Asset allocation is a bit like inviting people to a dinner party. You have to try to find the right mix and ideally the people should have different backgrounds – that reduces the risk of boring discussions. Also, you can not fully rely on past experience – it’s an indication, but no guarantee that the same crowd will make a good party again.

Similar mechanisms work when you construct a portfolio. Diversification, i.e. mixing different asset classes, is in general a good thing. It should make you less vulnerable to single events and should therefore stabilize your portfolio value. But there is no guarantee that asset mixes that worked in the past, will also work in the future. Correlation is an unreliable friend, as we learned, especially in difficult times.

Asset allocation is therefore more an art than an exact science. It’s in fact already difficult to find the right mix between equities and bonds. When it comes to hedge funds, many asset allocators find it even more difficult. Stretching the above metaphor a bit further, hedge funds are a bit like your new neighbours: you know that you should invite them to your party, but you find it hard to judge how they will fit in.

In practice, many asset allocators therefore treat hedge funds as a separate asset class with a separate risk budget and typically allocate a relatively small part of their allocation to it. In our experience this approach is often chosen, not because allocators are particularly convinced about it, but because “everyone” does it this way (including most regulators) or because they lack a better alternative concept.

However, just to have no better alternative does not necessarily mean that treating hedge funds as a separate asset class is appropriate. In our view there are some straightforward considerations that lead to a potentially better approach. In essence hedge funds should not be viewed as a separate asset class, but as an actively managed part of “traditional” asset classes, such as equities, fixed income, credit and commodities.

In this article we want to lay out a simple and practical asset allocation concept incorporating traditional assets and hedge funds. We will provide an allocation grid and examples of how to use it for strategic as well as tactical asset allocation decisions.

Dimensions of asset allocation
There are two generic dimensions to asset allocation. The most obvious one relates to the allocation split between different asset classes such as equity, fixed income, credit or commodities. The decision can be extended to include regions, sectors and currencies as well as other classifications. Such a split allows steering the portfolio’s overall characteristic as each asset class has its unique return drivers.

The second dimension is often introduced in the context of risk budgeting. It defines how much of the risk should come from market exposure and how much from active decisions. Or to use financial terminology, it sets the allocation between beta risk and alpha risk exposure.

Our goal is to align hedge funds along these two dimensions and thus integrate them into a practical asset allocation concept. At first glance this might seem easy. Hedge funds invest across all types of asset classes and have a high amount of alpha risk. They can easily be fitted into an asset allocation grid as shown in Fig.2.

It is not so easy at second glance. The challenge is that the hedge fund universe is very heterogeneous. We need to be more granular and look at single hedge fund types or strategies. But even single hedge fund strategies can invest across different asset classes or be exposed to alpha as well as beta risk, making a classification less straightforward.

Developing an asset allocation grid
To help us with the classification, we apply a statistical trick. For each hedge fund strategy or hedge fund product we ask how the best replicating portfolio would look like if we could only invest in traditional asset classes and in a hypothetical alpha risk portfolio.

Fig.3 shows the composition of the replicating portfolios for three different hedge fund strategies. The replicating portfolio for the equity hedge strategy is – not surprisingly – dominated by equity exposure (exposures to developed equity, emerging markets and company size). It also includes small weights to commodity and currency, and has a 20% allocation to “pure” alpha. The replicating portfolio is derived from a statistical factor model based on empirical data from 2000-2010.

The composition of the replicating portfolio for credit long/short and equity market neutral strategies is very different, with the latter being almost exclusively exposed to alpha.

Armed with this statistically derived information we can be more granular when aligning hedge funds along the asset class and alpha-beta dimension. The result is a much more detailed asset allocation grid as shown in Fig.4. Hereafter we use this grid to show how hedge funds can be incorporated in typical asset allocation decisions.

The asset allocation grid suggests a new two-step approach when designing a portfolio and deciding on a strategic asset allocation. The first step of this new approach consists of allocating between (traditional) asset classes using established methods such as risk/return projections and scenario analysis as well as client expectation and constraints. Because hedge funds are no longer considered a separate asset class, they are not allocated an individual risk budget initially. Only allocating to traditional asset classes at this stage allows defining what generic risk drivers will affect the portfolio in the long term.

The second step addresses the alpha-beta dimension. Here the question becomes: “How do we want to be exposed to each asset class?” Or in other words, how do we want to be exposed to the risks driving the performance of each asset class? To what degree do we intend to take passive beta risk versus active alpha risk?

To give an example, we assume that the first step resulted in a 20% allocation to equities. As part of the second step, we decide to split this 20% allocation and invest 5% in a purely passive way, 10% through traditional long only or 130/30 mandates, and the remaining 5% allocation to high-alpha risk exposure. Hence we can invest 5% into hedge fund strategies such as equity hedged, trading, event driven or equity market neutral. Just as a reminder, these are all hedge fund strategies that are exposed to equity risk drivers in one way or another and are thus aligned with the decision made in step one.

For our purpose we have focused on liquid asset classes. For the purpose of asset allocation, we consider hedge funds to be liquid as long as they provide at least quarterly liquidity. Real estate, infrastructure and private equity can be integrated into this framework by introducing a third dimension – liquidity.

High equity valuations, fading growth expectations at the end of a business cycle, events triggering an increase in risk aversion, or supply shocks are all reasons for wanting to tactically reduce equity exposure and risk in general. Within the asset allocation grid, this can be implemented by either moving vertically out of equity into other asset classes or horizontally away from beta to alpha exposure. A vertical move out of equity will predominantly flow into fixed income especially if valuations and growth concerns are at its origin (see Fig.5).

Yet if this move is triggered by a spike in risk aversion or a supply shock, allocations to commodity and anti-carry currencies offer interesting tactical possibilities. Anti-carry currencies are low-yielding, “safe haven” currencies which tend to appreciate during times of heightened global uncertainties. A horizontal move toward alpha risk makes most sense if tactical changes are expected to have a mid- or longer-term horizon simply due to the liquidity aspect of hedge fund products. A horizontal move must not restrain itself to equity-based hedge fund strategies. Other hedge fund strategies can be considered. Systematic trading strategies (CTAs) for example can be very useful to diversify equity risk.

A booming economy and inflation pressure can underpin rising interest-rate expectations and the desire to move out of fixed income and long credit positions. Translating this situation into the asset allocation grid, we can again consider the merit of vertical and horizontal adjustments. Commodity prices should profit from a booming economy, and increasing interest rate differences between countries can create opportunities in the FX space. If investors’ liquidity needs permit, real assets such as infrastructure or real estate can be considered. While rising interest rates do not support equity prices in general, moving into selected equity sectors such as consumer staples, telecoms or technology offers another vertical possibility.

Horizontal moves open up new doors to address a scenario of rising interest rates (see Fig.6). A switch to credit and fixed income-related hedge fund strategies can be complemented with more market neutral equity strategies or active commodity products. The asset allocation grid offers a new approach to think about ways to implement tactical views.

Conclusion
As initially stated, our goal was to provide a straightforward, simple and practicable asset allocation concept embracing hedge funds. We therefore introduced a two-dimensional asset allocation grid and gave examples of potentially beneficial horizontal and vertical shifts. This approach gives investors more options to express not only their strategic, but also their tactical views (like in the examples of falling equity markets or rising interest rates).

The key difference to the models we typically see in practice is that hedge funds are not treated as a separate asset class with a separate risk budget. Instead, different hedge fund strategies are seen as actively managed risk exposure within the traditional asset classes they best fit into. Clearly this is not an exact science, but looking at the explaining factors of a replicating portfolio gives a fairly useful idea how to classify the different hedge fund strategies.

Once this exercise has been completed and the asset allocation grid finalized, the strategic and tactical changes become increasingly clear. Under most scenarios, both vertical and horizontal moves will be viable options. Ultimately this should lead to a better asset mix. It’s like getting to know the new neighbours before you send out the dinner invitations.

Vincent Couson is Head of Strategic Alternative Solutions in Global Investment Solutions with UBS Global Asset Management. Wolfgang Batt is Hedge Funds Product Specialist in Alternative and Quantitative Investments with UBS Global Asset Management.