Competing objectives? The YFYS performance test and portfolio construction

When constructing MySuper and choice portfolios, all funds will need to assess the priority they attach to not failing the Your Future Your Super performance test.

Our analysis shows that, based on existing strategic asset allocations, even funds that are currently well positioned against their YFYS benchmark are likely to be confronted with the possibility of failure at some stage.

This risk can be reduced through more closely hugging the underlying indices used for the performance test, but this will have broader implications for portfolio quality. As such, funds will need a very strong governance framework and clear beliefs on the weight they give the YFYS test in order to ‘stay the course’ when performance falls below expectations, and the risk of failure escalates.

Integration into the portfolio construction process

The YFYS performance test effectively creates another lens through which portfolios can be viewed. We believe that the best way of incorporating it into the portfolio construction process is by adding it to the broader framework used to assess overall portfolio quality.

To do so, funds will need to first address:

• where the performance test ranks relative to other metrics of portfolio quality
• what probability of failing the performance test is acceptable, and what is the most appropriate timeframe over which to assess this.

Funds should also consider second order impacts of the performance test on other portfolio quality aspects. For instance, if there is a meaningful probability of failing, what implications does the potential impact on member cashflows have on the level of liquidity required?

The use of reverse stress testing (or ‘pre-mortems’) is also likely to be helpful – i.e., thinking through the circumstances in which the fund actually fails the test and identifying the actions that could have been taken in order to prevent this. In short, funds should have a ‘game plan’ for how to deal with underperformance as it unfolds in real time.

Avoiding test failure will be a high priority for funds, but this will not override the obligation to act in members’ best financial interests, nor will it remove the need to manage a variety of other objectives and risks in constructing portfolios.

To help with these different (and at times competing) objectives, we suggest trustees use a framework that assesses their portfolios across a range of factors, such as expected return, risk (e.g., frequency of negative returns and tail risk), liquidity, fees, sustainability, along with the risk of failing the YFYS test. It is impossible to optimise for all of these and so trade-offs must be made. Establishing the importance of each factor, as well as the metrics available to assess each one, is crucial to managing portfolios in a consistent and coherent way.

Going beyond the silos

Using a total portfolio approach can enable funds to more effectively weigh up the competing objectives and constraints, rather than a siloed approach in which asset class buckets are considered in isolation.

In our view, some asset classes do offer a higher reward-risk ratio for taking on active risk – assuming that they are implemented using high quality managers and at reasonable fee levels. Examples include:

• Alternative Credit – under the draft regulations, most sub-sectors of alternative credit will be benchmarked against a lower yielding fixed interest benchmark, which does result in tracking error, but also a significantly higher expected return
• Global equities – where active management can generate alpha in a highly risk-aware manner and in a way that has a low correlation with other sources of alpha.

However, every fund will need to assess where it has the best chance of outperformance, based on the make-up of their portfolio and their own beliefs.

It will also be important to assess portfolio-level impacts and interactions between active positions, noting that ‘active’ in a YFYS performance test context means both traditional active management, and investments that simply differ from the benchmark against which they are measured. Using a total portfolio lens will help to avoid two potential pitfalls:

• Active positions that are positively correlated – and likely to underperform at the same time. For example, inflation-linked bonds and alternative credit are both currently benchmarked against a fixed interest benchmark, meaning they are both likely to underperform in a ‘risk-off’ environment, where credit spreads widen, and nominal yields fall more than real yields
• Considering investments in isolation – investments with lower expected returns than their YFYS benchmark or with high tracking error may appear unattractive when viewed in isolation, but they could be paired with off-setting investments. For example, long-duration nominal bonds and floating-rate credit may each pose significant risks relative to the standard fixed interest benchmark used in the YFYS test, but combined will have much lower tracking error, with potentially attractive characteristics when viewed at a total portfolio level.

A framework for integrating the test

To develop a framework that will incorporate the YFYS performance test into the portfolio construction process, we suggest that super funds determine:

1. The likely implications of failing the test
2. The weight placed on passing the test relative to other factors when assessing the overall quality of their portfolio and member outcomes
3. The probability of failing the performance test deemed acceptable and the timeframe over which this will be assessed
4. The overall level of active risk or tracking error they are willing to take relative to the YFYS benchmark and the level of outperformance this is expected to generate at the total option level
5. The areas in which they have most conviction in taking active risk (in terms of the expected reward relative to the specified YFYS asset class benchmark) and whether this results in any differences to the desired or ‘optimal’ portfolio before YFYS considerations
6. How frequently total fund outcomes relative to the YFYS benchmark will be assessed, and in what circumstances these should lead to changes in the level of active risk being taken. This should include consideration of the actions that can be taken as relative performance deteriorates to reduce the likelihood of failing the test.

The next article in this series will explore the potential impact of stapling and the increasing importance for funds to have strong brand recognition and a compelling direct to member proposition.

Author

---

Jonathan Grigg, CFA

Director, Investments

email Email phone +61 3 9655 5155

---