|
MINISTERS’ DEPUTIES |
CM Documents |
CM(2019)170 |
23 October 2019[1] |
|
1361st (Budget) meeting, 19-21 November 2019 11 Programme, Budget and Administration
11.4 Co-ordinating Committee on Remuneration (CCR) – Review of the staff contribution rates to the pension schemes c. Third Pension Scheme Item to be considered by the GR-PBA at its meeting on 4 November 2019 |
1. Under Article 43 of the Rules governing the Third Pension Scheme (TPS), in force at the Council of Europe, the Pensions Administrative Committee of the Co-ordinated Organisations (PACCO) “shall give technical opinions” on draft amendments to the Rules. It is further to this provision that PACCO had given a favourable technical opinion to the preceding review of the contribution rate for staff members affiliated to the TPS.
2. The contribution rate for staff members affiliated to the TPS is currently 9.4%. The deadline for the update of this contribution rate coincides with that for the Co-ordinated Pension Scheme (CPS) and for the New Pension Scheme (NPS), i.e. 1 January 2020.
3. The technical opinion of PACCO is to increase to 10.6 % as of 1 January 2020 the contribution rate for staff members affiliated to the TPS.
Technical note – Review of the staff contribution rate
to the Third Pension Scheme (TPS)
1. Methodology background
1.1. The Third Pension Scheme (TPS) was implemented on 1 April 2013 at the Council of Europe, for staff members who took up duty on or after that date.
1.2. The TPS was established on the basis of a number of guidelines which involved the following amendments:
- to the New Pension Scheme benefits (in particular age of retirement and accrual rate);
- the proportion of staff contributions of the total cost of the scheme (45% instead of 40%).
1.3. Article 41 of the Rules governing the TPS provides, as for the CPS and the NPS, that the rate must be reviewed every 5 years. The previous revision, which took effect on 1 January 2015, led to an increase of the contribution rate to 9.4%.
2. Timetable
2.1. An actuarial study must be conducted every five years for all of the organisations. In accordance with the results of the study, the staff contribution rate is adjusted automatically, with effect from the fifth anniversary of the preceding adjustment.
2.2. The preceding adjustment, calculated in accordance with the method, was recommended by the Pensions Administrative Committee of the Co-ordinated Organisations (PACCO) in 2014.
2.3. The next adjustment must take place on 1 January 2020. Following the same timetable as the previous exercise, the actuarial studies began a year and a half prior to the next scheduled adjustment date, on the basis of data available at the end of 2017.
3. Actuarial parameters
3.1 Discount rate
3.1.1 Further to a technical opinion delivered during PACCO’s 190th meeting, the discount rate used for this review of the contribution rate for staff members affiliated to the NPS and TPS is identical to the discount rate used for the actuarial study covering the CPS.
3.1.2 The Annex to Article 41 of the TPS Rules precisely defines the principles to be used for calculating the present value of the various quantities. These principles are based on observation of the rates of return of long-term bonds issued in the reference countries. PACCO had been requested to assess the relevance of the statistical series used for the discount rate calculation; it reached the conclusion that no statistical series other than the one currently in use met the criteria of technical robustness and reliability.
3.1.3 Therefore, the data used for each year is the average of government bond rates as extracted from the OECD statistical series “Long-term interest rates from the Monthly Monetary and Financial Statistics database (Main Economic Indicators – MEI)” for each of the reference countries, weighted by the number of staff members stationed in the country concerned, with data for Spain being taken into account as from 2007.
3.1.4 However, PACCO recommended a number of corrections, the implementation of which ensures that the calculations are in conformity with the spirit of the text and mathematically relevant:
- rate of government bonds:
o correction of rates for Luxembourg, that were formerly aligned on the Belgian rates;
o correction of certain national rates so as to align them with the values mentioned in the latest version of the OECD statistical series;
o use of data from the Eurostat “Central government bond yields – annual data” series when the OECD series is incomplete;
o for the years in which certain national data are absent from both the OECD and Eurostat series: non-inclusion of the country (or countries) concerned and re-scaling of the weighting;
- effective date of the study:
o for the calculation of the weighting by reference to the population: use of data as of 31 December 2017 (“effective date of the study” as by the Pension Scheme Rules), instead of data as of 30 June that was used to date;
o for the calculation of HICP: use of an annual running average of monthly rates (from January to December) for each country, instead of the difference between the values at 30 June of the current year and the previous year;
- calculation of real rates:
o use of HICP as published by Eurostat as from 1 January 2012, and use of the rates consistent with the salary adjustment method for previous years;
o use of a division, rather than a difference, for deflating national rates.
3.1.5 Amongst these corrections, only the method for deflating national rates requires a minor amendment to the actuarial method foreseen under the Annex to Article 41 of the TPS Rules.
Recommended amendment:
|
Current French version |
Recommended amendment |
|
9. Le taux réel moyen pour une année passée déterminée s'obtient à partir du taux réel de chaque pays, calculé comme étant la différence entre le taux de rendement brut des obligations et le taux d'inflation correspondant, tel qu'il est retracé par l'indice national des prix à la consommation. […] |
9. Le taux réel moyen pour une année passée déterminée s'obtient à partir du taux réel de chaque pays, calculé sur la base du quotient du taux de rendement brut des obligations par le taux d'inflation correspondant, tel qu'il est retracé par l'indice national des prix à la consommation. […] |
|
Current English version |
Recommended amendment |
|
9. The average real rate for a given past year is obtained from the real rates in each country, calculated as the difference between the rate of gross return on bonds and the corresponding rate of inflation, as shown by the national consumer price index. […] |
9. The average real rate for a given past year is obtained from the real rates in each country, calculated as the quotient of the rate of gross return on bonds by the corresponding rate of inflation, as shown by the national consumer price index. […] |
3.1.6 The final discount rate is a real rate, equal to the arithmetic mean of average real rates observed over a running period of 30 years preceding the effective date of the study. As of the effective date of the study, the discount rate is equal to 2.83%. The rate used for the review of the TPS contribution rate was 3.95% for the review implemented on 1 January 2015.
3.2 Rate of increases in salary scales
3.2.1 The rate of increase in salary scales must be set in real value. This rate is used in the actuarial calculations to increase future salaries (which are the basis of calculation for future pensions as well as the basis to which the rate of contribution is applied) as well as to increase pensions. This parameter must have a prospective dimension that is necessarily difficult to assess.
3.2.2 PACCO had been requested to assess the relevance of the calculation method. On the basis of an analysis done by the ISRP, PACCO recommended to keep the temporal scope unchanged, since its range is sufficient to capture a complete salary scale.
3.2.3 The modalities by which the salary increase assumption “take[s] account of available forecasts in that respect” comes from a practice based on the application of a value of 0.5% above inflation, which was retained as a result of a political compromise when setting the actuarial method. However, as this historical value is moving increasingly away from the recorded adjustment values, it may no longer legitimately be considered as a relevant “forecast” of salary increase. Consequently, PACCO established a new methodology.
3.2.4 Due to the effective date of the study, two full remuneration adjustment exercises are carried out prior to the entry into force of the revised rate. PACCO considered that the data for these two years of adjustment constitutes a relevant “forecast” of salaries evolution after the effective date of the study, hence allowing for a reinforcement of the clarity of the calculation method, which would be based purely on mathematical and objective factors. In addition, whereas an arithmetic average of yearly rates had been used to date, PACCO recommended the use of a geometrical average to the CRSG, since the data under study are rates that are correlated with the previous year.
3.2.5 The salary increase assumption, as calculated according to the recommended method, is the sum of the geometric average of the remuneration adjustment rates for staff members of the Co‑ordinated Organisations over the 15 years preceding the effective date of the study, this average being assigned a weight of 15/17, and of the geometric average of the remuneration adjustment rates for staff members of the Co‑ordinated Organisations over the two years following the effective date of the study, this average being assigned a weight of 2/17.
3.2.6 The salary increase assumption calculated according to the recommended method is 0.24%. It was 0.27% at the time of the previous review of the contribution rate.
3.3 Demographic assumptions
3.3.1 Actuarial studies require the calculation of a series of demographic assumptions that are specific to each organisation (turnover before and after vesting, annual probability of retiring, annual probability of becoming invalid, salary increases linked to career advancement, new entrants’ profiles, marriage rate and age difference between spouses, loading of family allowances).
3.3.2 All the demographic assumptions were calculated for the review of the contribution rate to the CPS and the NPS as of 1 January 2005 and have then been updated.
3.3.3 In the context of its work for the Co-ordinated Organisations, the ISRP undertook a thorough analysis of these assumptions based on historical data up to 31 December 2015, which led to their review.
3.1. Mortality tables
3.4.1 The mortality table developed by the United Nations, based on their own affiliated population, has been used for several previous studies. It was first used in its static form, that is, without factoring in life expectancy trends, which were only taken into account as from 2004.
3.4.2 Because of questions raised regarding the use of this table, in particular linked to the underlying population, which is certainly more diverse from a socio-economic point of view than that of the Co-ordinated Organisations, the Pensions Section (JPAS) worked in collaboration with Eurostat on a mortality table project for international civil servants based in Europe in order to better take into account the specificities of the population of the Co-ordinated Organisations from a life-expectancy point of view. This project was finalised in mid-2008. The International Civil Servants Life Table (ICSLT) was implemented using mortality data from several international organisations headquartered in Europe, including the Co‑ordinated Organisations as well as the European Union. In order to obtain a prospective version of the table, the JPAS took the trend of the United Nations over a period of 20 years. This work was validated by an outside expert, a professor of actuarial sciences at the Catholic University of Louvain.
3.4.3 In 2013, the ISRP and Eurostat updated the mortality tables with additional data collected over the years 2008 to 2013, across an increased number of international organisations based in Europe. In addition, in order to respond to feedback regarding the prospective version of the mortality tables, the ISRP developed a specific prospective trend, applied over a thirty-year period and calculated as an average between that of the United Nations selected in 2008 and those of the national tables for Germany, France, the Netherlands, Switzerland and the United Kingdom.
3.4.4 The ICSLT 2018 table was developed using data from international organisations based in Europe over the 2013-2017 period, after extensive discussions with a group of actuarial experts within the participating organisations and with the OECD’s Insurance, Private Pensions and Financial Markets Division. The table has characteristics comparable with ICSLT 2013, including a prospective trend. A growing number of international organisations use ICSLT tables (the Co-ordinated Organisations, ECB, EIB, EPO and EUROCONTROL).
Comparison of life expectancies between the ICSLT 2013 and ICSLT 2018 tables
|
ICSLT 2013 (projected to 2018) |
ICSLT 2018 |
|
|
Males |
||
|
At age 60 |
26.7 |
28.3 |
|
At age 60 for current age 40 |
27.4 |
29.0 |
|
Females |
||
|
At age 60 |
28.4 |
31.0 |
|
At age 60 for current age 40 |
29.1 |
31.5 |
4. Summary of results
4.1. As in the previous exercise, the calculations were entrusted to the ISRP.
The staff rate of contribution to the TPS is 10.6% of monthly salary – an increase of 1.2 percentage points compared to the current rate (9.4%).
4.2. It is recommended to proceed with an update to 10.6%, as of 1 January 2020, of the contribution rate for staff members affiliated to the TPS.
4.3. With regard to the method, it is recommended:
‐ to use the 2018 update to the mortality table ICSLT2013 ©;
‐ to use the recommended calculation method for the salary increase assumption;
‐ to implement the amendment recommended for the discount rate calculation in paragraph 9 of the Annex to Article 41.
4.4. Applying the method, it is recommended to modify Article 41, paragraph 3 of the TPS Rules as follows:
“The rate of the staff contribution shall be set so as to represent the cost, in the long term, of 45% of the benefits provided under these Rules. The rate shall be 10.6%. This rate shall be reviewed on 1st January 2015 and thereafter every five years or whenever necessary, on the basis of an actuarial study, the procedures for which are appended hereto. After that date, the staff contribution rate shall be adjusted, with effect from the fifth anniversary of the preceding adjustment, the rate being rounded to the nearest first decimal.”
Annex 1 – Article 41 – Staff member’s contribution – Costing the scheme
1. Staff members shall contribute to the TPS.
2. The staff members' contribution shall be calculated as a percentage of their salaries and shall be deducted monthly.
3. The rate of the staff contribution shall be set so as to represent the cost, in the long term, of 45% of the benefits provided under these Rules. The rate shall be 10.6%. This rate shall be reviewed on 1st January 2015 and thereafter every five years or whenever necessary, on the basis of an actuarial study, the procedures for which are appended hereto. After that date, the staff contribution rate shall be adjusted, with effect from the fifth anniversary of the preceding adjustment, the rate being rounded to the nearest first decimal.
4. Contributions properly deducted shall not be recoverable. Contributions improperly deducted shall confer no rights to pension benefits; they shall be refunded at the request of the staff member concerned or those entitled under him without interest.
Annex 2 – Annex to Article 41 – Actuarial studies
1. Calculation, as at the effective date of the study of the rate of contribution payable by staff in order to finance 45% of benefits provided under the Scheme, establishing the present value of future entitlements and salaries.
2. Projections of annual amounts of future entitlements will be calculated, on the one hand, for staff affiliated to the TPS at the date of the study and, on the other hand, for the population of staff who will be recruited and affiliated to this scheme in the years to come. Projections of salaries for these populations will also be established year by year. Each of these amounts will be projected over a period of 80 years and discounted to present worth.
3. Combining these results will make it possible to determine the rate of contribution needed to finance 45% of benefits provided under the Scheme.
Demographic and salary-related assumptions
4. The demographic assumptions are derived from detailed demographic studies for the Organisation. These studies examine past experience over a period of 15 years, where the information is available, and also take account of available forecasts regarding future staff numbers.
5. The assumptions relating to salaries are based on detailed observation of the past, over a period of 15 years, where the information is available, and also take account of practices and forecasts available in this field.
6. The rates obtained are adjusted so as to eliminate distortions resulting from insufficient data.
Economic assumptions
7. The discounting process is based on observed rates of return on long-term government bonds issued in the reference countries, as from the date when they become a reference country.
8. A discount rate net of inflation shall be used. It shall be equal to the arithmetical average of average real rates observed over the thirty years preceding the date when the actuarial study is conducted.
9. The average real rate for a given past year is obtained from the real rates in each country, calculated as the quotient of the rate of gross return on bonds by the corresponding rate of inflation, as shown by the national consumer price index. The average is obtained by weighting the real rate in each country by the number of serving staff in that country at the effective date of the study.
Annex 3 – Discounting process
Real rate of return of medium and long-term government bonds
(1988-2017)
|
Country |
BEL |
DEU |
ESP |
FRA |
GBR |
ITA |
LUX |
NLD |
Weighted average |
|
Population |
2,161 |
1,467 |
81 |
5,599 |
432 |
630 |
981 |
2,057 |
13,408 |
|
Weight without Spain |
16.2% |
11.0% |
42.0% |
3.2% |
4.7% |
7.4% |
15.4% |
100% |
|
|
1988 |
6.8 |
4.9 |
6.1 |
4.6 |
n/a |
5.6 |
5.3 |
5.8 |
|
|
1989 |
5.3 |
4.0 |
5.1 |
4.8 |
n/a |
4.2 |
6.0 |
5.0 |
|
|
1990 |
6.3 |
6.0 |
6.5 |
4.5 |
n/a |
4.7 |
6.3 |
6.2 |
|
|
1991 |
5.9 |
4.5 |
5.8 |
2.4 |
n/a |
4.7 |
4.3 |
5.2 |
|
|
1992 |
6.1 |
3.1 |
6.2 |
4.6 |
7.9 |
4.6 |
5.1 |
5.6 |
|
|
1993 |
4.4 |
2.1 |
4.7 |
4.8 |
6.5 |
3.2 |
3.8 |
4.2 |
|
|
1994 |
5.2 |
4.1 |
5.5 |
6.1 |
6.4 |
4.9 |
3.9 |
5.1 |
|
|
1995 |
5.9 |
5.0 |
5.6 |
5.4 |
6.3 |
5.3 |
4.9 |
5.5 |
|
|
1996 |
4.3 |
4.8 |
4.2 |
5.2 |
5.9 |
4.8 |
3.9 |
4.4 |
|
|
1997 |
4.1 |
3.8 |
4.5 |
5.1 |
5.0 |
4.0 |
3.4 |
4.2 |
|
|
1998 |
3.8 |
3.7 |
3.9 |
3.9 |
2.9 |
3.7 |
2.6 |
3.6 |
|
|
1999 |
3.6 |
3.7 |
3.8 |
3.7 |
2.9 |
3.7 |
2.4 |
3.5 |
|
|
2000 |
3.0 |
3.2 |
3.7 |
4.5 |
2.8 |
2.3 |
2.5 |
3.2 |
|
|
2001 |
2.6 |
2.7 |
3.2 |
3.6 |
2.5 |
2.1 |
0.7 |
2.6 |
|
|
2002 |
3.3 |
3.6 |
2.9 |
3.6 |
2.5 |
2.6 |
1.8 |
2.8 |
|
|
2003 |
2.5 |
3.0 |
2.0 |
3.1 |
1.6 |
1.2 |
2.0 |
2.1 |
|
|
2004 |
2.0 |
2.3 |
1.9 |
3.5 |
2.0 |
0.7 |
2.8 |
2.1 |
|
|
2005 |
0.6 |
1.4 |
1.6 |
2.3 |
1.5 |
-0.1 |
1.7 |
1.3 |
|
|
2006 |
2.0 |
2.0 |
2.1 |
2.1 |
1.9 |
0.6 |
2.6 |
2.0 |
|
|
Weight with Spain |
16.1% |
10.9% |
0.6% |
41.8% |
3.2% |
4.7% |
7.3% |
15.3% |
100% |
|
2007 |
2.5 |
1.9 |
1.4 |
2.7 |
2.6 |
2.4 |
n/a |
2.6 |
2.5 |
|
2008 |
-0.1 |
1.2 |
0.2 |
1.0 |
0.9 |
1.1 |
n/a |
1.7 |
1.0 |
|
2009 |
3.9 |
3.0 |
4.2 |
3.5 |
1.4 |
3.5 |
n/a |
2.5 |
3.3 |
|
2010 |
1.2 |
1.6 |
2.4 |
1.4 |
0.3 |
2.4 |
0.7 |
1.7 |
1.4 |
|
2011 |
0.7 |
0.1 |
2.1 |
1.0 |
-1.3 |
2.5 |
-0.4 |
0.5 |
0.7 |
|
2012 |
0.3 |
-0.6 |
3.3 |
0.3 |
-0.9 |
2.1 |
-1.0 |
-0.9 |
0.0 |
|
2013 |
1.2 |
0.0 |
3.0 |
1.2 |
-0.2 |
3.0 |
0.2 |
-0.6 |
0.8 |
|
2014 |
1.2 |
0.4 |
2.9 |
1.0 |
1.1 |
2.7 |
0.6 |
1.1 |
1.1 |
|
2015 |
0.2 |
0.4 |
2.4 |
0.8 |
1.9 |
1.6 |
0.3 |
0.5 |
0.6 |
|
2016 |
-1.3 |
-0.3 |
1.7 |
0.2 |
0.6 |
1.5 |
-0.2 |
0.2 |
-0.1 |
|
2017 |
-1.5 |
-1.4 |
-0.5 |
-0.4 |
-1.4 |
0.8 |
-1.6 |
-0.8 |
-0.8 |
|
Annual average over 30 years (1988 - 2017) |
2.83 |
||||||||
Sources
Gross rates: “Long-term interest rates from the Monthly Monetary and Financial Statistics database (Main Economic Indicators – MEI)” series published by the OECD Statistics and Data Directorate, completed by the “Central government bond yields – annual data” series published by Eurostat.
Inflation: HICP published by Eurostat as from 1 January 2012, and rates consistent with the remuneration adjustment method are used for previous years.
Annex 4 – Rate of increase in salary scales
|
Real Rate of Annual Salary Inflation (% p.a.) |
||||||||||
|
Belgium |
France |
Germany |
Italy |
Luxembourg |
Netherlands |
Spain |
UK |
Weighted Average |
Geometric Average (15 years) |
|
|
2003 |
1.9 |
1.9 |
1.9 |
2.9 |
0.8 |
1.9 |
1.9 |
1.86 |
1.05% |
|
|
2004 |
1.0 |
1.2 |
1.0 |
1.4 |
1.0 |
1.0 |
1.0 |
1.08 |
1.06% |
|
|
2005 |
-0.7 |
-0.2 |
-0.7 |
1.9 |
-0.8 |
-0.7 |
-1.2 |
-0.38 |
0.75% |
|
|
2006 |
0.0 |
0.0 |
0.0 |
3.0 |
0.3 |
0.0 |
-1.3 |
0.13 |
0.70% |
|
|
2007 |
-0.2 |
1.2 |
-0.2 |
0.0 |
-1.5 |
-0.2 |
-0.2 |
-0.2 |
0.30 |
0.52% |
|
2008 |
-0.2 |
-0.2 |
-0.2 |
-0.2 |
-0.9 |
-0.2 |
-0.2 |
-3.2 |
-0.34 |
0.39% |
|
2009 |
-0.6 |
-0.6 |
-0.6 |
0.1 |
0.9 |
-0.5 |
-0.6 |
-0.6 |
-0.43 |
0.28% |
|
2010 |
2.5 |
2.5 |
2.5 |
2.5 |
1.7 |
2.5 |
2.4 |
5.0 |
2.52 |
0.58% |
|
2011 |
-2.4 |
-1.7 |
-2.4 |
-3.3 |
-1.9 |
-3.9 |
-3.2 |
-1.6 |
-2.31 |
0.45% |
|
2012 |
-1.4 |
0.3 |
-1.4 |
-2.2 |
-1.3 |
-1.4 |
-1.4 |
2.1 |
-0.61 |
0.37% |
|
2013 |
-1.0 |
0.0 |
0.1 |
-2.9 |
-1.4 |
-1.0 |
-1.0 |
1.3 |
-0.50 |
0.29% |
|
2014 |
-1.8 |
-1.8 |
-0.5 |
-2.2 |
-2.3 |
-1.8 |
-2.9 |
-1.8 |
-1.71 |
0.14% |
|
2015 |
1.0 |
1.5 |
1.0 |
-1.7 |
0.5 |
1.0 |
-0.4 |
1.8 |
1.05 |
0.13% |
|
2016 |
1.4 |
1.4 |
1.4 |
1.4 |
1.4 |
1.4 |
-1.8 |
1.4 |
1.37 |
0.24% |
|
2017 |
1.3 |
1.3 |
1.3 |
1.0 |
1.3 |
2.6 |
0.7 |
1.3 |
1.48 |
0.23% |
|
15 year geometric average real salary inflation rate |
0.23% |
|||||||||
Note that the calculation of the geometric average uses unrounded figures in the intermediate steps.
|
Element |
Rate |
Weighting |
|
15 year weighted geometric average salary increase (2003-2017 inclusive) |
0.23% |
15/17 |
|
2018 weighted salary increase |
0.76% |
1/17 |
|
2019 weighted salary increase |
0.00% |
1/17 |
|
Weighted average salary increase assumption |
0.24% |
Source: Salary Adjustments and inflation rates drawn from historic annual adjustments applied at the Co-ordinated Organisations.