Absolute Return Strategies: Understanding the underlying drivers of returns is key.

With interest rates at all-time lows, there is an increased focus by investors on how to enhance returns without assuming ‘excessive levels’ of interest rate or credit risk. Against this backdrop there is growing interest in what is broadly referred to as ‘absolute return’ strategies. However, care must still be taken by investors as not all absolute return strategies are the same.

What is an absolute return strategy?

The term ‘absolute return’ can encompass a broad array of strategies and are therefore difficult to clearly define. At one level absolute return strategies can be viewed as strategies which add value irrespective of the direction of markets. The issue with such a definition is that it risks becoming a bit of a logical tautology depending on realised returns. To deal with this, it is more useful to take a narrower perspective by defining them as strategies which are concerned with the absolute return of assets within an asset class as opposed to the exposure relative to an asset class benchmark. Another way of phrasing this is that they constitute benchmark unaware strategies within an asset class (1).

Differentiating Between Styles

Though often the two may be consistent, there is a clear distinction between saying that a strategy is benchmark unaware and that it is not impacted by the directionality of markets. To understand why, one needs to start by considering that most absolute return strategies exploit some form of relative value-based trading to generate returns without exhibiting an explicit net directional bias. With no explicit net directional bias the key drivers of value add become those factors which drive convergence. This means that any absolute return strategy can be conceptualised as sitting along a spectrum based upon the convergence factors which are driving value add. Though there is no standardised terminology, for convenience and clarity within this paper convergence factors will be considered as being either :

• Macro driven
• Non macro driven

While both types of convergence factors rely upon capital flows as the drivers which generate the additional return from convergence, the key difference is in the nature of the capital flows. More specifically, macro driven convergence trades rely much more upon capital flows into and out of markets in order to bring about convergence. In turn, these capital flows are more likely to be driven by a view of the world panning out for convergence, and thereby value add, to be generated by the strategy; i.e. value add exhibits some form of macro directionality.

By contrast non macro driven convergence trades rely more on flows between securities within a market rather than flows into the market itself. Accordingly, the ability to add value through such trades is less impacted by macro views being vindicated. The distinction with respect to the form of capital flows highlights that with macro driven convergence trades there is likely to be a greater macro directional bias to return profiles. Knowing and understanding both the nature and extent of any directional biases becomes the key to ensuring that an absolute return strategy is consistent with the desired investment objectives.

How to distinguish between styles in practice? : A Case Study

While it is all very well saying that there are two broad classes of convergence trades, how can one distinguish the extent to which they are being used by different managers? Though there are many approaches to analysing absolute return strategies, to address this question the following sets out an illustrative approach using two actual strategies. To add more context around the analysis it is useful to define which directional bias the investor is particularly sensitive to. With interest rates across the developed world being at, or near, all-time lows, it is not unreasonable to assume that an investor would be interested in an absolute return strategy which is less correlated to the directionality of interest rates. Reflecting this investor bias, two actual absolute return strategies have been chosen which are explicitly managed so that on a net exposure basis there is no interest rate exposure; i.e. net duration exposure is close to zero. Given the negligible net duration exposure in each strategy, on the face of it, the two strategies should be comparable with neither exhibiting a directional bias to movements in interest rates.

When considering any strategy an investor needs to look in detail at the types of trades utilised to fully understand its underlying characteristics. Due to such prior analysis, it’s known that one of the two strategies chosen utilises macro driven trades while the other places a much greater reliance on non-macro driven trades. But can an investor tell ‘which is which’ simply by looking at the return history for each strategy? Fortunately, it is often possible to understand the drivers of convergence behind an absolute return strategy by considering its return history in the context of fixed income risk factors. A starting point for an investor could involve simply comparing the correlations of the two strategies with some standard measures of fixed income market return risk. For the purposes of this comparison the macro factors chosen are credit spread and interest rate risk.

A comparison of the correlations for the two strategies highlights that, though the correlations to credit spread risk are similar, there is a marked difference in the correlation to interest rate risk. In contrast to Strategy 1, Strategy 2 appears to be quite correlated with interest rate risk. Such a high correlation provides prima facie evidence for Strategy 2 making materially greater use of macro driven strategies despite it not exhibiting an overall duration exposure.

Beyond simply utilising correlations the analysis can be taken to the next level by considering how the strategies perform under different interest rate environments or scenarios. By looking at different environments an investor is in a better position to understand whether the strategy is consistent with their objectives. With interest rate risk being the area of most behavioural divergence between the two strategies it is worth looking at this macro factor in more detail. The following sets out one approach to considering different interest rate scenarios. As a ‘back of the envelope’ approach the movements in Interest Rate Risk can be broken down into three key risk factors:

Level

• Interest rate on 1 year bonds.

Slope

• Difference in yield between 10 year and 1 year rates.
• Negative (positive) number denoting a yield curve steepening (flattening).

Cross sectional dispersion

• Higher number indicates that more markets are moving in the same direction.

The results from the decomposition can be seen in Figure 2 which plots the distribution of cumulative returns for each Environment Score in a rising and falling interest rate environment.

While the distributions of each strategy are interesting, how they compare to a long duration bond exposure (calibrated to a long duration exposure of around 3 years) is of more relevance given the predefined investment objectives. Of the two strategies, as highlighted by Figure 2, Strategy 2 exhibits characteristics which are more closely aligned to those of a long duration exposure. This is particularly evident when considering an environment where yield curves are steepening (Environment Score is negative). From such an analysis of returns an investor can ascertain that Strategy 2 has exhibited return characteristics more in line with those of a long global duration strategy. Hence by using historical returns the investor has been able to identify that :

• Strategy 2 is likely to be making greater use of macro driven trades and
• the macro drivers of convergence utilised by Strategy 2 are mimicking the behaviour of a long duration exposure.

It follows that, based on the historical return profile, Strategy 1 is more likely to be consistent with the investment objectives of an investor wanting an absolute return strategy which is uncorrelated to interest rate movements.

Just because strategies are benchmark unaware and exhibit no outright duration exposure does not mean that they will exhibit the same characteristics to macro factors. Based on the types of relative value trades utilised within the strategies they may exhibit materially different return profiles. Understanding the difference in return profiles, and how they behave under varying environments, is critical if an investor is to ensure that the absolute return strategy selected is indeed consistent with their investment objectives.