	0%	0%	0%	0%
TOTAL FIXED INCOME	0%	0%	29%	42%
Global Direct Real Estate	26%	26%	23%	18%
Global Direct Infrastructure	0%	0%	4%	3%
TOTAL ALTERNATIVE INVESTMENTS	26%	26%	27%	21%
TOTAL STRATEGIC ASSET ALLOCATION	100%	100%	100%	100%
*Expected Return*	*4.2%*	*4.2%*	3.4%	3.1%
*Volatility*	13.2%	13.2%	7.1%	5.8%

Source: ISRP

50. The Board notes that first, the target returns of 5.57% and 5% are unachievable, 4.2% being the maximum feasible target. Second, the results of the optimisation exercise were very extreme in general; therefore, it was decided to put limits to certain asset classes^{^[10]} with the aim of creating a better-balanced portfolio.

OPTIMISATION WITH CONSTRAINTS

51. To set such limits, the Board considered the financial target of the PRF, the SRI constraint, its views on investment risk, the size of the PRF, and the liquidity and availability of investment products.

52. As regards equity, a limit of 10% for emerging markets equity was retained, in line with peers' practices. This represents a reduction compared to the current SAA, where this class still represents 15%, since in the present analysis other sources of risk coming from other asset classes are included in the portfolio and therefore it was considered convenient to reduce the risk coming from emerging markets equity.

53. Concerning fixed income, the optimisation model favours government bonds. However, the Board considers that the diversification into corporate bonds is highly recommendable^{^[11]} and proposes a split allocation between government and corporate bonds in line with the aggregate market benchmarks,^{^[12]}i.e. two thirds and one third respectively.

54. Finally, the Board decided that, taking into account the characteristics of alternative investments - real estate and infrastructure -, an initial maximum of 5% of the global SAA should be allocated to each. This would allow to gradually gain experience with these new asset classes, as well as to limit asset manager risk and illiquidity risk while reflecting the high uncertainty about return and volatility assumptions.

55. The SAAs resulting from this exercise, putting limits to the different asset classes, are shown in Table XIV, XV and XVI below. The Board notes that, with the current return hypotheses and the limits applied to the SAA, all three, the current target and the target required to maintain sustainability while setting the GCR at 32.08 (5%), and the target required to maintain sustainability and keep the GCR unchanged (5.54%), are unachievable as the highest possible return is 4.0%.

Table XIV - SAA with Limits - Target Return (maximum achievable): 4.0%

ASSET CLASS	STRATEGIC ASSET ALLOCATION
World Equity	80%
Emerging Markets Equity	10%
Euro Area Equity	0%
TOTAL EQUITIES	90%
World Government Bonds (EUR hedged)	0%
Euro Area Corporate Bonds	0%
TOTAL FIXED INCOME	0%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%
*Volatility*	12.0%

Table XV - SAA with Limits - Target Return: 3.4%

ASSET CLASS	STRATEGIC ASSET ALLOCATION
World Equity	45%
Emerging Markets Equity	10%
Euro Area Equity	0%
TOTAL EQUITIES	55%
World Government Bonds (EUR hedged)	23%
Euro Area Corporate Bonds	12%
TOTAL FIXED INCOME	35%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%
*Volatility*	7.8%

Table XVI - SAA with Limits - Target Return: 3.1%

ASSET CLASS	STRATEGIC ASSET ALLOCATION
World Equity	32%
Emerging Markets Equity	10%
Euro Area Equity	0%
TOTAL EQUITIES	42%
World Government Bonds (EUR hedged)	32%
Euro Area Corporate Bonds	16%
TOTAL FIXED INCOME	48%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%
*Volatility*	6.2%

56. The Board made a sensitivity analysis of the SAA's expected returns to variations in the assumptions by calculating them using the asset classes' projections from the five sources previously mentioned. The results of this analysis are shown in Table XVII thru XIX below:

Table XVII - Real Return Projections of the SAA Targeting 4.0% with Different Return Assumptions

ASSET CLASS	SAA	SOURCE
ASSET CLASS	SAA	Board	SSGA	J.P. Morgan	UBS	BlackRock	Goldman Sachs
World Equity	80%	4.0%	3.7%	3.8%	4.8%	3.8%	3.6%
Emerging Markets Equity	10%
Euro Area Equity	0%
TOTAL EQUITIES	90%
World Government Bonds (EUR hedged)	0%
Euro Area Corporate Bonds	0%
TOTAL FIXED INCOME	0%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%

57. Based upon the different sources of estimation of the returns considered, the projected real returns of the SAA targeting 4.0% may vary between 3.6% in the worst case scenario (of Goldman Sachs), to 4.8% in the best case settings (of UBS).

Table XVIII - Real Return Projections of the SAA Targeting 3.4% with Different Return Assumptions

ASSET CLASS	SAA	SOURCE
ASSET CLASS	SAA	Board	SSGA	J.P. Morgan	UBS	BlackRock	Goldman Sachs
World Equity	45%	3.4%	2.4%	2.6%	4.0%	2.3%	2.3%
Emerging Markets Equity	10%
Euro Area Equity	0%
TOTAL EQUITIES	55%
World Government Bonds (EUR hedged)	23%
Euro Area Corporate Bonds	12%
TOTAL FIXED INCOME	35%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%

58. Based upon the different sources of estimation of the returns considered, the projected real returns of the SAA targeting 3.4% may vary between 2.3% in the worst case scenario (of Goldman Sachs), to 4.0% in the best case setting (of UBS).

Table XIX - Real Return Projections of the SAA Targeting 3.1% with Different Return Assumptions

ASSET CLASS	SAA	SOURCE
ASSET CLASS	SAA	Board	SSGA	J.P. Morgan	UBS	BlackRock	Goldman Sachs
World Equity	32%	3.1%	1.8%	2.0%	3.6%	1.7%	1.8%
Emerging Markets Equity	10%
Euro Area Equity	0%
TOTAL EQUITIES	42%
World Government Bonds (EUR hedged)	32%
Euro Area Corporate Bonds	16%
TOTAL FIXED INCOME	48%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%

59. Based upon the different sources of estimation of returns considered, the projected real returns of the SAA targeting 3.1% may vary between 1.7% in the worst case scenario (of BlackRock), to 3.6% in the best case setting (of UBS).

CONCLUSION vi:

The Board concludes that both 3.4% and 3.1% are associated with portfolios that are reasonably well diversified and could be considered reasonable target returns. The 3.1% could be qualified as more conservative, as its volatility is lower.

RISK IMPLICATIONS OF THE PRF INVESTMENT STRATEGY

60. This chapter first describes the risk framework specific to the PRF, i.e. the main risks the Fund is exposed to: depletion risk and shortfall risk (including market risk or volatility, relative risk, liquidity risk, counterparty risk and operational risk). The conclusion from this appraisal will be that the highest risk in terms of exposure (and under the control of the Management Board) comes from fluctuations in the market value of the investments, i.e. the market risk or volatility. Second, an overview will be given of the market risk or volatility associated to the SAAs proposed for the different target levels, and how this volatility influences the objectives of the COE for the PRF.

Risk framework

61. The Board believes^{^[13]} that the objective of the COE for the PRF investments is to produce a return that makes the Fund last as long as possible (for a certain level of contributions^{^[14]} and benefits). Therefore, when the PRF's objective is Sustainability (the Fund can pay pension benefits as they come due^{^[15]}) the primary risk faced by the PRF is its potential depletion.

62. To monitor the depletion risk, actuarial assessments of the life cycle of the PRF are carried out during periodical mid-term reviews in order to spot divergences between the expected and the actual Fund’s life cycle. It is at that stage that the COE can undertake corrections in the SAA or in the other parameters impacting the Fund's life cycle, should this be necessary.

63. Two elements intervene in the risk of depletion: the risk related to the assets not progressing adequately and the risk of the pension payments evolving unfavourably, i.e. pension benefits payments being higher than expected.

64. The risk linked to pension benefits payments is subject to the longevity risk, structural changes in the workforce, wage and inflation risk, whilst the assets are exposed to market risk or volatility, relative risk, liquidity risk, counterparty risk and operational risk.

65. The risk linked to inflation can be limited by the assets having a real target of return. Moreover, the longevity risk of the pension payments could in theory be hedged with the assets through longevity swaps: a derivative contract that offsets the risk of pension scheme members living longer than expected. In practice, this still remains a narrow market and important costs are associated to such deals.

66. With regard to assets’ risks, there is a brief description below of their nature and of the tools the Management Board has to monitor and reduce them. They all lead to shortfall risk: the risk of falling short of the investment performance target.

· Market risk or volatility: this is the risk of decrease in the value of the portfolio due to adverse movements in financial markets. The PRF needs to bear some market risk in order to reach its target return; market risk can be reduced through portfolio diversification and monitored through different measures. Market risk has several sources, impacting different parts of the portfolio:

ü Interest rate and credit risk - in bonds.

ü Currency risk - in any non-domestic investment.

ü Company or business risk - in equity and corporate bonds.

ü Political/regulatory risk - in emerging markets equity and infrastructure.

ü Economic risk - in equity and bonds (alternative investments are supposed to be more resilient to economic shocks).

· Relative risk: this is the risk of deviation from the benchmark’s performance – representing the target of return – due to the portfolio’s mechanical drift as asset classes have different growth patterns, to tactical decisions related to the asset allocation, or to active management within asset classes. The relative risk stemming from the first two factors can be controlled through the rebalancing strategy and by keeping the investment portfolio close to the long-term strategic asset allocation. With regard to the latter, it can be limited by investing in index-linked products, by imposing certain deviation limits to active asset managers and by monitoring them closely.

· Liquidity risk: the risk of big losses when liquidating positions or when there is no possibility of liquidation at all. As a long-term investor the PRF can afford some liquidity risk, as it has annual net inflows and its eventual outflows can be predicted in both, timing and amount. This risk is mitigated with the selection of liquid asset classes for the biggest share of the portfolio and by limiting the share of less liquid asset classes like real estate and infrastructure. In parallel, as the investment vehicles are mutual funds, they should be easy to redeem.

· Counterparty risk: this is the risk that a counterparty be incapable of honouring its obligations. The nature of the PRF’s long-term investment vehicles – mutual funds – does not entail a direct counterparty risk as the assets are always under the ownership of the Organisation. In the event of an asset manager bankruptcy, the PRF’s ownership of investments will not be affected.^{^[16]} Within the funds, for some asset classes counterparty risk exists, especially within fixed-income which is managed by the asset managers, and it translates into the PRF under the form of market risk.

· Operational risk: the risk of loss resulting from inadequate or failed internal processes, people and systems, or from external events. This risk could exist on the one hand, within the administration of the PRF and, on the other hand, within the mutual funds at the level of the asset managers. To reduce this risk, the ISRP has implemented procedures and control mechanisms to optimise the execution and administration of the investments. With regard to operational risks inside the mutual funds, the latter is analysed at the time of the asset managers’ selection, the solidity and the risk control procedures of the candidates being heavily weighted elements of the evaluation.

Very short-term market risk, measured on the basis of financial markets’ daily volatility, is a highly disturbing factor for rebalancing operations and investment of incoming contributions. This risk, even though it comes from market fluctuations, is very much related to operational risk. Indeed, large movements in securities prices during volatile periods make portfolio valuations and corrections through rebalancing very erratic. Moreover, exit and entry prices are more uncertain during such periods than under normal conditions and largely perturb both the rebalancing and investment executions. Certain levels of market instability can be monitored, eventually leading to the decision to modify the policies related to rebalancing operations and/or the investment of contributions.

67. Some pension fund managers define the main risk of pension funds like the PRF as the risk to lose money, whereas the risk of fluctuations in value or volatility should be considered irrelevant due to the long-term investment horizon.^{^[17]} Within this approach, the liquidity, counterparty and operational risks are considered the highest, as they lead to actual losses.

68. From the point of view of the compensation the PRF obtains for bearing risk, liquidity, market and counterparty risk are all rewarded while relative risk^{^[18]} and operational risk are not.

69. Finally, the PRF's biggest exposure is to market risk or volatility as it is the most challenging to reduce; in parallel, it is the only risk having a budget, i.e. market risk is explicitly allowed by the investment strategy approved by the Committee of Ministers.

Market risk – Volatility

70. The “risk-free option” for the COE as regards the coverage of its pension obligations would be to make the Fund grow now with contributions and invest it in assets with a guaranteed notional value at maturity, producing a stream of income that matches the benefits payments and having the same sensitivity as the liabilities to factors such as inflation or population/demographic changes.

71. However, it is the Board's understanding that the COE wishes to partially finance pension obligations with investment income (capital gains). This entails that the PRF needs to invest in assets generating the required rate of return to make the Fund grow and eventually reach the long-term objective of sustainability. These investments will carry a certain risk or price volatility and their evolution will not necessarily be linked to that of the benefits to be paid. This implies that the future sustainability is not 100% certain. In parallel, investing in financial markets implies a certain short-term volatility in the value of the assets. The COE trades the certainty of the need to increase contributions now and invest in the risk-free portfolio, for the probability of having to increase them in the future only if investment returns are lower than the target.

72. The Board has examined the level of market risk or volatility, both in terms of variations of the long-term objective (future sustainability) and of the PRF short-term market value and variations in the sustainability situation, to estimate what could be a reasonable return. Apart from this quantitative analysis, qualitative or fundamental views have been used to determine which of the SAAs producing 4.0%, 3.4% or 3.1% were the most reasonably diversified, and which reasonable target return should be set up accordingly.

73. It is challenging to predict what will happen in the future, especially with such a long-term horizon as for the targeted life of the PRF. Even though the expected rate of return is the most probable average result in the long term, there is a certain probability of returns being higher (and the PRF growing beyond its target) and a certain probability of returns being lower (and the sustainability not being possible). In parallel, it is likely to observe a very bad return one year. Even though the PRF can currently sustain negative fluctuations of its market value^{^[19]} - as losses do not need to be realised to pay pensions from (until 2030) -, the COE should understand the possible results linked to approving a reasonable return.

74. The actual return can turn out lower than the target or expected return^{^[20]} under two scenarios, one short-term and another long-term.

Scenario A – Short Term

75. There could be a short-term market crisis one year (or over a short period of time) involving a significant decrease in the value of the PRF. As a consequence, for that period, the PRF would have lost its sustainability status (in the long term), as it is calculated with the current level of assets of the PRF, which would have decreased as regards the target, while the assumption of expected return is kept at the original level. The probability of the SAA targeting a 4.0% net real return having a return below -5%, for example, in one year is 23%.^{^[21]} This probability decreases, of course, when the target return is lower: 14% for a SAA with a target of 3.4%^{^[22]} and 10% when the target is 3.1%^{^[23]}. At the same time, lower target returns (in the future) imply higher contributions now.

76. The return and probability scenarios described above would lead to a temporary lack of sustainability. Nonetheless, this would be temporary because 3.4% (or any other target return) is considered to be the long-term average return, which implies that lower returns will be compensated with higher ones, in order to reach an average of 3.4%. This has proven to be empirically correct as shown in Graph I below, depicting the long-term evolution of US stock market returns represented by the Standard & Poor’s 500 index.

Graph I - Evolution of the Standard&Poor’s 500 returns 1963 - 2017

Source: Thomson Reuters Datastream

77. Graph I shows that the market experienced severe short-term downturns such as in 1973-74 (oil crisis: -42% accumulated return in two years), 1987 (“Black Monday”: -24% in two months), 2000-2002 (explosion of the technological bubble: -46% accumulated return in two years) and 2008 (Financial crisis: -38% return in one year). Despite these declines, the long-term annual average return for the whole period was +6.6%.

78. The Board wishes to note that until the year 2030, the return of the Fund will not be used to pay pensions; therefore, short-term returns lower than the target or expected return or even negative returns can be borne, as losses do not need to be realised to pay pensions and because short-term returns above 3.4% (or any other target) will counterbalance or compensate temporary lower returns.^{^[24]} However, as from 2030, the cost of paying pension benefits might surpass the combined contributions to the Fund, meaning that the Fund return would be used to pay benefits. In this event, the investment strategy should be adapted to reduce the volatility of returns (negative returns should be avoided) and to include cash and yield-generating investments to pay pensions, hence lowering the expected Fund return.

Scenario B – Long Term

79. The long-term risk or worst-case scenario in the long term would be that the actual long-term return in the future be lower than the target. In that case, the sustainability would not be reached because the investment strategy would have generated a long-term return lower than assumed (as depicted in Graph III). With the life cycle depicted below in Graph II,^{^[25]} any long-term return below 3.4% leads to a depletion of the Fund.

Graph II - PRF Life Cycle as of End 2016
3.4% Long-Term Return and GCR 37.85%

Source: ISRP

Graph III - PRF Life Cycle as of End 2016
1% Long-Term Return and GCR 37.85%

Source: ISRP

80. To calculate the probabilities in the long term, the Board has simulated - with a stochastic application - how the SAA with target return 3.4% could behave over time. The Board examined the probabilities of the returns being lower than the target knowing that any long-term return below the target would lead to the worst-case scenario of depletion, and that the probability of this happening would be around 50%. In parallel, there are high chances (50%) of the return being higher and of the COE improving the funding of its schemes with the investment return. These results are given by the assumption of returns following a normally distributed pattern, which is often found in stock market analysis. The results of the simulation are presented in the Table XX and Graph IV below.

Table XX - Long-Term Volatility (Probabilities of Annual Real Returns over time)
SAA with target return 3.4% (Volatility 7.8%)

Annual Real Return for the Period in %		Probabilities of the Annual Real Return
Annual Real Return for the Period in %		95% probability of being lower than:	50% probability of being lower than:	5% probability of being lower than:
Time Period	5 years	9.3	3.4	-2.0
	15 years	6.8	3.4	0.3
	30 years	5.8	3.4	1.2

GraphI IV - Evolution of returns over the long term
SAA with target return 3.4% (Volatility 7.8%)

81. The situation described in Graph IV should not occur if the Fund status and its investment strategy are reviewed regularly on its way to the long term, and adjustments are made whenever needed. The lag in return can be analysed during these reviews to determine if it is a short-term situation that can be compensated by a higher expected market return or if the target return (and associated investment strategy) should be modified.

CONCLUSION vii:

In the Board's view, the reasonable target return in terms of portfolio diversification should be 3.4%. Likewise, given its current cash flow profile and the planned Reviews (Adjustments) the Fund shall go through, the PRF is able to bear the associated volatility of an investment strategy aiming at 3.4% - both in the long term and short term. The Committee of Ministers should note that with 3.4%, the COE would still be partially financing pensions with investment risk.

The Board would like to propose to the Committee of Ministers for approval the following SAA necessary to achieve a 3.4% real return in the long term:

TABLE XXI - Proposal for SAA - Target Return: 3.4%

ASSET CLASS	STRATEGIC ASSET ALLOCATION
World Equity	45%
Emerging Markets Equity	10%
TOTAL EQUITIES	55%
World Government Bonds (EUR hedged)	23%
Euro Area Corporate Bonds	12%
TOTAL FIXED INCOME	35%
Global Direct Real Estate	5%
Global Direct Infrastructure	5%
TOTAL ALTERNATIVE INVESTMENTS	10%
TOTAL STRATEGIC ASSET ALLOCATION	100%
*Volatility*	7.8%

Given the long-term horizon for which the investment strategy has been created, the extreme difficulty of generating higher returns when reacting to changing market conditions and the higher volatility and costs involved by a dynamic implementation, the asset allocation will continue to be implemented in a static way, i.e. the Board would faithfully implement it over time, with no consideration to short-term market expectations or fluctuations. Only in periods of exceptional circumstances may the Board decide to diverge from the Strategic Asset Allocation.

This SAA remains valid until the next review (the next intermediary assessment or the Fourth Four-Year Review).

C.2. EXTRA-FINANCIAL (SRI) INVESTMENT STRATEGY

82. The extra-financial strategy defines the best way the SRI policy of the COE can be implemented to the investment portfolio. It takes into account the SRI policy of the PRF, the financial characteristics of the Fund, and the availability of SRI products.

Current situation

83. The present SRI strategy has the following main axes:

· The current SRI Policy of the COE should apply to at least 90% of the investment portfolio.

· Exclusion, when possible, of companies with more than 5% of their activities involving alcohol, tobacco, weapons, pornography and gambling and the selection of companies should preferably be made with a 'best in class' approach. For those asset classes for which this is not possible, the strategy shall rely on an Engagement Overlay, consisting of running a portfolio with traditional financial criteria while companies in the portfolio are eventually approached by an external SRI expert with the aim of carrying out a process of dialogue and negotiation in order to convince them to improve their corporate behaviour in terms of Environmental, Social, and Governance criteria (as expressed in the SRI policy of the PRF).

· In the evaluation of companies, the asset manager should use his own environmental, social and governance criteria, which will be examined by the Management Board in order to assess their compliance with COE's SRI principles.

· Government bonds: When investing in euro denominated bonds issued by euro area countries, each of them being member of the COE and thus respecting the values of the Organisation, no specific SRI criteria will be applied.

· Real estate: Investments in real estate may be placed in traditional (non-SRI) funds.

Assessment

84. The Board considers that the current SRI strategy is in line with the COE’s objectives of the investment policy, with the financial constraints of the Fund and with the present development of the SRI industry.

85. The Board has looked at the application of the SRI policy to the new asset classes included in the proposed SAA and considered that in Global government bonds, all non-member countries of the COE represented in the global government bond universe - except for Australia - are observers of the Organisation.^{^[26]}Countries granted observer status by the Committee of Ministers, accept its guiding principles of democracy, the rule of law, human rights and fundamental freedoms.^{^[27]} Therefore the debt they issue might be considered compliant with the main principles of the SRI policy of the PRF and no specific SRI filter should apply to this part of the portfolio.^{^[28]}

86. In infrastructure, the Board shall ideally look for funds or portfolio solutions respecting certain environmental, social and corporate governance criteria. Nonetheless, if such investment products do not exist and in order for the PRF to be able to benefit from the diversifying impact of infrastructure investments, the share of this asset class in the portfolio (currently limited to 5%) could be included within the 10% of non-SRI allowance.

CONCLUSION viii:

The Board proposes to the Committee of Ministers for approval to expand the current SRI strategy of the PRF to make it relevant to the other asset classes included in the SAA proposal. The updated SRI strategy would read as follows:

· The current SRI Policy of the COE should apply to at least 90% of the investment portfolio.

· Exclusion, when possible, of companies with more than 5% of their activities involving alcohol, tobacco, weapons, pornography and gambling and the selection of companies should preferably be made with a 'best in class' approach. For those asset classes for which this is not possible, the strategy shall rely on Engagement Overlay, consisting of running a portfolio with traditional financial criteria while companies in the portfolio are eventually approached by an external SRI expert with the aim of carrying out a process of dialogue and negotiation in order to convince them to improve their corporate behaviour in terms of Environmental, Social, and Governance criteria (as expressed in the SRI policy of the PRF).

· Government bonds: When investing in government bonds from global developed countries or EMU countries, as most of these issuers are members or observers of the COE and thus respecting the values of the Organisation, no specific SRI criteria will be applied.

· Alternative investments (real estate and infrastructure): Investments in real estate and infrastructure may be placed in traditional (non-SRI) investment vehicles as long as combined they represent less than 10% of the portfolio.

D. INVESTMENT STRUCTURE

BACKGROUND

87. The present investment structure of the PRF is as follows:

Table XXII - Investment Structure of the PRF

ASSET CLASS

(share and rebalancing limits)

BENCHMARK

MANAGEMENT STYLE

INVESTMENT VEHICLE

ASSET MANAGER

World equity

(35% +/-5%)

MSCI World

(Net Return)

Index-linked

(traditional) + SRI Engagement Overlay

Mutual fund:

iShares Developed World Index Fund

Engagement Overlay Provider: BMO

BlackRock

Emerging markets equity

(15% +/-3%)

MSCI Emerging markets

(Net Return)

Index-linked

(traditional) + SRI Engagement Overlay

Mutual fund:

Vanguard Emerging Markets Equity Stock Index Fund

Engagement Overlay Provider: BMO

Vanguard

Euro area equity

(10% +/-2%)

MSCI EMU

(Net Return)

Active (SRI)

+/- 50% Mutual fund:

Amundi Actions Euro ISR

Amundi Asset Management

+/- 50% Mutual fund:

Allianz Valeurs Durables

Allianz Global Investors

Euro area government bonds

(24% +/-5%)

Citigroup EMU Government Bonds Index

(Total Return)

Index-linked

(traditional)

Mutual fund:

SSgA EMU Government Bond Index Fund

State Street Global Advisors

Euro corporate bonds

(16% +/-5%)

Markit iBoxx Euro Corporates

(Total Return)

Active (SRI)

Mutual fund:

BNP Parvest Sustainable Bond Euro Corporate

BNP Paribas Asset Management

88. During the period under review, 2013-2016, the following changes to the investment structure have been implemented:

· The euro area equity allocation of the SAA was reduced from 55% to 10% of the PRF in January 2016 to comply with the SAA approved in the Second Three-Year Review, approved by the Committee of Ministers in November 2014. This share of the portfolio remained equally split between the two actively-managed SRI funds already in place: Allianz Valeurs Durables and Amundi Actions Euro ISR.

· 35% of the PRF was invested in world equity through the index-linked fundBlackRock Developed World Index Sub Fund, also in January 2016. In July 2017, this fund was renamed iShares Developed World Index Fund.

· Also to comply with the new SAA, 15% was invested in emerging markets equity through the index-linked fund Vanguard Emerging Markets Equity Stock Index Fund.

· The euro area listed real estate allocation, which represented 5% of the PRF, was entirely disinvested. The total allocation in the actively managed fundAllianz Foncier was therefore sold in January 2016, along with the above changes.

· The Committee of Ministers approved in November 2014 that, for those asset classes for which the offer of SRI products did not allow investment with a best-in-class approach, the prescribed SRI strategy would be Engagement Overlay on an index-linked portfolio. This Engagement Overlay strategy was to be applied on the investments in world and emerging markets equity, totalling 50% of the PRF. After the market research on Engagement Overlay service providers conducted by the Secretariat, three companies where invited to send in a proposal: F&C Investments,[29] Global Engagement Services (GES) and Hermes EOS. The Board analysed the offers received through a quantitative evaluation grid taking into account the following criteria: company, experience, service offered, engagement team, fees, reporting and client relationship. This analysis ranked F&C first, Hermes EOS second and GES third. Based on this result the Board decided to select F&C overlay at a cost of EUR 50 000 per year.

· For the fixed-income part of the portfolio, the allocation of 24% to EMU government bonds and 16% to euro corporate bonds was left unchanged. The EMU government bonds bucket remained solely invested through the index-linked traditional fundSSgA EMU Government Bond Index Fund. The euro corporate bonds part was modified as it was originally equally split between the two actively-managed SRI funds BNP Parvest Sustainable Bond Euro Corporate andAmundi Crédit Euro ISR. In November 2014, Amundi Crédit Euro ISR was entirely sold as it underperformed its benchmark and peer group since the start of investments in 2012 due to its overly conservative investment policy. The proceeds were reinvested in BNP Parvest Sustainable Bond Euro Corporate, which then became the only investment vehicle for this asset class in the PRF.

· Additionally, the Board noted that the benchmark index for the euro corporate bond allocation differed from the index used to build the assumptions of the proposed SAA. Therefore, the Board proposed to the Committee of Ministers to approve a change in benchmark index from the Markit iBoxx Euro Corporates 3-5 years index to the Markit iBoxx Euro Corporates All Maturities index. This change became effective as of 1 February 2016.

Management style

89. To improve the volatility-return profile of the PRF, the SRI strategy was modified during the Second Three-Year Review. The new SRI strategy provided a double approach: the use of the existing SRI strategy (best in class) for those asset classes for which the current offer of SRI products allowed to do so. For the asset classes for which this was not possible, an Engagement Overlay strategy was initiated. This strategy consists of running a portfolio with traditional financial criteria while an external SRI expert approaches the companies in the portfolio to carry out a process of dialogue and negotiation so as to convince them to improve their corporate behaviour in terms of Environmental, Social and Governance criteria. This solution was considered optimal in terms of risk as it allowed a further portfolio diversification into global and emerging markets equity, reducing the market risk; moreover, it diminished the asset manager risk as it was implemented through passively managed index-linked funds. In parallel, it allowed to respect reasonably well the reputation of the COE as it reflected the effort to soundly manage the PRF in accordance with the COE's principles in terms of social responsibility to the extent permitted by the current development of the SRI industry. Finally, costs remained close to those of the previous strategy with a decreasing trend for the future.

90. Consequently, as shown in Table XXII above, the major part of the PRF (74%) is now managed through index-linked funds: world equity and emerging markets equity (50% of the portfolio) are passively managed and covered by an Engagement Overlay. EMU government bonds (24% of the portfolio) are also passively managed but without SRI criteria given the nature of the issuers. As these markets are heavily traded and efficient, these investments followed an index-linked strategy, thus reducing management fees and relative risk.

91. In view of their SRI component, it was originally decided to actively manage euro area equity and euro corporate bonds. The main reason for this was the lack of eligible index-linked funds tracking SRI indices while there was an appropriate offer of SRI active funds. In any case, tracking an SRI index implies accepting some relative risk as the PRF will not be exposed to the market as a whole, as with traditional index-linked management, but only to the universe of the SRI index.

92. Compared to index-linked management, where asset managers replicate a market index, active management aims at over-performing the index. More resources are therefore required (e.g. constant monitoring), inducing higher management fees and transaction costs. Active management can involve higher risk-taking and represents an extra source of volatility for the portfolio, since it can lead to lower than the benchmark returns (on which the strategic asset allocation is based). Moreover, active management has been known to show poor results in heavily researched, intensively traded and highly correlated markets, where the possibilities of consistently beating the benchmark are more limited.

93. To limit the active management risk in the euro area equity and euro corporate bond parts, two mutual funds have been chosen for each asset class, thus diversifying the asset manager risk and reducing the relative risk. Nevertheless, as explained above, one of the euro corporate bond funds was fully redeemed in 2014 due to its inadequate investment policy. It was not replaced as the Board considered that the strategy of the remaining fund already involved low relative risk.

Investment vehicles

94. The Board decided to invest using mutual funds. Mutual funds are simple to trade and account for, which represents an advantage for the operation of the Fund and the integration of the PRF accounting into that of the Organisation. In addition, mutual funds benefit from a sound legal framework providing transparency and control.

Asset managers

95. The assets of the PRF are managed by BlackRock (world equity allocation); Vanguard (emerging markets equity allocation); Amundi Asset Management (50% of the euro area equity allocation); Allianz Global Investors (50% of the euro area equity allocation); State Street Global Advisors (euro area government bonds allocation) and BNP Paribas Asset Management (euro corporate bonds allocation).

96. All asset managers have been selected following a competitive selection process carried out by the ISRP. For government bonds, the Board decided in 2007 to invest in a mutual fund initially selected by a consultant for the Pension Budget and Reserve Fund (PBRF) of the OECD. In 2011, the Board directly selected the euro area equity and euro corporate bonds funds, and, in 2015, the world equity and emerging markets equity funds. The fees of the funds the PRF invests in are very attractive. Whenever possible, the ISRP tries to negotiate fees with each asset manager that apply to all the Funds under its administration as well as across the asset manager's mutual funds invested in, as in general fees decrease with the size of the investment.

97. Table XXIII presents a comparison of PRF investments (in terms of returns, volatilities and costs) with their benchmarks and with a peer group during the period January 2014 - June 2017.[30] The peer group is composed of similar funds (usually the competitors from the original Request for Proposal of the fund the PRF is invested in) and the entire Morningstar category, which encompasses both active and passive funds, as well as SRI and non-SRI funds.

Table XXIII - Comparison of PRF’s Investments with Benchmark and Peer Group

Source: Morningstar Direct (categories of open-end funds in the Europe/Asia/Africa universe) and asset managers

Note (1): Some share classes have performance data calculated prior to their inception date, in accordance with Morningstar’s Extended Performance Methodology. This is based upon a simulated/extended track record, using the track record of an older share class or a previous fund following the same strategy.

Note (2): Management costs and fees are net of rebates for those funds in which the PRF is invested. For the other funds, figures shown are either provided in the original Request for Proposal or by Morningstar, whichever is the lowest. Morningstar figures are used for Morningstar categories and funds that were not included in the original RfP (i.e. HSBC Sustainable Euroland Equity, iShares Eur GovtBdIdx, Vanguard EUR Gov Bond Index and Allianz Euro Credit SRI).

Note (3): Following its acquisition by New York Life Investments on 3 February 2014, Dexia Asset Management changed its name to become Candriam.

98. In order to analyse the risks taken by active asset managers and the results obtained, Table XXIV below compares the volatility of each fund with the volatility of its benchmark and its peer group, as measured over the last three years. The volatility comparison indicates whether the asset manager takes excessive risk or not compared to the market risk as expressed by the benchmark’s or peer group’s volatility. To determine if the risk taken translates into return, the Sharpe ratio[31] of each investment is compared to the Sharpe ratio of its benchmark and its peer group, which allows comparing how well the risks taken by asset managers are rewarded versus the risk reward of the market and of the competitors.

Table XXIV - Comparison of Volatility and Sharpe Ratio of PRF's Active Managers

Source: Morningstar Direct (categories of open-end funds in the Europe/Asia/Africa universe)

99. In order to analyse the risks taken by passive asset managers and the results obtained, Table XXV below shows the tracking error of each fund. The tracking error is the standard deviation of the difference between the returns of an investment and its benchmark. For a passive fund, the tracking error should be minimal.

Table XXV - PRF's Index-Linked Mutual Funds' Tracking Error

Source: Morningstar Direct and asset managers [BlackRock for iShares Dev Wld idx (IE) Flex Acc EUR (world equity fund)].

ENGAGEMENT OVERLAY SERVICE PROVIDER - BMO GAM

100. BMO GAM engages with companies in the investment universe according to one of the following three approaches:

· on a proactive basis, i.e. it contacts companies with perceived ESG weaknesses, defined as priority companies;

· on a reactive basis, i.e. it contacts companies after incidents have occurred;

· on a thematic basis, i.e. it engages with companies on a sector or regional level focussing on specific ESG themes.

101. In addition, BMO GAM also engages with policy makers and regulators on ESG topics.

102. In practice, BMO GAM prepares an annual priority list of companies to engage with, based upon the risk scoring of each company as defined by their internal ESG-risk tool, and the most material financial holdings within their clients’ portfolios. During the year, other companies can be added to this list, in case reactive engagement is urgently needed. For each company on the list, BMO GAM sets targeted outcomes it wants to reach with the company in terms of improved ESG-practices.

103. The engagement actions undertaken by BMO GAM cover most of the topics described in the SRI Policy of the Council of Europe, most of them being connected to conventions and international standards originating from the United Nations and the International Labour Organisation.

104. BMO GAM engages with companies using different methods, including in-person and telephone meetings, written correspondence and emails. Such contacts can be established at different levels within the company, including Board level, C-level and with operational specialists. While BMO GAM typically engages on a one-to-one basis, it may decide to participate in group meetings or engage including other stakeholders, if this could lead to a better outcome.

105. Engagement procedures with companies can have different time horizons. In BMO GAM’s experience, an average period of two to three years is needed to either reach a positive outcome or to come to the conclusion that the engagement has failed.

106. To follow up on the process and to determine its success, BMO GAM sets milestones for every engagement started. A milestone is reached when a pre-defined change in behaviour (the behaviour BMO GAM is asking for) is adopted by the company. At the same time, each topic consists of multiple intermediate milestones allowing to measure progress towards the final outcome.

REVIEW OF THE INVESTMENT STRUCTURE – ASSESSMENT

Management style

107. The active management style shows good results in the SRI euro area equity allocation. For the period under review, the gross investment results of the PRF funds are significantly above those of the benchmark. At the same time, the managers do not take excessive relative risks and the risks taken are well rewarded, as expressed by the comparison of their Sharpe ratios to those of their benchmark. For corporate bonds, the BNP Parvest fund performed better than its benchmark, hence doing better than the passive managers.

108. Passive managementin the world and emerging markets equity and government bonds allocations has shown results in line with what could be expected, i.e. the results and risks taken closely track those of the benchmark index. The gross investment results over the last three years are in line with the benchmark for government bonds, and above it for the world and emerging markets equity funds thanks to the securities lending activity and a favourable tax treatment of the fund.

CONCLUSION ix:

The Board considers that, given the current investment strategy of the PRF, each asset class is managed following the most appropriate management style.

Investment vehicles

CONCLUSION x:

As regards the investment vehicle, taking into account the sound legal framework of mutual funds, their simplicity in terms of accounting and administration, their higher external control and the moderate costs, the Board considers that mutual funds are best suited for the investments of the PRF in traditional asset classes. For the new alternative asset classes, other investment vehicles such as funds owned by limited partnerships and trusts will be considered.

Asset managers

109. Asset managershave been evaluated in terms of client service, performance and costs. Performance figures are compared to those of the benchmark and the peer group, and costs were compared to those of their closest peer group. For the period under review, the gross investment results of the PRF funds are significantly above those of the benchmark and the peer group, except HSBC.

110. With regard to client service, all managers have proved to be very service minded, professional and reactive. There have not been significant changes in the management teams of the funds during the period under review. A change occurred at BMO Global Asset Management, where in 2017 Matthias Beer became the new contact for the Board, under supervision of the former contact, Vicki Bakhshi, Head of Governance & Sustainable Investment (GSI).

111. SSGA, the manager of the EMU government bond fund, has gone through some structural changes in recent years described below.

112. The SSGA French-domiciled funds merged on 28 September 2015 into an SSGA Luxembourg SICAV; the investment process, fund philosophy and track record were not affected by this merge.

113. Moreover, the Board noted the decision from SSGA, communicated end 2016, to consolidate all of their existing portfolio management activities in London. The London office is SSGA's main operating entity in Europe with over 350 staff, including 120 investment professionals. The team of relationship managers and client account managers remained in France. The transition was completed by the end of April 2017. The location of the portfolio management team relative to where the underlying assets are invested has no consequences, as the same processes, systems (e.g. portfolio management, accounting, corporate actions, and order management), philosophy, and regional trading desks remain in place.

114. In terms of net performance (after management fees and costs), the active SRI equity managers performed better than their benchmark and peer group except for the fund from HSBC which performed better. Compared to the other fund in the PRF, asset manager Allianz shows the highest relative return for the period under review, after a slight underperformance during its first two years of investment.

115. Since the start of the investments in June 2012, the BNP Parvest fund shows results above those of the benchmark and its peer group over the same period.

116. The passive equity managers performed in line with their benchmark. The net performance for the BlackRock world equity fund is still above its benchmark after taking into account the fees. For emerging markets equity, the Vanguard fund posted a net performance below its benchmark due to high trading and custody fees in this universe. In order to mitigate these costs, the fund employs a so-called “optimised” management process, giving a higher weight tolerance for each line in the portfolio compared to a strict physical replication, which is more common for developed markets funds. Despite this optimisation, the fund failed to replicate the return of its benchmark on a net-of-fees basis.

117. The government bonds fund kept on track with the benchmark as expected in view of its management style.

118. The cost^{^[32]} comparison is complex, as the peer group shows a large dispersion and the fees reported by the peer group managers are not negotiated (fees could significantly decrease after rebates, as is the case with the present PRF funds). Indeed, as shown in Table XXVI below, the Secretariat negotiated the following fees with the different asset managers (figures at end June 2017 versus the standard fee for institutional clients; in percentage of asset under management):

Table XXVI - PRF's Mutual Funds' Annual Fees Before and After Rebates at End June 2017

119. The euro area SRI equity funds show costs well below the average of its peer group, with the exception of Fédéris who proposed a very competitive 20 bps management fee in its response to the Request for Proposal (RfP). The costs for the corporate bond investments are in line with the average of the peer group. The expense ratio for the government bond fund is excellent for this type of investment, even if Blackrock (iShares) now proposes even lower management fees so the Board will explore the possibility of a change. For the two funds selected in the latest selection processes, BlackRock and Vanguard, respectively proposed the most interesting management fees in world and emerging markets equities, far below the average of their Morningstar categories. At the same time, it should be noted that, for all asset classes, passively managed index-linked funds show lower fees than actively managed funds.

CONCLUSION xi:

On the evaluation of asset managers, the Board concludes that the assets of the PRF are well managed by the selected managers according to their investment objective, in a cost-effective manner as compared to fees charged by the peer group, and with gross performances above those of their respective benchmarks and above most of their peer groups. The Engagement Overlay Service provider is also producing the expected results.

E. ADMINISTRATION

E.1. TREASURY MANAGEMENT

120. Since the PRF receives contributions that are partly designated to pay benefits and administration costs during the year, a part is invested as treasury. The average treasury part of the PRF amounted to 3.27% for the period 2013 - 2016.

121. The treasury management is carried out by the ISRP, following a Treasury Management Plan (TMP) originally approved by the Management Board in September 2007 and reviewed every year. The TMP makes assumptions on the amount and timing of the contributions to be received and on the expected benefits and administration costs to be paid by the PRF during the year. The main objective of the TMP is to minimise the amount of treasury held, while allowing the payment of benefits and administration costs in a timely manner without undoing long-term investment positions. In parallel, treasury is invested with an objective of maximising its return while keeping the counterparty and interest rate risk at a reasonable level.

122. With regard to treasury instruments, the Board examined bank deposits, money market funds, certificates of deposits[33] and French saving accounts “Comptes sur Livret Association” (CLA). The instruments eventually selected were bank deposits, certificates of deposits, and in recent years especially CLAs given the current level of short-term interest rates.

123. The risk-limitation measures have evolved over the period under review in order to adapt to the historically low-rate environment. During its meeting of 29 September 2014, the Management Board revised the treasury investment procedures so as to widen the spectrum of eligible bank counterparties in a context of low and potentially negative market rates, avoid the concentration of counterparties and meet the objective of capital preservation. The Management Board modified the procedure to preferably involve counterparties with a minimum rating of “AA-/Aa3” at the time of the transaction according to at least one of the three main rating agencies – Fitch, Standard & Poor’s and Moody’s.^{^[34]} The maximum lifetime of such deposits was limited to six months. The Board even broadened the range of eligible counterparties to those banks rated “A-/A3” by all three main rating agenciesunder the following conditions: (i) less than three “AA-”-rated banks propose positive rates, (ii) they are considered systemically important by the Bank for International Settlements, (iii) they passed the latest Asset Quality Review, (iv) they have no negative credit outlook from any of the three main rating agencies, and (v) deposits with these banks are limited to three months maximum.

124. However, the conditions to operate with “A-/A3” banks and the maturity limits proved to be too restrictive in the 2015 rates environment as the universe of eligible counterparties proposing positive rates for deposits within the maturity limits was again very limited. Thus, the Secretariat, after having looked into several possibilities, proposed to enlarge the maturity limit for all eligible banks to one year, remove the condition that “A-/A3” banks should not have a negative credit outlook.[35]

125. To summarize, at present, the ISRP implements the following internal risk-limitation measures for treasury investments, after the Board's approval:

· Bank deposits should preferably involve counterparties with a minimum rating of “AA-“ (or equivalent) at the time of the transaction according to at least one of the three main rating agencies, i.e. Fitch, Standard & Poor’s and Moody’s. If less than three “AA-” or better-rated banks proposed zero or positive interest rates, bank deposits may be made with banks:

- rated at least “A-” (or equivalent) by all three main rating agencies or at least two agencies if a third rating is not available,

- listed as “systemically important institutions” by the Bank for International Settlements, and

- have passed the latest Asset Quality Review test from the European Central Bank.

· The maximum lifetime of deposits should be limited to 12 months.

· At least three banks shall be consulted at the time of the investment.

· Counterparties are usually included in an SRI index, are signatories of Equator principles^{^[36]}or have high standards of Corporate Social Responsibility.

· The ISRP aims at diversifying the counterparty risk between banks, if not (excessively) hurting investment returns.

· Moreover the monetary policy of the European Central Bank and market expectations / reactions are monitored, noting that the time of inflows does not always match the timing of interest rate moves.

126. The treasury performance is reported separately from the performance of the long-term investment portfolio. The ISRP calculates it with reporting methods established after consultation with the COE's and OECD’s treasury departments, and with reference to international standards. Table XXVII shows the average performance of the treasury part for the period 2014-2017 (at 30 June) that has been higher than both the benchmark (the Euro Overnight Index Average) and the average of Euro short-term money market funds.

Table XXVII - PRF Treasury Performance

Source: ISRP, EONIA homepage and Morningstar

CONCLUSION xii:

The Board concludes that the Treasury Management Plan has proved effective during the period under review and that short-term investments have been managed in a prudent way while producing a return above that of the benchmark and above the return of Euro short-term money market mutual funds.

E.2. ACCOUNTING AND BACK-OFFICE

Execution and administration of investments

127. The negotiation of contracts with asset managers as well as the execution and daily administration of the investment operations (long-term and treasury) are done by the ISRP, as provided for by the COE Secretary-General’s mandate of 1 July 2006, and in accordance with the Investment Procedures of the Fund that are based on the Investment Strategy as approved by the Management Board and Committee of Ministers.

128. The ISRP has created procedures for the execution of investments for all Funds under its administration, adapted to the characteristics of each Fund, and which are updated on a regular basis. The ISRP is, for example, currently setting up a new electronic signature validation process with SGSS to facilitate the transmission of orders. This new process should be adopted by the beginning of 2018.

CONCLUSION xiii:

The Board notes that no operational incidents during the period 2013-2016 are to be reported, meaning that the operational risk in the administration of the PRF has been successfully minimised.

Investments and rebalancing during exceptional circumstances

129. The Management Board may deviate temporarily from the SAA and its rebalancing limits in periods of exceptional circumstances, such as the implementation of a new strategy or the occurrence of extreme uncertainty or volatility in financial markets.

130. The decision to declare exceptional circumstances for the implementation of a new investment strategy can be taken during Board meetings. However, exceptional circumstances due to extreme volatility or uncertainty in financial markets should be declared at the time of the investment or rebalancing. The process to determine exceptional market circumstances is as follows:

1. As net contributions are received or on the date of rebalancing,^{^[37]} the ISRP monitors the daily returns of the MSCI All Countries World Index and the VIX Index to see if they reach abnormal levels^{^[38]}. If either of the two indicators is at abnormal levels, the ISRP sends an email to the Board to report on the level of the index raising the alert and the situation of the other indices (MSCI All Countries World daily returns / VIX, iTraxx Crossover and the spread Euribor 3M - Three-month German government debt). Investment operations[39] will be carried out by the ISRP in equal parts over the five business days following the day on which the ISRP monitored the indicators.

2. If neither of the two indicators is at abnormal levels, the ISRP will carry out the entire investment operations in one business day.

CONCLUSION xiv:

The Board notes that the procedure on exceptional circumstances effectively dealt with extreme market volatility, e.g. in 2011, and was not triggered since.

Accounting

131. The accounting of the PRF, carried out by the ISRP in accordance with its mandate, complies with the International Public Sector Accounting Standards (IPSAS) issued by the International Public Sector Accounting Standards Board (IPSASB), of the International Federation of Accountants (IFAC).

132. Until 2015, the accounts were verified, at the ISRP level and on a monthly basis by an accounting firm – Cabinet Quantin – after which they were transferred, on an annual basis and after the Board’s approval, to the COE’s financial department for integration in the Organisation’s consolidated accounts. The latter in turn, are audited by the COE’s External Auditors and approved by the Committee of Ministers. As agreed with the Secretariat of the Organisation, the verification performed by Cabinet Quantin has been replaced by an advisory mandate with PricewaterhouseCoopers that will advise the ISRP on accounting issues related to the implementation of new investments and the application of new relevant IPSAS standards or other.

133. During its meeting of November 2013, the Board decided to review financial statements on a half-yearly basis instead of quarterly.

134. Finally, IPSAS 28, 29 and 30[40] were published and have been applied to the financial accounts since 2012. The main impact for the PRF is that transaction costs and entry fees into the mutual funds are no longer capitalised with the cost of the investment, but expensed in the performance statement, and the disclosure of the PRF risk framework and risk management measures. The latter is being reviewed by the ISRP and the Organisation for publication in the 2017 financial statements.

Administration costs

135. The administration costs of the PRF are forecast in the annual operating budget that is prepared by the ISRP and submitted to the Board for approval and next to the Committee of Ministers, for approval in the consolidated Operating budget of the Organisation. Table XXVIII below describes in detail the evolution of the actual expenses (commitments included)[41] for 2013 – 2016.

Table XXVIII - Evolution of the PRF’s Operating Costs for 2013 – 2016

136. The sharp increase in 2016 operating costs is mainly due to the new Engagement Overlay strategy with BMO Global Asset Management (EUR 50 000 for a full year), appearing in the 'Consultant contracts' item. F&C proposed to charge an annual fee of EUR 75 000 that the Secretariat managed to negotiate to EUR 50 000.

Control of management costs

137. Table XXIX presents the management costs of the PRF for 2013 - 2016, as controlled by the ISRP.

Table XXIX - Evolution of the PRF’s Management Costs for 2013 – 2016

138. Management costs, as a percentage of assets held by the Fund, diminished from 0.40% in 2013 to 0.26% in 2016. As from the implementation of the new SAA at the beginning of 2016, the share of actively-managed funds decreased in the portfolio, resulting in a sizeable decrease in management fees (net of rebates). This was partially offset by the increase in dilution fees (paid to the fund) and the new Engagement Overlay strategy with BMO Global Asset Management. Simultaneously, the increase of custody fees between 2015 and 2016 comes from the Transfer Agent not using Euroclear anymore, a low-cost settlement platform, for the World Equity and Emerging Market Equity funds (representing 50% of the portfolio).

139. The Secretariat negotiated management fee rebates with fund managers which are calculated as a percentage of the TER for the funds managed by Allianz Global Investors, Amundi and SSGA, and as a percentage of management fees for the fund managed by BNP Paribas Asset Management. Depending on the contract, the rebates are calculated by the asset managers on an average or daily basis. The table hereafter depicts standards fees (available from funds' prospectuses or from asset managers' information) vs negotiated ones.

Table XXX – Standard vs Negotiated Fees (2016)

140. Total costs, including administration costs, showed the same trend, diminishing from 0.56% in 2013 to 0.35% in 2016. This also reflects the growth of the portfolio size, as compared to the fairly steady administration costs.

E.3. REPORTING

Custodian bank

141. The custodian bank is responsible for keeping the Fund's assets, reporting on the PRF investment positions and calculating the performance.

142. In 2006, the ISRP carried out a request for information that was sent to six custodian banks to compare their services, fee schedules and reporting capabilities. Based on the competitive fee schedule and cash remuneration of the current account, in addition to a good credit rating and adequate reporting capabilities of the firm, the Management Board decided to appoint Société Générale Securities Services (SGSS) as a custodian bank for the PRF.

143. In March 2012, SGSS informed the ISRP that it planned to raise its custody tariffs. During its market consultation to study other alternatives, the ISRP contacted four other well-known global custodians, i.e. J.P. Morgan, Brown Brothers Harriman, Bank of New York Mellon and State Street Global Services. Only two of the firms contacted made a price proposal. They were respectively 132% and 153% more expensive than the then prevailing tariff of SGSS. Following negotiations between the ISRP and SGSS, the latter decided not to pursue the announced fee increase.

144. In 2017, the ISRP thoroughly reviewed the custody services provided by SGSS and concluded that keeping SGSS as the custodian bank for the PRF was the most efficient option. The reason for this related to the fact that SGSS was the custodian bank for all Funds under the administration of the Secretariat generating important economies of scale. Also, the SGSS custody services had been in place for a long period of time and had allowed both the Secretariat and SGSS to get to know the respective working environments.

Performance calculation

145. The performance calculation is carried out by a special unit at the custodian bank. The ISRP contrasts the performance figures on a monthly basis with the numbers reported by the asset managers through Reuters or Morningstar and sends the information to the COE’s financial department. The performance is reported periodically to the Management Board during its meetings. Moreover, the Fund’s performance is presented to the Committee of Ministers in the Board's half-yearly reports.

146. The performance report provided by the custodian includes a comparison to the benchmarks for the total portfolio, for each asset class and for each external fund manager. The report also includes various risk parameters. The SIRP adapts the report to the specific requirements of each recipient party.

147. The external auditor of the COE recommended the Management Board to add, in their report to the Committee of Ministers, the Year-on-Year performance together with the monthly figures for the corresponding period. This was implemented in 2016.

148. During the course of 2016-2017, the ISRP and SGSS substantially reviewed the performance service following the ISRP's request of calculating the risk measures with weekly observations as from the implementation of the new strategy, and the evolution in SGSS's systems in view of automating as much as possible the procedures and reducing errors.

Methodology

149. The monthly, Year-To-Date (YTD), annual, three-year, five-year and since inception performances are calculated with a Time Weighted Return measure, indicating the performance of the asset manager only, without taking into account the effect of the investment of new contributions. This performance is to be compared to the benchmark to evaluate the skills of the asset managers and the Board's mandate to implement the SAA as approved by the Committee of Ministers.

150. To obtain the annual and actual performance since inception of the PRF, taking into account both the performance of asset managers and the effect of the investment of new contributions, the ISRP uses the Internal Rate of Return on investments. This figure gives an indication of the actual return obtained from the financial management of the Fund, i.e. the return obtained (by the asset managers) on the initial capital plus the impact on the performance of the investment of new contributions. This rate is to be compared to the target return.

151. In order to best reflect the investment universe of the asset managers, benchmarks are all reported by using the net return version of the indices (i.e. including reinvestment of net dividends) for equity investments and the total return version of the indices (i.e. including reinvestment of gross coupons) for the fixed-income investments.

152. The total performance of the PRF, as well as the performance of each asset manager, is reported net of management fees (taking into account negotiated fee rebates), brokerage costs and mutual funds’ administration costs, but gross of the PRF’s administration costs included in the operating budget.

153. The performance methodology used is compliant with international standards (Global Investment Performance Standards – GIPS) and, both, the ISRP and the custodian bank continuously work to improve the performance reports’ accuracy, the level of information and the ongoing compliance with international standards.

154. In 2013, the Board carried out a thorough review of the Risk management framework of the Fund and took some decisions aimed at improving the monitoring and diminishing the risks the PRF is exposed to [PRF/MB(2013)1/REV1].

155. In order to remain compliant with international standards and to better monitor the volatility of the portfolio, several changes to the performance reporting were proposed in the Second Three-Year Review, along with the implementation of a new Strategic Asset Allocation. It was thus decided to report risk measures on a one-year, three-year, and five-year sliding scale, as well as since inception of the investment strategy. Moreover, it was decided to use weekly observations as a basis for the calculation of the risk measures. This new methodology was put into place by SGSS in 2017.

CONCLUSION xv:

The Board considers the current performance reporting to be adequate and will continue to monitor the risk of the portfolio while remaining compliant with international standards.

ANNEX I: SOCIALLY RESPONSIBLE INVESTMENTS POLICY OF THE COUNCIL OF EUROPE[42]

BASIC CRITERIA FOR THE ANALYSIS & EVALUATION OF SOCIALLY RESPONSIBLE INVESTMENTS

1. The most extended definition of Socially Responsible Investments and most of the analyses and strategies available are based on three pillars: Environment, Social, and Corporate Governance (ESG).

· Environment - companies are evaluated on their consumption of natural resources and raw materials, emissions into the atmosphere (contribution to climate change), control of pollution risks, respect of biodiversity and ecosystems, water collection, recycling and disposal, energy consumption, product innovation, and disclosure of environmental information, conformity with the United Nations’ Global Compact‘s environmental requirements.

· Social - respect of human and workers’ rights: compliance with human rights universal declarations and the International Labour Organisation (ILO) tripartite declaration, lifelong learning, respect of consumers (product security and public health), fair trade, quality relations with suppliers, commitment of the company with the local communities where they operate, clean transportation, measures against corruption, and money laundering.

· Governance - traditional corporate governance issues and governance of ESG risks, deontology, election of equilibrated executive bodies, internal audit and control organisation, respect of shareholders’ rights.

2. These criteria are very much compliant with the COE’s views. Actually, many of the COE’s conventions (or parts of them) are already incorporated into the criteria that asset managers and research companies use for the SRI evaluation of companies, either directly or by assimilation with other international norms.

VIEWS OF THE COUNCIL OF EUROPE W.R.T. ENVIRONMENTAL, SOCIAL & GOVERNANCE CRITERIA [43]

Environment

3. Generally speaking, the criteria on the protection of the environment must be in conformity with European environmental regulations whose main principles are:^{^[44]}

· The principle of precaution; the principle of prevention; the principle whereby priority is given to dealing with environmental problems at source, and the 'polluter pays' principle;

· The European Directives concerning environmental impact studies (EIS);

· European Directives concerning industrial production, water and waste management, soil and air pollution and the protection of nature.

Moreover, companies must respect all relevant international conventions and agreements.

4. The policy of the COE concerning Nature Protection is based on the Berne Convention relative to European wildlife and natural environment preservation (1979). The Berne Convention coordinates the actions of the European States for the adoption of norms and common policies for the sustainable use of the biological diversity, contributing this way to the improvement of the life of Europeans and the promotion of an enduring development.

5. On the other hand, the European Landscape Convention of the Council of Europe (2000) can be regarded as complementary to the UNESCO Convention concerning the Protection of the World Cultural and Natural Heritage of 16 November 1972. The two conventions have different purposes, as do the organisations under whose auspices they were drawn up. One is regional in scope, the other world-wide. The COE convention covers all landscapes, even those that are not of outstanding universal value, but does not deal with historic monuments, unlike the UNESCO Convention.^{^[45]}

6. Related to water use and protection, companies must be dutiful with the European Agreement on the Restriction of the Use of certain Detergents in Washing and Cleaning Products (1968).

7. Animal rights: the COE also asks for the respect of the European Convention on the protection of vertebrates animals used for experimental and other scientific purposes (1986), the European Convention for the protection of animals for slaughter (1979), the European Convention for the protection of animals kept for farming purposes (1976) and the European Convention on the protection of animals in international transport (1968).

Social

8. The COE has three fundamental pillars concerning social policy: human rights, democracy and social cohesion. The most relevant for the definition of the SRI policy of the PRF relate to human rights and social cohesion.

v Human and workers’ rights – (PRIORITY ISSUE)

9. The European Convention on Human Rights and its protocols guarantee civil and political human rights. The European Social Charter (revised) (1996), its natural complement, guarantees social and economic human rights. The European Social Charter sets out rights and freedoms and establishes a supervisory mechanism guaranteeing their respect by the States Parties.

10. The rights guaranteed by the Charter concern all individuals in their daily lives and many of its criteria can be used for the definition of companies’ social/labour policy.

· prohibition of forced labour;

· security and hygiene at work;

· prohibition of the employment of children under the age of 15;

· special working conditions between 15 and 18 years of age;

· the right to earn one’s living in an occupation freely entered upon;

· fair working conditions as regards pay, holidays and working hours;

· protection from sexual and psychological harassment;

· freedom to form trade unions and employers’ organisations to defend economic and social interests; individual freedom to decide whether or not to join them;

· information to workers, promotion of joint consultation, collective bargaining, conciliation and voluntary arbitration;

· protection in case of dismissal;

· the right to strike;

· access to work for persons with disabilities;

· the right of women and men to equal treatment and equal opportunities in employment (art. 20);

· a guarantee to all nationals and foreigners legally residing and/or working that all the rights set out in the Charter apply regardless of race, sex, age, colour, language, religion, opinions, national origin, social background, state of health or association with a national minority;

· prohibition of discrimination on the basis of family responsibilities.

11. To be consistent with the Organisation’s views, companies where education, vocational training, and the creation of stable jobs are promoted should be considered for investing.^{^[46]}

12. Finally, the Guidelines to assist Internet Service and Online Games Providers, developed by the COE in co-operation with the European Association of Internet Services Providers (ISP) in 2008, establish a reference for their activities. The Guidelines reckon the significance that these suppliers have as providers of internet access, email, information, and online games and at the same time, they underline the importance to protect users, especially as regards their right to privacy and freedom of speech. The Guidelines highlight the important impact the activities of internet services and online games providers may have on human rights.

13. Safeguarding human rights in times of economic crisis (2013): In order to ensure the effective and equal enjoyment of all human rights –civil, political, economic, social and cultural – in times of economic crisis and fiscal austerity, the Commissioner for Human Rights calls on COE member States to:

· institutionalise transparency, participation and public accountability throughout economic and social policy cycle;

· conduct systematic human rights and equality impact assessments of social economic policies and budgets;

· promote equality and combat discrimination and racism;

· ensure social protection floors for all;

· guarantee the right to decent work;

· regulate the financial sector in the interest of human rights;

· work in concert to realise human rights through economic co-operation and assistance;

· engage and support an active civil society;

· guarantee access to justice for all;

· ratify European and international human rights instruments in the field of economic and social rights;

· systematise work for human rights, engage and empower national human rights structures in response to the economic crisis.

v Social cohesion

14. In its strategy for social cohesion, the COE defines social cohesion as the capacity of a society to ensure welfare of all through the shared responsibility of its various stakeholders (public and private players, citizens). A link is established with sustainable development, through the inclusion of the welfare of future generations.

v Health

15. Biotechnology:

· GMO: As long as companies and countries fulfil the following regulations, GMOs are not forbidden (Resolution 870, 1986 adopted by the Parliamentary Assembly of the COE):

- the Biological diversity convention signed in Rio de Janeiro at the UN Conference about environment and development (1992);

- the Cartagena Protocol to prevent biotechnology risks, related to the Convention on biologic diversity (29 January 2000, Montréal);

- farmer’s rights as it comes from the Resolution adopted at the 25^th session of the FAO Conference (November 1989).

· Embryo research: the Oviedo Human rights and biomedicine Convention (1997) has two articles (14 and 18) concerning embryo research. The agreement forbids using medical assistance techniques in order to select the sex of the foetus (unless the aim is to avoid a serious hereditary condition linked to gender) and the constitution of human embryos for research purposes. If this kind of research was admitted by law, this law must ensure an adequate protection to the embryo.

· Cloning of human beings: there is an additional protocol to the Oviedo Human rights and biomedicine agreement on forbidding this point (1998).

· Genetic research: the Oviedo Human rights and biotechnology Convention has four articles (11 to 14) concerning genetic tests and interventions on human genome.

· Consumers’ health: the COE has published the Guidelines for good practices in the production of cosmetics (1995). In parallel, the Resolution ResAP(2007)2 on good practices for distributing medicines via mail order intends to protect patient safety and the quality of the delivered medicine.

16. Emerging technologies (2015): the Council realises the possibility of abuse as a reason for concern from the perspective of human rights and human dignity. The International Conference on Emerging Technologies and Human Rights took place on May 2015. Interactions between the life sciences and the engineering sciences are increasing as more and more innovations in the biomedical field are emerging from the convergence of developments in different domains, including nanotechnology, cognitive sciences and information technology.

17. MEDICRIME Convention (2016): The COE drafted a convention which constitutes a binding international instrument in the criminal law field on the counterfeiting of medical products and similar crimes involving threats to public health. This is the result of a longstanding issue about the absence of harmonised international legislation, non-deterrent sanctions that were not proportionate to the harm caused to patients, and the involvement of criminal organisations which operate across borders. The Convention establishes as offences:

· the manufacturing of counterfeit medical products;

· supplying, offering to supply and trafficking in counterfeit medical products;

· the falsification of documents;

· the unauthorised manufacturing or supplying of medicinal products and the marketing of medical devices that do not comply with conformity requirements.

v Corruption, money laundering and cybercrime

18. The COE has also shown its concern regarding the fight against money laundering, corruption, and cybercrime. While respecting the Civil Law Convention on Corruption (1999), the Criminal Law Convention on Corruption (1999), the Convention on Laundering, Search, Seizure, and Confiscation of the Proceeds from Crime (1990), the Convention on Laundering, Search, Seizure, and Confiscation of the Proceeds from Crime and on the Financing of Terrorism (2005), and the Convention on Cybercrime (2001), the COE will use general SRI criteria (from asset managers).

Governance

19. In the Parliamentary Assembly of November 2004, a definition on what business ethics should be was drafted that is very similar to the positions already adopted by some Pension Funds.

· use of auditors of quality and proven independency;

· strong Chinese walls when the activities of business counselling and investment are done at the same company;

· training in Corporate Social Responsibility to managers, non-executive administrators and to the members of the Administration Board;

· encouragement of regular shareholder voting;

· separation of the roles of President of the Administration Board and General Manager for big publicly quoted companies;

· control over managers’ salaries in terms of quantity and transparency.

20. European Label of Governance Excellence Benchmark (EloGE) (2016): this label is awarded to local authorities having achieved a high overall level of governance by a national or regional stakeholders’ platform accredited by the COE. The evaluation allows local authorities to understand their strength and weaknesses when providing public services to the local community and exercising public authority.

[1] This document has been classified restricted at the date of issue; it will be declassified in accordance with Resolution Res(2001)6 on access to Council of Europe documents.

[2] The remuneration scheme should be defined by the Committee of Ministers and specify the starting date (as from a specific date or as from the mandate renewal date), the amount and adjustment method, whether or not applicable to all Board members, and the source and terms of payment (from the PRF or directly from the CoE).

[3] The remuneration scheme should be defined by the Committee of Ministers and specify the starting date (as from a specific date or from the mandate renewal date), the amount and adjustment method, whether or not applicable to all Board members, and the source and terms of payment (from the PRF or directly from the CoE).

[4] Some topics will be presented and analysed covering a broader period for a better understanding.

[5] PRF/MB/WD(2015)8/REV1

[6] The remuneration scheme should specify the starting date (as from a specific date or from the mandate renewal date), the amount and adjustment method, whether or not applicable to all Board members, and the source and terms of payment (from the PRF or directly from the CoE).

[7] CM(2014)4

[8] All asset classes are denominated in EUR.

[9] For details on the study of the universe of asset classes and the related decision, see PRF/MB/WD(2016)13/REV1, PRF/MB/WD(2016)13/REV1/ADD1 and PRF/MB/M(2016)4.

[10] Setting limits to the strategy is a common practice among pension funds of other international organisations. Such limits are usually based on peer practices, liquidity, risk preferences, availability of investments and Fund regulations (related to Socially Responsible Investments or country exclusions, for example).

[11] See document PRF/MB/WD(2016)11.

[12] Indices representative of the global fixed-income market.

[13] As per the interpretation of the Fund Statute: "The objective of the Pension Reserve Fund (“the Fund”) is to smooth, in the medium and long term, the financing of the member States’ obligations under the Organisation’s Pension Schemes."

[14] The document SIRP/E(2017)16/REV1 Global Contribution Rate Update as at 31/12/2016 shows that, with 3.1% and 3.4% target returns, the objective of sustainability can be attained only if the Global Contribution Rate increases (from 30.1%) to 38.88% and 37.85% respectively.

[15] This implies that, in the current situation of the COE as regards the funding of its pension schemes, every year the pension contributions plus the Fund return meet the pension payments and that the Fund does not have an expected depletion date.

[16] In case of asset manager’s bankruptcy, there may be a temporary lack of liquidity, while the legal / administrative procedures are carried out.

[17] Indeed this view is valid within the environment of pension funds having net inflows, and mean reversion models (where market prices revert to their long-term trend).

[18] The efficient markets’ theory as well as empirical evidence shows that the relative risk coming from active or tactical management is not consistently rewarded over the long term.

[19] As per a mean-reversion model, meaning that negative returns will be compensated by positive ones, in order to reach the (long-term) expected average.

[20] Changes in the pension debt are not considered here as a risk factor, i.e. having a probability of postponing the sustainability date, but they may have.

^{^[21]}As per the risk of an investment strategy designed to produce a 4.0% net real return (assuming normality of returns).

[22] As per the risk of an investment strategy designed to produce a 3.4% net real return (assuming normality of returns).

[23] As per the risk of an investment strategy designed to produce a 3.1% net real return (assuming normality of returns).

[24] Having said that, diversified portfolios with moderate risk are in any case desirable as they will have less severe downturns and therefore they will require less important compensations.

[25] Where the global contribution rate has been adjusted to make the Fund sustainable in the long term with a 3.4% investment return.

[26] Observer States of the COE are Canada, Holy See, Israel (Observer to the Parliamentary Assembly), Japan, Mexico, and the United States.

[27] Observer Status with the COE: Criteria (CM/Del/Dec(99)667/2.4, 668/2.4, 671/2.3, 674/2.3, CM(99)58-rev. 2).

[28] This is an important point as there is not an appropriate offer of SRI global government bond EUR hedged funds.

[29] F&C Investments became part of BMO Global Asset Management in July 2015.

[30] The BlackRock and Vanguard funds were only invested in January 2016 but performance and risk figures are computed over the last three years in this section in order to allow for a more meaningful comparison.

[31] The Sharpe ratio indicates the excess return above the risk free return per unit of risk taken. The higher the ratio, the better risk is rewarded.

[32] A detailed list of all costs and fees of the PRF investments for 2013 - 2016 is given in Table XXIX of this document.

[33] Certificates of deposit (CDs) are similar to bank deposits in that it is a placement in a bank with a fixed term and a fixed interest rate. However, while certificates of deposit are tradable financial products, a deposit is a bilateral agreement between the counterparty bank and the client (the Council of Europe).

[34] The Secretariat uses the rating agencies' long-term rating.

[35] The applicable investment procedures for treasury management were approved by the Management Board in 2015 [PRF/MB/WD(2015)5, TMP for 2015, and PRF/MB(2015)4/REV2] and adjusted in April 2016 [PRF/MB/M(2016)01].

[36] The Equator Principles is a risk management framework, adopted by financial institutions, for determining, assessing and managing environmental and social risk in projects. The Equator Principles are based on the environmental standards of the World Bank and the social policies of the International Finance Corporation.

[37] The process starts only if rebalancing is needed.

[38] Abnormal levels are defined in document PRF/MB/WD(2015)21: when the daily movement of the MSCI All Countries World Index shows two or more readings above “average since inception +/- 2 times the standard deviation” over the last five business days and/or when the VIX index, over the last five business days shows more than two readings above “average since inception + 2 times the standard deviation”.

[39] Investment operations are defined as investing new incoming contributions and/or rebalancing operations.

[40] IPSAS 28 defines the requirements for presenting information about financial instruments, IPSAS 29 establishes the principles for recognising and measuring financial assets and liabilities and IPSAS 30 defines the requirements for disclosing information about financial instruments.

[41] The difference between the approved budget and the actual expenses (according to the financial statements at the end of the year) has been deducted from the ISRP invoice of the following year.

[42] Version 2017, where paragraphs 16 and 17 have been added as regards the previous version.

[43] Include rules and policies relevant to SRI and in the case of the Conventions, only those signed and ratified by the majority of euro area countries.

[44] The COE’s Development Bank states these principles in its general philosophy to assess projects to finance.

[45] In the work leading up to the drafting of the convention, constant reference was made to existing international and national legal texts concerned with landscape. Were included:

· the Mediterranean Landscape Charter;

· the European Community regulation on agricultural production methods compatible with the requirements of the protection of the environment and the maintenance of the countryside;

· the European Community directive on the conservation of natural habitats and of wild fauna and flora;

· the European Community directive on the assessment of environmental effects, and other important national, European Community and international instruments;

· the Convention on Biological Diversity (Rio, 5 June 1992); and

· the Convention on Access to Information, Public Participation in Decision-making and Access to Justice in Environmental Matters (Aarhus, 25 June 1998).

[46] At the COE Development Bank, priority is given to projects respecting or promoting these principles.