Introducing Australia’s Performance Test may not be the panacea for UK DC pension schemes

Last week’s King’s Speech noted one component of Labour’s upcoming Pensions Bill will be the introduction of ‘a standardised test that trust based defined contribution schemes will need to meet to demonstrate they deliver value’. Speculation in the media has suggested that the UK may adopt the ‘Performance Test’ metric introduced in Australia in 2020. Coincidentally, I’ve recently returned to the UK having spent the last 5 years in Sydney where I spent considerable time advising Australian DC schemes and policymakers on the Performance Test. A back of the envelope calculation tells me I spent more than 1,000 hours in this space which probably explains why I’ve returned without a decent suntan. 

The Australian Performance Test was a well-intentioned but poorly implemented attempt at improving outcomes for DC members. In this blog I’ll summarise how the test works, some of the key challenges it has faced and provide some suggestions for how a UK equivalent could be adopted. I firmly believe the UK can do better, but it’s important we don’t sleepwalk into copying the Australian version just because many other aspects of the Australian pensions system is market leading. It’s also worth noting that there has been a change in government in Australia since the Performance Test was first introduced. The current government is consulting with the industry on improvements to the methodology, with many of the considerations raised in this blog currently being examined.

The Performance Test explained

There are two key components:

• Firstly 10-year p.a. performance of an investment option is compared to a composite benchmark. Each investment option has its own bespoke benchmark based on the asset classes used in the portfolio (for example, if 10% of the strategic asset allocation is to Emerging Market Equity then 10% of the composite benchmark will be to the MSCI Emerging Market Index, etc.). 
• The second component of the Performance Test involves comparing the administration fee of the investment option (over the past 12 months rather than 10 years) to the median administration fee of competitor options. 
• The results of both components are added together with a score of -0.50% and below being deemed a failure.  

The Trustees of a failing option are legally compelled to write to members and inform them of the failure and advise them to consider alternatives. Additionally, an option that fails in consecutive years can no longer accept new members. 

Impacts

From the outside, it might be tempting to believe the Performance Test did its job. The inaugural Performance Test in 2021 covered default options only and 13 out of 80 options failed. In the majority of cases, the DC schemes operating a failing default option have since merged with another scheme. Administration fees have also reduced considerably. However, one doesn’t need to dig very deep to see the flaws. 

Arguably the biggest concern is the limited link between performance relative to a composite benchmark and member outcomes. The successful provision of member outcomes should be based on a range of investment and non-investment factors. Even within investment aspects alone, the decision regarding the proportion of the default strategy in growth assets versus that in defensive assets (and how this changes over time for a lifecycle) has a much bigger impact on member outcomes than benchmark relative performance. There are examples where an option meeting its investment objective would likely result in the Performance Test being failed. For example, a key objective for the latter stage of a lifecycle strategy is to manage sequencing risk (i.e. negative returns are more detrimental when experienced around the time drawdown from a portfolio is commencing compared to when it’s near completion). Portfolio construction techniques recognising sequencing risk might reasonably involve a bias toward ‘quality’ stocks or the use of derivatives to change the return profile of an equity allocation, particularly to reduce downside risk. A portfolio construction approach that runs below benchmark risk would also likely have a below benchmark return hence an expectation of failing the Performance Test.

The implications of failing the Performance Test are severe, particularly the letter that must be sent to members holding a failing option (see excerpt in italics below). 

“Your superannuation product [XYZ] has failed the annual performance test at least 2 years in a row. We are now banned from accepting new members into the product until it passes a future test. You should consider moving your money to a different super product. …….you could save thousands of dollars more for when you retire by switching to a better product”

This letter has been watered down since the inaugural test was conducted but readers will note it still encourages members to switch to an alternative option and makes this as easy as possible with a QR code that displays better performing competitor options.

The severity of the implications of failing the Performance Test understandably changes investment decision-making for those managing the investment portfolio. Herding has become more prevalent (already an issue in Australia where performance versus peers is a key objective). Rather than herding toward peers, the Performance test incentives portfolios to look more like benchmarks where possible. Some of these reversions back towards the benchmark have been detrimental to Australian members, most notably the unwinding of an underweight duration position in place across most DC schemes just prior to one of the quickest and largest interest rate rises in history. There are also other implications which will likely conflict with wider Labour policies if the Performance Test is introduced in the UK, the most obvious being undermining the push for DC Schemes to invest in UK private market assets (given private market benchmarks are un-investible and therefore provide unmanageable tracking error). Net Zero policies are likely to be similarly undermined.

Some thoughts on improvements

It’s important to emphasise that the purpose of this piece is by no means to argue against policy that aims to uplift outcomes for members, rather it aims to emphasise that the implications of the policy need to be considered carefully to avoid unintended consequences. Two alternatives are summarised below.

Firstly, a suite of metrics could be used that more holistically assess whether high quality outcomes have been provided to members. In the investment space, this might involve including a risk-adjusted metric. A peer-relative metric may also have merit although clearly increasing the number of metrics comes with increasing complexity.

Secondly, an improvement might be amending the implications of failing a Performance Test. Rather than triggering a letter to members, a failure could instead be a trigger for enhanced regulatory scrutiny where the Trustees need to submit evidence to show why they believe member outcomes are still being provided. The more flexible approach would need to be considered carefully as the high number of UK trust-based DC schemes would likely lead to a greater regulator workload.

To wrap up

In many ways, the Australian DC system is world leading and the new Labour government will be able to cast their eye ‘down under’ for ways to improve the UK system. However, the Performance Test introduced in Australia is anything but world leading. Whilst we currently don’t know Labour’s intentions, the application of such a test to the UK market should be approached with caution. 

The new government has signposted a collaborative approach to pensions reform and we look forward to engaging with the industry on this and other areas over the coming months.