New lease of life for active asset allocation

In the 1990s, the benefits of active asset allocation (AAA) were called into question, mainly for two reasons.

First, both bond and equity returns were consistently positive over the period. Therefore, even if some portfolios experienced a negative relative performance (ie, compared to their strategic benchmark), the absolute performance was high enough for the investor to ignore the underperformance of his fund managers. Second, a new paradigm emerged during the second half of the 1990s: the business cycle is dead. Whereas strategic asset allocation (SAA) is mostly based on medium-to long-term economic outlook (economists usually use the terminology of potential growth), AAA depends more on short-term outlook (the business cycle). To cut a long story short, if the business cycle dies, so does AAA.

But when the bubble burst investors were reminded that periods with valuation gaps do exist and that tactical asset allocation (TAA) did make sense after all. At the same time, the 2000-2003 world inventory/investment adjustment proved that business cycle had obviously not disappeared. Conditions were in place for a reincarnation of AAA, both for the man in the street and for institutional investors. However, this reincarnation is a slow process.

Asset allocation

Explicitly or implicitly, individual investors require dynamic management of their asset allocation; it has to move over time with the evolution of their wealth, age and risk aversion. The bubble burst showed investors that they cannot manage this alone and that AAA is a specific job with a specific competence. In the US, individual investors’ appetite for asset allocation is clearly improving but supply is still evolving.

On the institutional investor side, the demand for TAA is still low. They probably still have to realise that when breaking down the risk budget (ie, the tracking error compared to the strategic benchmark) between the selection tracking error (the active risk devoted to management of assets within individual asset classes) and the allocation tracking error (the active risk devoted to the management of assets between asset classes), the latter is considerably higher than the former. The time for “benign neglect” on the asset allocation risk is over; investors cannot accept this implicit active risk anymore without managing it. Progressively, institutional investors are becoming aware of this necessity.

Rebalancing

Investors are faced with two options regarding the management of the risk allocation of their portfolio. The first is to implement a systematic rebalancing process and the second is to dedicate a specific part of the risk budget to AAA.

Implementing a systematic rebalancing policy explicitly means that after having decided the rule, the tracking error is directly taken on by investors. The most widespread rebalancing processes are based on a calendar rebalancing (monthly, quarterly or even yearly basis) and threshold rebalancing (return to strategic mix or allowed range). There is no unique solution when calibrating the rebalancing for an investor, as the optimal rebalancing policy depends on the portfolio’s structure, volatility/correlation assumptions, and most of all the investor’s risk tolerance and transaction costs he bears.

Beyond the systematic rebalancing, there is what some call intelligent rebalancing with a dedicated risk budget for timing the rebalancing. It is after all very close to the active management of TAA. It is just a matter of risk tolerance magnitude.

So, why it is optimal to implement AAA? First, it is obviously inefficient to stick to SAA particularly during periods of apparent valuation gaps or in some particular phases of the economic cycle (see chart one). In other words, rebalancing is definitely sub-optimal compared to a well-built AAA process.

Second, AAA is an additional and complementary source of return because the allocation active risk is said to be uncorrelated with market risk, so it adds very little to the overall portfolio risk. This simplified description has to be fairer. Active returns, defined as returns relative to a benchmark, are not necessarily completely uncorrelated with market risk. We commonly decompose active risk into that due to the impact of the benchmark return and what is left, called the residual risk. This residual risk is often called the pure active risk or the alpha, and is in theory uncorrelated with the market.

Asset management

On the asset management side of the story, AAA can be broken down in two separate sets of expertise: global products allocation which has more to do with architecture and the construction of well-diversified portfolios, and AAA which is faced with the market timing challenges.

Global products allocation requires the ability to assemble many asset classes – both benchmarked and alternatives, such as private equity, commodities, real estate and infrastructure – different regions and different strategy/products (for example, structured products to put in place convexity positions). Individual and medium-sized investors need this type of asset allocation expertise. It is probably best managed in an open architecture style. This open architecture framework is based on the idea that it is nigh impossible for an asset manager to have internally all the relevant products. So global product management consists of producing well-balanced and well-diversified portfolios based on a mixture of both internally and externally manufactured products. This requires sophisticated optimisation tools¹ and has notably two challenges. The first is to manage to combine management services and an advisory role. For instance, implementing an overlay programme involves the set up of a framework which is by necessity specific to each investor (currency management, benchmark choice, performance measure and dedicated budget). The second challenge is to link this active management to the dynamic strategic allocation².

The second part of the AAA job is the pure tactical asset allocation. It mainly consists of defining the market timing and implementing it via exchange-traded funds and derivatives (futures and options) contracts. Using such assets allows us to benefit from higher liquidity and lower transaction costs, but obviously requires a deep expertise in risk analysis.

In an efficient asset management organisation, those two different jobs in the AAA should be clearly distinguished. Typically, a global products/portfolio construction department is responsible for the transversal added value, whereas a TAA department will be in charge of the market timing added value (see chart two).

As time goes by and the markets regain pace, the still unconscious demand for AAA, both on the institutional and individual investors side is bound to become a conscious need. Facing this huge potential demand, supply is still nascent. Now in a reincarnation phase, AAA will probably not have to wait for long before entering the growth phase. Maturity is still far away.

Notes

¹ The more you integrate “alternative” asset classes, the less the normal distribution assumption is relevant. So the risk/return framework is no more suitable and an optimisation should notably take into account asymmetric and extreme event risks.

² Both in textbook and in practice, SAA is based on a medium/long-term economic and financial scenario, liability-driven matters and on the investor’s risk aversion. It was symptomatic to observe that “equilibrium” assumptions (return/volatility/correlation) were globally unchanged for instance both during the 1997-1999 bull market and in the 2000-2002 bear market. Nevertheless, everything being equal, it is a nonsense not to introduce valuations diagnosis in the SAA process. In other words, SAA can obviously not be the same if DJ EuroStoxx 50 stands the 5400 or 2000. Thus, SAA should be dynamic and in such conditions, the frontier between SAA and AAA is not obvious. It is a matter of time, but this concept may evolve in a changing environment.

Roland Lescure (right), deputy CIO in charge of active asset allocation --- Christophe Morel, head of tactical asset allocation