Now that every super fund seems to have worked out how to play the APRA annual performance test game, it’s time to get on with the task of redesigning the test to make sure it really works in members’ interests.
Criticisms of the test structure and its likely impact on fund behaviour were raised before the Your Future Your Super regime officially began and have continued as it’s had the predicted effect.
In the first year of the test, 13 MySuper products failed – some because they genuinely performed poorly or were excessively expensive, but also because funds generally were still working out how to pass the test; in the second year five failed; the year after, just one. And this year, none.
Passing the test is relatively simple: stick closely enough to benchmarks, and you can’t go too far wrong. But if passing the test becomes a primary objective – that is, if the negative publicity and the consequences of failing are viewed as a big enough risk to require active management – then it seems reasonable to ask whether funds are acting in their own or members’ interests.
Ironically, the current test rests in part on the assumption that a fund’s past performance is likely to be a reliable indicator of its future performance, something that, in a different context, regulators insist investors should never assume.
Recognised shortcomings of the test in its current form prompted the government to open consultation on alternative approaches. The first thing to say about this is that while there’s little doubt the test can be improved, there’s probably no perfect solution.
In a consultation paper published in March, Treasury noted that “the test has improved member returns by holding trustees to account for investment performance and encouraging continual improvement or exit of poor performing products”.
“However, there is evidence that the test may be influencing investment decisions to the detriment of member outcomes, including discouraging investment in asset classes that may otherwise be in the best financial interests of members,” it said.
In a submission to Treasury, the Conexus Institute* noted that the test should be designed for the benefit of members, but right now it creates behavioural biases within funds that might not serve members best.
The Institute said that “the industry is now actively managing the current test making it quite possible that no MySuper option will fail in the future”. That turned out to be right on the money. And it added: “As such, the current test may have become ineffectual yet nevertheless has some adverse behavioural impacts.”
In its submission to Treasury, consulting firm Willis Towers Watson said that relying on a single performance metric “will necessarily provide a clear incentive for funds (whether consciously or implicitly) to manage their investment program and product design to ensure a favourable outcome in regard to that metric (or, at least, to minimise the likelihood of an adverse outcome)”.
WTW said this can “work against the overall long-term interests of members where the outcomes captured by the single metric do not align perfectly with overall member outcomes”.
In other words, if a fund’s performance is gauged against a benchmark, the fund will potentially prioritise performance relative to that benchmark over setting an asset allocation that potentially deviates from the benchmark but which suits members’ interests better.
“The potential for such distortion is high given the outcome of failure of the test is so grave,” WTW said.
A common theme canvassed in submissions is that the test has clearly become of more importance to funds for self-preservation or marketing purposes than it has for meeting members’ best financial interests.
Treasury floated four performance test design options for comment and feedback: maintain the status quo; introduce an alternative single-metric test; introduce a multi-metric test; and “any alternative performance test frameworks that may more effectively address the principles set out in this paper” – essentially, a “start-again” option.
There is broad consensus that the current single-metric test – especially a backward-looking one based on a flawed assumption – should be replaced. The favoured alternative seems to be a muti-metric test, but then agreement needs to be reached on exactly what those metrics should be.
Nevertheless, the latest round of performance test results lend weight to the argument that the existing test has passed its use-by date and it’s time for a review. At the end of the day, the current annual performance test is more an assessment of whether funds can follow the rules of a game than it is a way to ensure the best outcomes for members.
Any test that distorts funds’ behaviour, however much, and distracts from a focus on members’ best financial interests is counterproductive.
* The Conexus Institute is an independent, not-for-profit think-tank philanthropically funded by Conexus Financial, the publisher of Investment Magazine
