Strategic Asset Allocation: Practical Considerations for Alternative Investments

By Andreas Rothacher, CFA, CAIA, Head of Investment Research, Complementa AG 
& Thomas Breitenmoser, CFA, CAIA, Head of Investment-Consulting, Complementa AG

Allocations, backdrop and historical development

The assets under management in alternative assets such as private equity, infrastructure, and private debt have increased significantly over the past decade. This is reflected in institutional portfolios, which have higher shares of alternative assets. In addition, alternative assets are becoming increasingly prominent in private wealth portfolios[1]. The access and availability of subcategories have increased significantly over the past two decades. Therefore, alternative assets are now one or multiple components besides traditional assets in a multi-asset context. As alternative assets have become more mainstream, asset allocators and other investment professionals must think about ways to integrate them into their asset allocation process and modelling. To achieve this, various models and approaches have been developed. The most famous of these models include the Yale/endowment model, the Canadian model and the Total Portfolio Approach by the CAIA Association[2].

This article highlights some of the key considerations when it comes to alternative assets in a multi-asset context. Our remarks and concepts are general in nature and hopefully, will be useful for different types of allocators, regardless of the approach or model they follow. Our key considerations and comments are based on our experience with asset allocation studies that we conduct for institutional clients, as well as on our conversations with senior investment professionals at large Swiss and German pension funds.

Strategic asset allocation and modelling

Strategic asset allocation describes the long-term planning and definition of target weights for different asset classes. For many institutional investors, choosing an investment strategy will be the most important decision they make. According to various studies, the asset allocation choice determines more than 80% of return volatility and return levels achieved[3]. Additional key investment decisions include tactical allocation and fund/security selection. Although we believe that these decisions are important and can potentially add value, research shows that a mis-specified strategic asset allocation cannot be corrected by good selection or market timing.

The strategic asset allocation stands at the beginning of setting up a new portfolio, but it is also conducted at two-to-five-year intervals for existing portfolios. The chosen asset allocation must comply with portfolio restrictions and be suitable to achieve investment goals. Investment goals and restrictions (e.g. return goals or the amount and structure of any liabilities) are taken as a given and may be defined by law, by a plan sponsor, or by the ultimate beneficial owner. In the context of ALM, decision makers must balance the long-term return potential of different strategy variants with the corresponding risk of short-term volatility in stress scenarios. Scenario analysis can support decision makers in this process.

Ideally, the strategy process for institutional investors starts with formulating investment beliefs. This can include the analysis of quantitative data as well as tapping into the experience of senior investment staff and investment committee members. The data and experiences should then be used in structured discussions to arrive at shared investment beliefs for the organisation. In general, the conceptual work should be done in house, while the implementation can be supported by external partners or an internal team.

In simple terms, asset allocation answers the question of which building blocks an investor should use in their portfolio construction. This simple question may have a complicated answer. On a conceptual level, it means that each component or asset class that is added to the investment strategy should benefit the portfolio in some way. This could include earning an additional yield, earning a different type of risk premium, improving portfolio diversification, or a combination of these elements. This rule of thumb applies to all asset classes or components, traditional and alternative assets. The CAIA Association calls this concept “Competition for Capital” in their Total Portfolio Approach[4]. To answer the above question, institutional investors conduct an ALM study or asset-only study. Traditionally, mean-variance optimization (as first described by H. Markowitz)[5] is used to derive a new asset allocation (SAA). However, when adding alternative assets, some challenges arise.

Unlike traditional assets, the investable alternatives universe may not be fully known or accessible. This is also reflected by the fact that there is often no representative benchmark available. These aspects make it difficult to determine the true market beta of alternative asset classes, especially private market assets. To create meaningful time series, allocators could, for example, use feedback from managers as inputs (e.g. historical data, return distributions) or listed benchmarks as proxies. Historical data provided by managers is likely to underestimate volatility, whereas a public market proxy will tend to overestimate volatility. The truth may lie somewhere in between. It is crucial that investors are aware of the drawback of each approach.

There is no universally accepted way of modelling different alternative asset classes. In our opinion, being systematic and consistent with the rest of the modelled investment universe is a key consideration. In addition, the ALM specialist (internal or external) should be transparent about the assumptions used and the methods applied when modelling a new strategy. Even though modelling is difficult, it can provide a sanity check for the investment beliefs and the investment thesis the organisation holds. The ALM study and related discussions should also include qualitative aspects.

Take illiquidity into consideration

A traditional mean-variance optimisation has an important disadvantage when it comes to illiquid alternative assets. Time series often show low volatility and underestimate the actual risk. Additionally, lagged time series result in lower cross-asset class correlations than non-lagged numbers and, therefore, overestimate the benefits of diversification. In an unconstrained optimisation, this could also lead to unrealistically high allocations in illiquid assets. Therefore, for real-world portfolios, it might make sense to add constraints to the optimisation. This could include guidelines like maximum allocations to certain asset classes or groups of asset classes. The illiquid nature of some alternative assets, such as private market investments, also means that illiquidity risk must be considered. The illiquidity aspect can be addressed by introducing a liquidity penalty into the optimisation function. This allows the cost of illiquidity and liquidity preferences to be incorporated. 

Since allocations in private market asset classes cannot be adjusted quickly, investors who include such asset classes should have an allocation plan for building, maintaining, and/or adjusting allocations. The investment team and managers should have a clear understanding of how much capital should be deployed in each vintage year. In a broader sense, this is an element of liquidity planning. This involves planning to ensure the ability to service capital calls and any other potential payments (e.g. pension pay-outs). Investors should position themselves to minimise the probability of becoming a forced seller. They should also remember that liquidity can dry up in times of market turbulence, and the redemption of investments may be subject to redemption fees, penalties, and gating.

Factor lens

As mentioned before, actual diversification can be overestimated because of lagged and infrequent time series. To address this, factor exposure should be considered as an additional measure of portfolio diversification. This approach should also allow greater transparency regarding underlying portfolio drivers. Factors can include equity, credit, interest rates, and real estate, for example. The factor sensitivity of an asset class can be determined via regression analysis or through a heuristic discussion among the investment team or investment consulting firm (ALM specialist). In any case, the results should be challenged and discussed. Investment professionals should see the factor lens as an additional, helpful tool, bearing in mind that it is also an approximation of the actual, unobservable factor exposure of asset classes. In addition, one should keep in mind that factor exposures may change over time.

Education

Before venturing into a new alternative asset class, it is recommended that the board, the members of the investment committee, and internal staff are educated about it. This includes education about return drivers as well as the corresponding risks. The goal of this education is to achieve a good understanding of potential new investments and to make sure that decision-makers and staff are comfortable with the asset class. When it comes to boards and committees, it is important to remember that their composition can change over time. Therefore, continuous education and refresher training is advisable. The long investment horizon of private market assets also implies that the board or investment committee responsible for selecting the investment may no longer be in place by the time the investment is fully paid back. Therefore, a new decision maker might inherit an existing portfolio or current decision makers may select investments for their successors.

Operational set-up and implementation

Once a strategic asset allocation has been defined, it must be implemented. For alternative assets, this means that investors must determine the way to gain exposure to an asset class (e.g. investment vehicle, segregated account, etc.). This goes hand in hand with the size of the investor, but also with the availability of existing solutions. Once these questions have been answered, investors should determine how to steer their position sizes. This includes, for example, a minimum investment size and maximum manager concentrations. It also includes vehicle concentrations as well as the share of an investment vehicle, if choosing anything but a segregated account or fund of one. For asset classes, regulatory bands should be established. Investors should also be aware that there can be a substantial implementation gap between the defined investment strategy and the actual implementation. The size of this gap depends on the underlying assumptions and time series used to model the specific alternative asset classes and the selected investment vehicles. As the product range and risk profiles in some asset classes can vary greatly, it is essential to diversify within asset classes and consider the role of each asset class within the portfolio.

Investors should also be aware that, from an operational point of view, most alternative assets are more labour intensive than traditional asset classes. When evaluating the attractiveness of an asset class, the resource requirement should be factored in. The staffing of the investment team as well as the back-office team should be appropriate for the complexity of the respective portfolio. Besides internal resources, this can also include external partners such as specialised investment consultants, external investment accounting, and investment reporting services. The organisation needs to have appropriate systems and adequate personnel to handle any pending capital calls. This also means that it should be easier for larger organisations (high AuM and larger teams) to manage more complex portfolios. As a rule of thumb, organisations must find a suitable asset allocation that matches their size and team resources available. Regarding resource constraints, it is also important to discuss how asset classes should be implemented. This can include an internal investment team, an external manager for an asset class, or a fund-of-funds approach. It is important to note that this decision can be made on an asset class by asset class basis. While an internal setup allows for more control, it also requires more resources. A higher share of alternative assets also means that more sophisticated investment accounting and reporting is necessary. To improve the governance standard, organisations should discuss whether they want an internal audit and/or external investment control. These functions should review both quantitative and qualitative aspects of the portfolio.

Conclusion

Regardless of the chosen model, institutional investors should keep a few key considerations in mind. ALM studies must take qualitative elements into consideration. Beyond the numbers, investors should have a sound investment rationale for their asset allocation and the respective asset classes. Decision makers should not underestimate the complexity of alternative asset classes. This means allocating enough time during ALM studies for due diligence efforts and the education of staff and board members. The handling and implementation of alternative assets may require a lot of resources. Therefore, decision makers should ensure that their organization is staffed appropriately and that they make use of external resources where this adds value. When adding illiquid investments, such as private market assets, liquidity management and planning must be a key element for future success.

About the Contributors

Andreas Rothacher, CFA, CAIA is the Head of Investment Research at Complementa AG. In his role, he advises institutional clients on strategic asset allocation and manager selection. In addition, he is a co-author of Complementa’s annual Swiss pension fund study (Risk Check-up). He is also the author of various articles and publications. Andreas is the Chapter Head of the CAIA Zurich Chapter and a member of the CFA Swiss Pensions Conference Committee. Before joining Complementa, he worked at a German family office and held various roles at UBS and Credit Suisse.

Thomas Breitenmoser, CFA, CAIA is the Head of Investment Consulting and Controlling at Complementa AG. He held management positions at Swisscanto/ZKB, JPMorgan Asset Management, Merill Lynch Investment Managers and Credit Suisse. He was also the president of the Foundation Board and the Investment Committee of JPMorgan Chase Suisse Pension Fund. Thomas studied economy and business administration at the University of Applied Sciences in Zurich and completed a dual degree Executive MBA in Asset and Wealth Management at the University of Lausanne and Carnegie Mellon University (Pittsburgh, USA).

Learn more about CAIA Association and how to become part of a professional network that is shaping the future of investing, by visiting https://caia.org/