Assuming the worst to get the best
Judge’s comment: “In 2014, Pension Fund SBB continued on its very positive path to implement a sound, strong and convincing investment philosophy”
Another excellent performance in 2014 lead to a record coverage ratio of 107.3% for Pension Fund SBB (PFSBB), the €13.6bn Swiss railways pension fund, which it attributes to an investment philosophy characterised by a liability driven process, strong investment governance, comprehensive monitoring and broad diversification.
During 2014 and 2015, PFSBB continued to develop the process it uses to systematically achieve its target asset allocation. This is based on the scheme’s strategic asset allocation that depends on the risk budget the trustees define and is split into two portfolios: a liability-hedging portfolio and a performance-seeking portfolio. The performance-seeking portfolio consists solely of equities and the level of overall exposure PFSBB can take in equities is determined by the level of its liabilities. At present, the risk budget allows for a 35% equity weighting.
Composed of a broadly diversified mix of assets and mandates, PFSBB’s target allocation was completely redesigned in 2014 and 2015, driven by the following key aspects:
- The development of its own return assumptions to maintain a consistent framework
- The target allocation must reflect solely the risk budget the trustees agree
- The expected return of the target allocation must be at least as high as the expected return of the strategic asset allocation
- The return assumptions behind the target allocation must be consistent with those of the risk budget determined by the scheme’s asset liability modelling.
A key element of the study was the development internally of a return assumptions model, which is based on a risk premium approach. These are the central drivers of SBB’s optimal allocation and understanding how these come about is crucial. Moreover, defining its own assumptions internally removes the need to discuss any return assumptions suggested by external consultants with the investment committee and thus enables the scheme to concentrate on the changes the asset liability modelling suggests for its target allocation.
PFSBB calculates its ultimate return target by adding its risk premium to the short-term interest rate. Also known as the return assumption, the fund then divides this into two. The first is a medium-term return assumption that covers a period of up to five years and the second is a long-term assumption that concerns five years or more.
To determine its target allocation, PFSBB uses a simple mean-variance approach that is enhanced by the Black-Litterman model, a mathematical model for portfolio allocation developed in 1990, The reason behind this approach is to enable the fund to better understand the economic drivers behind its returns and the target allocation it sets. Indeed, the fund believes that models that are perceived to be much more sophisticated may lead to results that are hard to interpret and that may subsequently form the subject of lengthy internal debate, especially given that the majority of the members on the investment committee are not investment professionals.
PFSBB currently relies on several different medium-term return assumptions and the corresponding target allocation based on different economic scenarios. These are the result of different economic views about the outlook for world growth, inflation and the evolution of the short-term interest rate in the medium and long term.
To adjust its current allocation to ensure it maintains its high funding level, PFSBB’s investment committee selects the scenario that best matches current market conditions and its liabilities profile. At present, the most likely scenario is one of low future growth. Since this is related to the medium-term assumptions, the fund does not consider the return potential from its investments to be sufficient over the next five years, triggering further tactical refinements of its risk budget to try to enhance its performance.
Pension Fund SBB
Founded in 1999
Hybrid corporate pension fund
- active: 28,828
- retirees: 26,552
- one year: 8.36%
- three years: 7.2%
- five years: 5.4%
- ten years: 4.2%
- Record funding level in 2014 of 107.3%
- Internal development of assumptions and risk budgeting
- Efficient tactical allocation model based on medium-term return assumptions
- Sulzer Vorsorgeeinrichtung
- Swiss Social Security Funds (AHV/IV/EO)
- Felix Kottman
- Stephan Lods
- Graziano Lusenti
- Mike McShee
- Murat Uenal