The Netherlands and the UK were among the first countries to establish pension funds. The oldest one in Holland, to my knowledge, is the Gist-Brocades Pension Fund founded in 1879.
In the first half of this century, the cash flow available for pension plans was mainly invested in Dutch government fixed-income securities. The fund manager’s sole concern was about interest rates, but their importance was limited because loans were kept until maturity and tactical asset allocation (TAA) was certainly not an issue.
In the 1950s, or perhaps even before, the more sophisticated funds started to invest in Dutch equities and – within the past 30 years – in domestic property in the Netherlands. At this point, funds had to consider what had to be allocated to these asset classes. Was this the dawning of TAA in Holland?
Currently, pension funds are confronted with an endless list of investment opportunities, from alternative investments to emerging market debt products and even arbitrage techniques or alpha transport strategies.
With the expected growth of the pensions industry and the likelihood that returns will probably be lower in the near future, the appetite for more risky, though risk-controlled, investments is inevitable. With this in mind, strategic and tactical asset allocation is an absolute must for pension funds.
The majority of Dutch funds have extensive experience with asset/liability modelling (ALM) the top of the investment pyramid. Pension funds have hired external consultants to assist them with ALM. Generally consultants and pension plans have relied on historical returns from mainly Dutch equities and fixed-income securities or, in some instances, the use of historical returns from broad asset classes, as input for these models.
Others were more advanced and refined their approach by including leverage, currency hedging and also derivatives. However, the liability side of the model has always attracted more attention than the asset side, mainly because the input was coming from actuaries. Some funds even hired two consultants: one for liabilities and the other to consult on the asset side.
Since the first generation of asset/ liability models, pension funds today emphasise the investment segment of the model. The funds are considering regional, country and style benchmarks and even sector returns. The development of tactical asset allocation models would be the next logical step. Some funds have developed a model themselves, but they do not have hired consultants to assist them. With so much effort dedicated to ALM, one would expect it to be neither complicated nor expensive to develop a global tactical asset allocation (GTAA) model.
At a recent GTAA seminar at Palladyne Asset Management in Amsterdam, Professor Guus Boender at Ortec Consultants, a highly respected ALM specialist, said he is investigating whether GTAA is a logical extension of the ALM model. TAA is certainly not a new phenomenon, but it has not received as much attention as ALM or strategic asset allocation, for example. In addition, it has rarely been a topic of discourse in the academic world.
In general, after ALM more research was done on other subjects, such as internal versus external management, passive versus active management, the selection of external managers and performance measurement.
After defining a strategic asset allocation, one of the more difficult challenges facing pension plans is how to monitor the strategic allocations and when to intervene to make reallocations as plans deviate from the strategic allocations due to performance.
Usually, pension funds write annual investment plans, in which they define how much money would be allocated and which asset class will be overweighted or underweighted versus the benchmark.
A general strategy is to make investments in a range of values around the ALM model (the normal benchmark). Some funds rebalance their portfolios more frequently than once a year, although the majority make adjustments annually.
Through this process, the pension funds have realised that little value is being added to the overall return. Some funds have hired external managers to run a country-specific TAA model for them, to overweight or underweight bonds versus equities or cash; others have been looking for managers that really can offer global TAA strategies.
My own experience is that there are a handful of managers who can run a GTAA model, and research has demonstrated that only very few can beat the benchmark consistently.
A more recent development is the so called ‘market-sensitive intervention’ method where capital is allocated to those managers, internally or externally, based on the excess return they expect to deliver. This technique has been developed recently and therefore results are only available for a very limited period and the normal benchmark, the outcome of the ALM model, is the basis of the discussion. Additionally, the CIO will ascertain from the board how much risk is acceptable for the pension fund in terms of tracking error. With this in mind, one can allocate capital or risk based on expected contributions to return and risk, thereby achieving a higher return while maintaining a desired level of aggregate risk. Diversification among styles, portfolios and managers will reduce the fund’s overall risk level.
Another technique recently introduced to pension plans is the opportunity to transport alpha from one asset class and/or manager to increase returns. Via long-short strategies, or other arbitrage techniques, institutional investors are attempting to add value to the return of their total portfolio. The investment environment is changing so rapidly and the endless list of investment opportunities is so long that it seems nearly impossible for pension funds to identify and investigate every new technique.
There is still no concrete evidence that GTAA will add value to the overall return of an institutional investor or that it diminishes risk. Consequently, in my opinion, further investigation is absolutely necessary.
Lou Ten Cate is head of Investment Management Factory in Amsterdam

Topics