Asset managers need to be primed for all eventualities. Our Global Chief Investment Of ficer Christian Nolting explains how strategic asset allocation helps achieve this.
You can read this article in Italian here.
Karl Popper said: “Our knowledge can only be finite, while our ignorance must necessarily be infinite.” If we already knew how investments would perform in future, there would be no need for strategy at all. But since we do not have a crystal ball, asset managers need to be primed for all eventualities. Strategic asset allocation (SAA) can help achieve this by following a robust approach.
Why the classical approach falls short
We understand a portfolio to mean a combination of different investments, comprising equities, bonds, cash and other asset classes. An optimal asset mix should maximise an investor’s overall return for a given level of volatility risk (optimum risk/return). To construct a portfolio, certain assumptions need to be made for each asset class. These assumptions are based on forecast investment parameters, which are used to determine the portfolio composition: expected returns, expected risk (volatility) and correlations. According to classical portfolio theory (Harry M. Markowitz), these parameters make it possible to determine the optimal weightings (or the share) of the individual assets in a portfolio, creating an “efficient” portfolio. However – and this is often forgotten – portfolio construction will only be optimal if all parameters behave exactly as forecast.
The first problem is the potential for over-optimistic assumptions in the classical approach. This unfounded optimism is due to failure to take account of particularly extreme, unusual market events (e.g. the tulip mania of 1637, the Great Depression of 1929, the 1973 oil crisis, the dot-com bubble in 2000, the 2008 financial crisis, or the Coronavirus market correction in 2020) that have a major negative impact on returns. In reality, these events occur far more frequently than forecast. The risk for portfolios that do not take this into account is much higher than the risk for those that do. In practice, this means that the equity allocation of portfolios that do not factor in these extreme risks is higher and so the risk exposure is greater. In short, the real risk is underestimated. Secondly, it is not possible to reliably predict the future, since no model in the world can fully represent reality and there is no such thing as perfect information, particularly when it comes to the capital markets. Models are merely tools that give us an approximate simulation of reality. How this uncertainty with regard to future events is dealt with thus plays a crucial role in portfolio construction.
The aim is robust positioning
By avoiding unnecessary market timing – timing the purchase and sale of equities (etc.) based on predicted market movements – SAA can reduce uncertainty and with it the likelihood that returns will suffer. To achieve the required robustness, the uncertainty of each individual parameter needs to be taken into account and destabilising positions avoided when constructing the portfolio. Robustness means that the portfolio has lower sensitivity to market movements that are less favourable than the developments assumed during portfolio construction. A robust SAA takes better account of the information and the level of uncertainty for the characteristics of each parameter in the portfolio modelling than a conventionally optimised portfolio.
Alongside expected returns and volatility, observing the correlation between the individual assets is a key element of SAA. Correlation is a statistical measure of how two assets move in relation to each other (i.e. the more A rises, the more B rises or falls). In classical portfolio design, correlations between assets are considered a given and are based on fixed assumptions. However, correlations are not stable over time. Consequently, robust SAA also takes into account changes in correlations. The aim is to ensure the portfolio construction remains intact, even if correlations become more volatile, in order to guarantee the desired risk/return over the long term. This reduces the risk of particularly poor relative performance.
Why robustness is crucial for SAA
Unfortunately, we cannot look into the future. However, we can get closer to reality. We are in a position to analyse the impact of the uncertainties described and use our findings to model a robust SAA that is able to deliver a realistic risk-adjusted performance, while avoiding over-reliance on parameters that are particularly uncertain. Modelling increases the chance of the SAA systematically achieving the investment objectives. Consequently, a systematically robust SAA is especially well suited to long investment horizons, as robustness beats efficiency over the long term.