Contiguous Swiss Pension Fund Benchmarks From 31.12.1984
Since occupational pension funds became mandatory at the end of 1984, the latitude of investment freedom for Swiss pension funds has been revised on several occasions. With each ammendment, Banque Pictet, the leading publisher of synthetic pension fund performance benchmarks in Switzerland, have designed numerous benchmark indices to reflect the universe of permitted investments.
Today, Pictet calculate and publish no fewer than nine pension fund indices daily (three index ‘families’ of three risk-tiers each), yet not a single one of them is representative across the entire history. Readers are referred to the link provided for details of these benchmark compositions. Regrettably, Pictet have removed all benchmark statistics prior to December 31, 1999 from public access. In order to address the resulting twilight thus cast over all genuinely long term analytical efforts regarding pension fund’s returns, Agathos compute three contiguous indices, combining archived benchmark data with currently published series.
The resulting benchmark indices, designated with the letter ‘C’, are shown in Graph-1 below, together with information on their ancestry. A monthly analysis of the ongoing performance of these three contiguous benchmark indices is found here.
In addition to the bird’s eye view of these indices, the tables below show selected metrics, giving a nunemerical profile. The first table relates to the full history since 31.12.1984, the second shows identical statistics, but only since 31.12.1999. A comparison across both tables illustrates some of the profound changes challenging Swiss pension funds, in part due to financial market patterns, in part arising from the greater freedom of choice given to pension fund managers.
1984 to 31.12.2019
Please note that all three indices reflect one and the same Pictet benchmark (Pictet LPP-93) prior to 31.12.1999.
While back data for Pictet’s ‘2000’ family of indices is (was) available as far back as 31.12.1984, the asset class weightings within in these indices were not yet permitted, making such data rather theoretical, if outright misleading until after 31.12.1999.
The risk/return profiles of pension fund benchmarks in all risk-brackets have undergone dramatic chances with the end of ‘Irrational Exuberance’. A modest increase in observed risk coincided with lower investment returns from windfall.
Certainly a testing time for Swiss pension funds, or any similar group of institutional investors operating with complex and slow decision making processes, and/or fragile decision-making methodologies.
Since the change of the millennium, pension funds have sought refuge in ever more benchmark-driven investment philosophies, embracing the full impact of deteriorating risk/reward ratios, not realising the inherent hazard of having increased risk exposure (modified investment guidelines), without adequate risk monitoring in place.
1999 to 31.12.2019
Across the twenty years since the end of 1999, overall performance from static asset allocations (which is what pension fund performance benchmarks describe) has been fairly humble. More to the point, the massive swings of performance of high risk and low risk assets have generated very similar long term results. It is not that higher returns from financial markets were impossible to achieve. Simply put, they needed to be earned by means of professional, active management and did not drop out from a magic cornucopia. The performance comparison across the three contiguous benchmarks would suggest that pension funds could have generated higher returns with less risk than they actually have.
The line graph below shows the emerging congruency of the Agathos Pension Fund Composite Index and all three risk-tier benchmark indices, in the form of trailing values for r-squared, calculated across trails of 120 months. More than ever before do pension funds seem to replicate higher, rather than lower risk benchmarks. Congruency with the middle risk tier is at a record high (r-squared = 97.2%), while congruency with the upper risk tier is only moderately behind (95.9%). The readings validate Agathos’ initiative to calculate contiguous benchmarks.
Even considering that any pension fund aggregate index will, by its nature average things out, such values for r-squared appear extremely high, suggesting an extreme and unquestioning bias among pension funds to effectively clone a benchmark index, in lieu of ‘active’ asset management, thus accepting the full risk inherent in benchmarks, regardless of how mature an investment cycle may have grown. Passive investing and risk-management are mutually exclusive and the high degree of congruency to benchmarks illustrated here make it unlikely that pension funds will secure their good fortunes in time. The dangers of this neglect of investment risk will become painfully evident with the next bear market. Swiss pension funds run the risk of seeing the recovery of their reserve capital melt away just as suddenly as happened already twice since the introduction of mandatory occupational pension schemes in 1984.
The two scatter diagrams further down show 120 monthly rates of change since 31.12.2009 as well as details of the regression equations.
The table below illustrates congruency between all benchmark indices, the three original benchmarks of Pictet’s ‘2000-Family’, and the corresponding contiguous indices calculated by Agathos, grouped by risk-tiers.