The client’s specific objectives, constraints, and associated risk factors are simply not taken into account in the design of the optimal allocation. By Professor Lionel Martellini, PhD, scientific director, Edhec-Risk Institute.
As a result, competition to improve existing client relationships and provide new tools to improve advisor effectiveness is increasing. While the private banking industry is relatively well equipped on the tax planning side, the software packages used on the financial simulation side typically suffer from significant limitations and cannot satisfy the needs of a sophisticated clientele.
Most existing financial software packages used by private bankers to generate asset allocation recommendations rely on single-period mean-variance asset-portfolio optimisation, which cannot yield a proper strategic allocation for at least two reasons. For one, optimisation parameters (expected returns, volatilities and correlations) are defined as constant across time. This is contradicted by empirical observation and does not make it possible to take into account the length of the investment horizon. More importantly perhaps, liability constraints and risk factors affecting them, such as inflation-risk on targeted spending, are neither modelled nor explicitly taken into account in the portfolio construction process.
Overall, dealing with a private client usually leads to a detailed analysis of the client’s objectives, constraints, and risk-aversion parameters, sometimes on the basis of rather sophisticated approaches. Yet it is striking that once this information has been collected, and sometimes formalised, very little is done in terms of customising a portfolio solution to the specific needs of the client. In general, the approach consists of providing several profiles, expressed in terms of volatility or drawdown; in some instances a distinction in how the capital will eventually be accessed (annuities or lump-sum payment) is made, but the client’s specific objectives, constraints, and associated risk factors are simply not taken into account in the design of the optimal allocation. While some industry players have recently developed planning tools that model assets in a multi-period stochastic framework, asset-liability matching for individuals remains an area for exploration.
A recent paper, Asset-Liability Management in Private Wealth Management, produced by Edhec-Risk as part of the private ALM research chair in partnership with Ortec Finance, sheds light on the ways new forms of welfare-improving financial innovation which were inspired by the use of asset-liability management techniques but originally developed for institutional money management, can be used in private wealth management.
Asset-liability management (ALM) refers to the adaptation of the portfolio management process to the presence of constraints relating to the commitments represented by the investor’s liabilities. We argue that suitable extensions of portfolio optimisation techniques used by institutional investors could be transposed to private wealth management.
These techniques have been engineered to incorporate an investor’s specific constraints, objectives and horizon (the liability value) in the portfolio construction process. While ours is a fairly stylised model, and effects such as taxes or mortality risk are not explicitly taken into account at this stage, we believe it is a significant first normative step towards a better understanding of private wealth management decisions.
Our paper can be regarded as an attempt to take a first step towards a rational framework for private investors’ financial decisions that extends standard portfolio optimisation techniques by recognising that the aforementioned factors seriously affect the optimal allocation decision. We show that a significant fraction of the complexity of optimal asset allocation decisions for private investors can be captured through the introduction of a single additional state variable, the liability value. In the framework of private wealth management, ‘liabilities’ encompasses any commitment or spending objective, which is usually self-imposed.
For example, an investor committed to a real estate acquisition will perceive such an expense as a future commitment or soft liability for which money should be available. Overall, it is not the performance of a particular fund or that of a given asset class that will be the determinant in the ability to meet a private investor’s expectations. Satisfaction of the investor’s long-term objectives is fundamentally dependent on an ALM exercise. What will prove decisive is the ability to design an asset allocation programme that depends on the particular risks to which the investor is exposed.
Similarly, the concept of a risk-free asset depends on the investor’s time-horizon and objectives. Hence, a five-year zero-coupon Treasury bond will not prove a perfectly safe investment for a private investor interested in a real estate acquisition in five years. The actual risk-free asset in this context (which we call below the liability-hedging portfolio) would instead be an asset perfectly correlated with real estate prices.
An investor whose objective is the acquisition of property would accept low and even negative returns in situations when real estate prices fall significantly, but will not be satisfied with relatively high returns if these returns do not match dramatic increases in real estate prices. In such circumstances, a long-term investment in stocks and bonds, with a performance weakly correlated with real estate prices, would not be the right investment.
So, the first benefit of the ALM approach is its impact on the menu of asset classes, with a focus on including an asset that exhibits the highest possible correlation with the liability portfolio.
Early papers on long-term financial decisions, which our paper is related to, highlight important aspects of life-cycle investing, including the usefulness of real bonds for inflation-hedging purposes. On the other hand, they mostly abstract away from some of the key complexities of private financial decisions. A large number of more recent papers have subsequently focused on integrating salient features of private wealth management including the impact of human capital, illiquid real estate allocation or borrowing constraints on optimal allocation decisions.
A different perspective
However, these papers fail to integrate a key dimension of private wealth management – that investment decisions should be designed to help investors achieve predetermined objectives. We argue that the literature on household finance has mostly taken an asset management perspective, as opposed to an asset-liability management perspective.
Our paper can be seen as an attempt to merge two somewhat separate strands of the literature, that is, the literature on long-term financial decisions for private investors, which has focused mostly on an asset-only perspective, and the literature on asset-liability management decisions, which have been analysed mostly from an institutional perspective.
We do so by casting the long-horizon lifecycle investment problem in an asset-liability management framework suitable for the private wealth management context. This allows us to show that pursuing an asset-only strategy usually involves a substantial opportunity cost. Taking an ALM approach leads to defining risk and return relative to a liability portfolio, a critical improvement on asset-only asset allocation models that fail to account for the presence of investment and/or consumption goals and objectives, such as preparing for retirement or for a real estate acquisition. As a result, taking an ALM approach leads to a focus on the liability-hedging properties of various asset classes, a focus that would, by definition, be absent from an asset-only perspective.
(Article based on research carried out within the Edhec-Risk/Ortec Finance ‘Private ALM’ research chair)
©2009 Funds Europe