Low-balance SMSFs under fire, but SMSF sector fights back

Mr George told the Productivity Commission at a public hearing that the data used to conduct this analysis into costs and returns between SMSFs and APRA-regulated funds in their report was not substantial enough to draw conclusions.

“The nature of costs in an SMSF are different to those in APRA regulated funds,” Mr George noted.

“Often we talk about advice an investment costs which are quite different to what an APRA regulated fund provides their members. So advice is obviously a direct cost that is very transparent in an SMSF and reported in the annual return, while for many APRA funds the way that cost is spread across the fund is quite different and isn't necessarily reflected as an administration cost or an investment cost as it would be in an SMSF.”

The SMSF Association, he said, is also worried by the quality of the data driving some of the analysis by the Productivity Commission in its report.

“What we are concerned about is the quality of the data and comparability of data. It is actually really hard to make comparisons between APRA regulated funds and SMSFs, and because of that, some of the return figures for lower balance funds can be distorted, and I think one of the key things we focus on is the difference between the APRA methodology for return on assets and the ATO methodology,” he said.

The SMSF Association, he said, is concerned by suggestions in the report that $1 million may be an appropriate minimum figure for setting up an SMSF.

“While we know the commission has not made a draft recommendation about this, we are concerned about people taking this $1 million figure to be the minimum amount needed to establish an SMSF. We don't think that is an appropriate figure for that, and the reality is that many people may start off in small SMSFs and achieve scale over time,” he told the Commission.

“That $1 million figure is probably a good aspirational figure, especially when you think about a two-member fund which 70 per cent of SMSFs are, that would be about $500,000 per person, which is an amount that we would regard as a good amount to aim to save for in their SMSF.”

Mr George said that more work needs to be done around what an appropriate starting balance is for SMSFs but there needs to be better data sources around the costs and returns for SMSFs.

“The reality is that the ATO collects SMSF data through the SMSF annual return and its primary job is to collect revenue and not collect statistics so the ATO does the best it can with what's given to it, and we think more work can be done around data,” he stated.

He also noted to the Productivity Commission to what extent the presence of small SMSFs are necessarily a problem.

“We don't think it is a problem because they generally do achieve scale over time, or they are exiting the sector. So we think the small funds with less than $100,000 are either going one way or the other. They're either going out and they're growing in scale,” he explained.

“There is some interesting research that was done by Class [that] shows in the 2015 financial year that 50 per cent of the funds with less than $50,000 and were either newly established or were entering this bracket due to drawdowns and rollovers and so they were probably very much in the process of winding their fund up.”

Also speaking at the public hearing, the Productivity Commission said that while they acknowledged that the data could have been more comprehensive, it was clear that if balances didn’t reach $1 million fairly quickly, an SMSF would be able to achieve a 5.9 per cent investment return over the 10 to 12 year period, which would make it competitive with the retail offerings.

“I think this is an area, where the regulator under a FOFA world is going to be doing more work so it would be helpful to know what advisers consider to be a cost effective balance and what time frame they need to get to that balance by in order for an SMSF to be something they feel comfortable advising a client to go into,” the Commissioner stated at the hearing.