In its recent ‘pulse check’[1], APRA and ASIC noted that the implementation of the retirement income covenant across superannuation funds has been varied and, in general, has lacked urgency and progress. While this may be true in a broad sense, there is no doubt that over the past year or two there has been an acceleration of development of funds’ retirement income strategies and an increase in product development, especially insured products.
Against that backdrop, it is critical for trustees to have a robust framework for member retirement outcomes that properly considers the role of alternative retirement products in the context of their broader retirement income strategies and members’ best financial interests. In this first article of our Retirement Focus series, we discuss a framework that trustees can use to assess member retirement outcomes and design retirement income strategies.
Under the retirement income covenant, trustees are required to establish a retirement income strategy that addresses how they will assist those members to achieve and balance the three competing objectives of:
For members and advisors there may, however, be a broader set of objectives to consider such as managing income/wealth variability, maximising Age Pension entitlements and leaving a bequest.
A key challenge for trustees, and indeed members and advisors, is how to assess the interaction of different product choices to achieve these overarching retirement income strategy objectives.
At the heart of the framework are four core assessment metrics, each aligned to one of the three retirement income covenant objectives. As these objectives comprise both expected outcomes and variability of outcomes, the assessment will almost certainly need to be stochastic in nature and consider a range of financial outcomes. It will also need to consider the longevity of the member and other behavioural aspects (e.g. risk appetite).
Income for life: measures the ability or likelihood of a strategy to deliver a target minimum or base level of income for life. This metric allows for the contribution of the Age Pension towards this minimum lifetime income need.
Income for living: when considering retirement spending patterns, many members wish to aim for a higher level of income in the early years of retirement when they expect to be living more active (and expensive) lifestyles. This metric measures the ability or likelihood of a strategy to deliver an additional level of target income over and above the ‘income for life’ target during those early years of retirement.
Income stability: measures the frequency and extent of year-on-year changes in total income.
Access to capital: measures the proportion of retirement savings that are liquid and accessible (i.e. could be moved to an alternative product or withdrawn in cash if the member so chooses) over a member’s retirement.
In order for the metrics to effectively assess a retirement income strategy, they need to:
The four core assessment metrics do this using a reference-dependent utility approach. The ‘reference’ or target can be varied across and within funds depending on the characteristics of the membership, and utility is measured using a loss aversion function.
Additional (secondary) assessment metrics can also be introduced into the analysis, such as Age Pension utilisation or ability to leave a bequest, if deemed relevant and an important consideration for members.
Recognising that the importance and balance of the retirement income covenant objectives will differ from member to member, a set of preference weightings can be used to define the contribution of each metric to an overall Retirement Score.
The preference weightings can be defined or captured in a number of ways. For example, the preference weightings could be set by default by the trustee based on the characteristics and expected behaviour of a cohort of members. Alternatively, the preferences could be captured by asking the members to rate the importance for each metric through a targeted questionnaire.
The Retirement Score, calculated as a weighted average of the metric results and the preference weightings, can be illustrated in different ways depending on the audience.
For example, Figure 2 shows a pie chart that could be used to communicate the results to members where the size of the wedge is aligned to the member’s preference weighting. This visualisation would allow the member to compare scores for different combinations of product choices and features and allow them to see the contribution of the preference weightings.
In this example, the member would observe that the nominated retirement income strategy yields a very strong result in terms of providing a stable income that meets their needs and wants but in doing so sacrifices some flexibility in terms of access to capital – which is consistent with their preference weightings.
Trustees using this framework to assess and design retirement income strategies are likely to require additional detail that includes the distribution of outcomes. An example is shown in Figure 3 where five retirement strategies are compared using the Retirement Score framework. The distribution of outcomes between the 5th and 95th percentiles is shown for each strategy along with the mean (blue dot) and median (white line) outcomes. In this example, retirement strategy 4 is shown to be the strongest.
The framework for assessing and designing retirement income strategies is part of a broader framework that includes:
Member advice is a key plank that underpins the broader retirement strategy and the recent passing of legislation in response to the Quality of Advice Review[2] provides clarity for funds that has enabled them to refine their advice offering.
This will be discussed later in our Retirement Focus series.
If you would like to discuss how this framework could be adopted to assess member retirement outcomes for your fund membership, or participating in the WTW Retirement Preferences Survey, please contact our team.
