Council decisions on personnel policy and social responsibility: a harmony to be restored



As explained in a previous article (Echo 195), in line with the Host State agreement between the Swiss Confederation and the Organization, the CERN Pension Fund must guarantee a pension insurance to its beneficiaries at least equivalent to that available in Switzerland. As the CERN Pension Fund is a capitalized defined benefit scheme, it is important to verify that the Fund is capable of guaranteeing the agreed benefits until the cessation of the rights of the last beneficiary. Therefore, the financial health of the Fund is reviewed at least once every three years by the Actuary, who, under the supervision of the Actuarial and Technical Commitee, must draw up an Actuarial Report that must be approved by the Pension Fund Governing Board (PFGB) before presentation to Finance Committee and Council.

In this article we look at various important changes in the pension parameters that were decided by Council in the framework of changes in the Organization’s human resources’ personnel policy and what effect they had on the Fund’s long-term financial status.

The early years

Since its creation in 1955 and until 1976 the Fund’s main pension parameters were the following:

  • contribution rate of 21% of basic salary (staff 1/3, Organization 2/3);
  • assumption of 3.5% net capital return;
  • retirement age: 65 years for men, 62 years for women;  
  • maximun pension : 60% of average salary of the last three years of service (maximum obtained after 30 years of service until 1967, 33 years of service until 1976);
  • early retirement leads to a corresponding reduction of pension entitlements.


The first revision (1967) of the Pension Rules introduced the following changes:

1. Adjustment of pensions for inflation to be borne exclusively by the Pension Fund;

2. Pension calculated as a percentage of the last insured salary (prior to 1967, increases in salary after the age of"61 were not taken into account in the pension calculation);

3. Termination of extraordinary contributions for increase in insured salary after the age of 61.

After the Oil price shock in 1973, which led to capital returns of –15% and –20%, the 1974 Actuarial study showed that with an annual net return of 2.5%, 4% salary and 3.5% pension adjustments, the contribution rate should be at 27% of basic salary (CERN/FC/1734). Nevertheless Council decided to leave the contribution rate unchanged at 21% (Fig. 1).

The second revision (1976) of the Pension Rules introduced the following changes:

1. Retirement age: 65 years;
2. Increase maximum pension from 60% to 70% of last basic salary after 35 years of service;


3. No reduction of pension entitlements when retiring at age 60;


4. Finance Committee does not recommend an increase of the 21% contribution rate (Fig. 1) despite negative capital returns in 1973 and 1974 and important increases in the  pension entitlements (CERN/1211);
5. Council suppresses the guarantee of capital return if the       capital return is less than the actuarial parameter (this guarantee had never been applied).

Fig. 1: Historical membership and contribution rate

The third revision (1981 to 1985) of the Pension Rules introduced the following changes:

1. RESCO working group: CERN pensions are lower than those of comparable European institutions due to taxation and family allowances (CERN/1349);
2. Increase in pension entitlements by introduction of reference salary to replace the previous insured salary, in particular to compensate for taxation effects (1986);


3. Introduction of allowances and complementary pensions (1982).


In view of point 1 above, Council decided to partially align CERN pensions in four phases with those of the Coordinated Organizations. Over the period 1981−1985, pensions were increased by an average of 23.3% with respect to the situation prevailing prior to 1981, without taking account of allowances (point 3). Retroactive payments were requested from active and retired staff as well as the Organization (without the total equivalent of the cumulative capital return) in order to benefit from the new pension entitlements. Proposals by CERN management to change to a budgetized system were refused by Member States in 1980, with a proposal to create a debt to the Pension Fund instead.

These 1967, 1976, and 1981 revisions were not accompanied by sufficient increases in funding. The effect in 2010 of this underfunding for a given individual measure as calculated by the actuary is indicated inside a box. This shortfall in funding was pointed out in various actuarial studies, for instance in 1983 (CERN/FC/2617), where the Actuary expressed the opinion that contributions should increase in view of the inflation rate and the pension complements approved by Council. Therefore, following the 1983 actuarial study, the Organization agreed to pay an additional 0.5% of reference salaries to the Pension Fund from 1986 onwards (Fig. 1).

In 1987 reduction factors below the age of 65 years were reintroduced for all new recruits with effect from 1 July 1987.
(reversal of measure 3 of 1976 revision of Pension Rules).

The CERN Pension Fund matures

Following the 1989 and 1992 actuarial studies, the Actuary proposed to reduce the real rate of return on investments used for the actuarial calculations from 3.5% to 3% and to progressively increase the contribution rate from 23.5% to 30.0% by 1995 in order to partly compensate the impact of previous decisions to introduce substantial improvements in benefits, with a view to bringing the Fund back into technical balance. Therefore, over the period 1990−1995 contributions increased in various steps to 30% of reference salary (Fig. 1). This was all the more necessary since the number of retirees drastically increased during the 1990s, together with the outflow which overtook contributions in the later years of the decade (Fig. 2). The CERN Pension fund had reached maturity and depended now much more on the return on its capital.

During the earlier years, the favourable levels of return achieved by the Fund in association with the measures taken in 1987 and 1992 partly concealed the Pension Fund’s underlying structural problems, such as: the real cost of previous measures taken to improve benefits and the changes in actuarial accounting practice and governance introduced to reflect a more realistic approach to the calculation of the Fund’s liabilities (i.e., the reduction of the discount rate from 6% to 4.5%).

Fig. 2: Contributions (blue), benefits (red), and contribution rates (green)

The start of a new millenium

Although following the 2001 actuarial review (CERN/2440) the results review showed a reassuring picture for the health of the Fund, the PFGB nevertheless proposed three changes in the pension parameters:

Return on investment: reduction of technical rate from      6% to 5% for the next actuarial review;
- Personnel policy: compensation of Fund for reduction in CERN staff complement (CERN/2441);
- Safety margin: introduction of a final funding ratio target  for the end of the period of between 100 and 125%, in order to take account of pension indexation  needs.

At the beginning of 2000, the structural problems of the Fund were somewhat obscured by the good results obtained during the 1990s, with an annual average return above 7%.
There was a major correction in the financial markets in 2001 and 2002, and despite a diversification of investments, a policy of prudence and generally satisfactory fund manager performances, the Fund could not be immune to this dip, and the 2004 actuarial review had to face this totally different situation (Fig. 3). In order to restore full funding by 2033, the PFGB in July 2005 took a certain number of measures, including the following:

- introduction of partial dynamic indexation of pensions with     maximal loss of purchasing power of 8%;
- change in technical interest (discount) rate (from 5.5% to 4.5%);
- increase of contributions to 30.88% (1/3 personnel, 2/3 Organization).

 Fig. 3: Annual and average performance of the Fund

The 2008 Crash in financial markets and its sequels

The 2007 actuarial review (CERN/2775) showed an important improvement in the Fund’s financial position. This favourable trend was essentially due to two factors: the general economic climate, which yielded a good return on investments due to the increase in financial markets from 2004 to 2006 (an annual average of 8.93%) and the change in the way the current value of future liabilities towards active members was calculated.

At the beginning of 2008, the position of the Pension Fund looked reassuring. Progress had been made on the financial side and by introducing a new set of actuarial parameters more in line with the current economic situation and its likely medium- or long-term evolution, and with those used by pension funds similar to CERN’s (i.e., Swiss pension funds or those of international organizations based in Switzerland). Moreover, notwithstanding the 2002 “dotcom” collapse, the Fund was on a positive track with an annual average return from 2001 to 2007 of 4.5%, exactly the new discount rate. Over the longer term, the Fund’s performance compared favourably with its Swiss peers.

The year 2008 was one of the worst years for equity and credit markets since the 1930s. The Fund’s performance was the worst of its history (-19.3%), which translated into an investment shortfall of around 900 MCHF, mainly due to its exposure to the equity markets.

The genesis of the 2010 balanced package of measures

In March 2008 Council set up a Working Group (WG2) to analyse measures to achieve full funding. This study produced a report (CERN/2897) which concluded that the investment strategy alone was not sufficient to enable the Fund’s recovery since the Fund had to face a deficit of around 2,100 MCHF (just under half generated by the 2008 crash). Without recovery measures the deficit of the Fund would increase by some 80 MCHF per year and the funding ratio would decline to 15.4% in 2033, with the Fund running out of money by 2038.

In its report WG2 studied a series of measures to reach full funding:

1) immediate injection by the Member States of 2,100 MCHF: rejected as “impractical”;
2) reduction in the outflows of the Fund:
     a) changes affecting current beneficiaries
         (under-indexation without constraint)
         and members (increasing retirement age,  
         reducing annual accrual rate of 2%, etc.):
         limited by  legal constraints and not further
     b) opening a completely new scheme for new
         members (defined contribution):
         abandoned due to its high cost;
     c) changing benefits for future members of the Fund within the existing scheme:
         implemented in 2010 package of measures;
3) increase in the intake of the Fund:
     a) increase in the regular contribution rate:
         implemented in 2010 package of measures;
     b) injection of capital by special contributions of the Member State:
          i)  shared equally by all Member States:
              implemented in 2010 package of measures;
          ii) paid by Member States which derive financial
              benefit  from taxation of pensions
             (via internal taxation of pensions by CERN or  retrocession of (part of)

              the taxes levied on pensions by Member States): rejected.      


This analysis showed that WG2 realized that a large part of the deficit had been caused by the unwillingness of Council to take timely measures to fully compensate the Fund for personnel policies introduced in the past (the individual cost of a given measure is indicated inside a box in the text which is the estimate quoted in document “Pension Fund — Analysis of Structural Causes” (CERN/2895) and corresponds to the effect on liabilities should the individual decision be reversed). Taking into account the WG2 report and the 2010 actuarial study (CERN/2948) Council decided in December 2010 (CERN/2947), June 2011 (CERN/2972) and March 2012 (CERN/3010) on a package of measures towards restoring full funding of the Fund. We recall
the main parameters of these measures:

1) contributions for pre-2012 members: increase from 30.88% to 34% (1/3 personnel, 2/3 Organization);
2) pensions of beneficiaries: freeze until individual accumulated loss of purchasing power reaches 8%;

3) special contributions of Organizations: (max 30 years or until full funding reached): CERN: 60 MCHF/year; ESO: 1.3 MCHF/year;
4) new pension plan for members recruited as of 2012:


Figure 4 shows the effect of each of these four measures on the full funding on the 2041 horizon (calculation based on 2010 actuarial study). The first three measures contribute towards full funding, while the newly recruited staff should make sure that the funding level remains above 100% after full funding is reached.

Fig. 4: Towards full funding with the 2010 package of measures (CERN/2972)


In this article we showed how changes in the pension parameters, decided by Council in the framework of the evolution of the Organization’s human resources’ personnel policy had long-lasting and sometimes drastic effects on the long-term financial health of the Fund. Of course, these past decisions were made in the broader context prevailing at the times, including CERN’s priorities with regard to human resource management. Nevertheless, the right balance between the role of CERN as a scientific organization which needs to build and run accelerators and physics experiments, and as a socially responsible employer, which has to honour its social insurance commitments to its staff, was sometimes difficult to find.

The large loss of around 900 MCHF in the 2008 crisis is of course not directly to blame on CERN, but the structural underfunding since several decades aggravated its effect and made the corrective measures introduced in 2010 more painful than they could have been. Therefore, it is all the more unacceptable that some delegations, after only three years, put into question unilaterally their commitment expressed in the June 2011 Resolution, to contribute to restoring the Fund to full funding on the 2041 horizon.

by Staff Association