We support the steps APRA is taking to improve data collection, reporting and transparency around superannuation products across the industry. However, we note that data collection is onerous for the industry and therefore do not propose any further metrics to be obtained from providers. We believe the greatest value would be achieved through further refinement of the data being collected, with standardisation and improved consistency, where possible, supporting usability. A summary of our responses to key areas requested by APRA is below.
We note that there is still very good reason for segmenting funds based on their fund type due to differences in their ownership structures and product offerings which flow down into impacts that will be seen in service provider agreements, fee structures and insurance arrangements.
We support the proposal to apply a representative member balance and believe a $100,000 account balance is reasonable as this is reflective of the average member balance observed.
In our experience we have found that Sharpe ratios are not broadly understood, especially amongst consumers, and are of limited usefulness if not considered alongside other risk metrics, due to their known shortcomings, particularly the inability to account for volatility of unlisted assets. Furthermore, disclosure of multiple performance measures which seek to describe similar underlying constructs, is likely to cause confusion among users of the data and particularly members if they view this information.
We note the challenges of determining growth asset proportions across the industry due to inconsistencies in classification approaches, particularly for unlisted property and infrastructure assets. We believe there is room for improvement here and that a consistent approach is needed for meaningful comparisons to be undertaken across providers. As a result, the computation and interpretation of growth asset proportions across the industry remains a challenging area. Therefore, the introduction of growth asset categories is likely to create further complexity and confusion given there are already a number of bands being communicated, including our own method.
We are unsure of the purpose of computing the difference in premiums paid by members versus premiums paid to insurers. If the goal is to understand the cost of insurance administration, there are more suitable methods to utilise. We note that there are likely to be several reasons for there to be a non-zero difference between premiums collected and remitted. These include timing differences, differences in accounting practices and factors inherent to benefit design.
One possible objection to the publication of default insurance cover and cost is that products and costs are not always directly comparable from fund to fund. Nevertheless, we do see some value in publishing the default product details for each fund to the extent that these are by definition the terms that will apply to members who make no active decision regarding their insurance cover.