The increasing integration of the world’s economies and financial markets, and the sharp drop in share prices are now forcing investors to be constantly on the lookout for new sources of diversification and performance relays in strategic allocation. According to a range of studies undertaken in the US and Europe, the performance of a fund over time depends more than 90% on its strategic asset allocation.

Diversification is at the heart of the challenge for investors
Today, when investors look for ways to diversify their portfolios, they may think of adding corporate bonds to gilts or increasing their international exposure to equity. However, the increasing integration of the large economic zones and the world’s financial markets reduces the impact of this type of diversification.
The challenge facing the investor is to find new, less-correlated sources of added value that take into account his objectives and constraints, whether at the financial or regulatory level. AXA Investment Managers (AXA IM) has developed a strategic asset allocation process which optimises the expected return per unit of risk taken. We use a broad universe of asset classes including investment-grade, high-yield and emerging market bonds; large-cap, small-cap and private equity; and property assets. Our risk model specifically takes into account the risks of market crashes. Finally, we can add non-correlated strategies based on hedge funds to improve the risk /return profile ratio of the portfolio.

How to evaluate and include alternative asset classes in overall asset allocation
Asset classes, such as private equity, property and hedge fund strategies, offer special risk/return profiles that make them difficult to model and to include in a traditional bond/equity portfolio.
AXA IM has developed a tool for asset allocation, which enables it to:
q take better account of the risk factors of financial assets and the risk of market crashes;
q better understand an investor’s degree of risk aversion, through understanding the distribution of the returns of potential portfolios; together with the possibility of modelling the client’s liability requirement;
q simulate better the different investment strategies used by investors.

AXA IM’s Asset Allocation Model:
AXA IM’s model stems from the observation that the ‘mean-variance’ approach relies on strong assumptions that are often not verified in practice.
q Returns do not follow normal distributions. Nowadays, it is admitted that the distribution of many asset returns does not follow a normal distribution. In general, the distribution is skewed (not symmetrical) and fat-tailed (leptokurtosis). The knowledge of wealth distribution over time and the use of indicators, such as Value at Risk (VaR), prove fundamental to guide the investor’s decisions.
q The parameters are not stable over time. The volatility and correlation of assets are well known to move over time. In particular, a rise in volatility and an increase in correlations between assets often characterise crisis periods in the market. This phenomenon is taken into account via a specific risk inherent to each asset class and a systemic risk that characterises the macro-economic risk.
q Asset management costs exist. They are certain and quantifiable, and, furthermore, a correlation between risk and transaction costs often occurs.
Determining the parameters of risk and return is a crucial step in the process of portfolio building. We do not determine risk premiums solely on an historical basis: we use the outlook for the economic cycle anticipated by AXA IM’s Investment Strategy team.
Risk parameters are assessed under three different market conditions:
q normal conditions (the markets are in equilibrium);
q specific shock conditions (inherent to an asset class, a particular security, a sector, etc); and
q systemic shock conditions (macro-economic events that affect all asset classes at the same time).
To determine whether portfolios are efficient, we use Monte Carlo simulations to generate the assets’ behaviour. 10,000 scenarios are tested for each allocation over the investment period being considered. We then have the distribution of returns for each allocation, which lets us calculate, for example, the probability of obtaining a given return over a particular period.

Building the ‘ideal’ portfolio
The process of developing asset allocation takes place in two major steps:
q Determining optimal strategic asset allocation. The starting point is an analysis of the implicit or explicit liabilities of the investor. After characterising the investor’s risk profile and taking into account regulatory, financial or client constraints – we seek to determine the portfolio that will maximise the expected return.
The method is to use the model to test all possible asset allocations and examine the one that offers the best risk/return payoff. According to each investor’s specific constraints, we test all the possible allocations that include the alternative asset classes and examine the one that offers the best balance between risk and return.
q Improving return at a given level of risk. Based on a risk budget allocated by the investor, the balanced fund manager can use three sources of ‘alpha’: tactical allocation of assets between bond and equities; diversification of investment manager style (growth, value, quantitative, judgemental, etc); and use of non-correlated strategies based on hedge funds.
The goal is to allocate the different ‘alpha’ sources to maximise the portfolio’s information ratio for a given level of tracking error (or maximum downside risk) specified by the investor.

Advice on strategic asset allocation: an extra service to our clients
The performance of equity markets over the last 10 years could lead someone to think that strategic asset allocation can be reduced to the question, “What percentage was invested in equities?” The recent period highlights the importance of strategic asset allocation at the centre of investment decisions. It also highlights the need to take into account the special risks associated with financial assets (brutal crash, long-lasting underperformance). Lastly, it shows the importance of managing and diversifying risks.
To manage the risk of a balanced fund, you have to analyse the underlying assets first. Investment management companies, like AXA IM, which have a long experience managing the broadest palette of asset classes and different sets of expertise, do this on a daily basis. The next step is to develop a model and a methodology to blend assets so as to deliver an optimal asset allocation. Each investor should seek the best compromise between reward and risk thanks to the effects of diversification.
AXA IM offers its clients a broad range of financial services and products. A rigorous and flexible framework for strategic asset allocation decisions is a cornerstone of our service to investors, whether they wish to optimise the return/risk profile of their traditional portfolios or start diversifying towards property, private equity, or alternative asset management.