SMSF return on assets
The following data is taken from Data tables, table 22 and table 23.
Care must be taken when using SMSF performance figures, particularly when making comparisons. While the methodology used to estimate SMSF performance resembles APRA’s methodology, the data collected is not the same.
In the five years to 2016–17 the performance of the SMSF sector has been positive. The estimated return on assets (ROA) for SMSFs in 2016–17 was 10.2%.
Graph 11 shows comparisons between the ROA for SMSFs and the rate of return (ROR) reported for APRA funds of more than four members.Footnote26
Graph 11: Average returns (%) for SMSFs and APRA funds 2012–13 to 2016-17
Graph 12 shows the proportion of SMSFs by their estimated ROA over a five-year period.
Over the five years to 2016–17, most SMSFs recorded a positive ROA (78% in 2016–17).
The proportion of SMSFs recording a zero or negative ROA from fell from 48% in 2015–16 to 22% in 2016–17. In turn, the proportion of funds with an ROA of greater than 5% increased from 18% in 2015–16 to 52% in 2016–17.
The estimated SMSF ROA continues to show a direct relationship with SMSF size. Generally, the larger the SMSF asset holding the higher the ROA.
Graph 12: SMSF return on assets 2013–17
Graph 13 shows the estimated average ROA by SMSF size for the five years to 2016–17. On average, SMSFs with more than $200,000 in assets had a positive ROA in 2017.
Graph 13: SMSF return on assets by fund size, 2012–13 to 2016–17
- Footnote 26
- APRA, Annual Superannuation Bulletins, 2013 to 2017
SMSF expenses
The following data is taken from Data tables, table 24 and table 25.
Care must be taken when comparing expense ratios of SMSFs with those of APRA funds. While the methodology used to estimate SMSF expense ratios is close to APRA’s method, the data collected is not the same (see Data issues).
We report on 'total expenses' incurred by SMSFs, broken down into 'investment expenses' and 'administration and operating expenses' (see Glossary).
In 2016–17, the estimated average total expense ratio of SMSFs was 1.17%. In dollar terms, the average total expense was $13,900.Footnote27
SMSFs solely in the accumulation phase had average total expenses of $13,300, while those in pension phase averaged $15,300.Footnote28
The ratio for 'investments expenses' was 0.66% and for 'administration and operating expenses' 0.51%. Over the five years to 2016–17, the expense ratios by type of expenses incurred have remained relatively consistent.
Lower balance funds tend to have higher expense ratios. For example, the average total expense ratio for funds with assets of $1 to $50,000 was 14.29% in 2017.
Graph 14 shows estimated expense ratios for 2016–17 by fund size. The ratios generally decline as funds get bigger, except for funds with assets of between $100,000 and $200,000.
SMSFs with $50,000 or less in assets had the highest average expense ratios for total expenses, administration and operating expenses, and investment expenses (14%, 8% and 6% respectively). SMSFs with more than $500,000 in assets had an average expense ratio of 2% or less across all expense types.
Graph 14: 2017 SMSF expense ratios by fund size
SMSF average and median assets
The following data is taken from Data tables, table 13, table 14 and table 15.
At 30 June 2017, SMSFs had assets of just over $1.2 million on average, an increase of 27% over five years and 10% over the previous year.
The median SMSF asset size was $693,000, an increase of 12% over the previous year and 31% over the previous five years. The growth in average and median SMSF assets was consistent with that of SMSF member balances (see Graph 8).
The difference between the average and median asset figures is due to the high proportion of SMSFs with more than $1 million in assets. These have increased from 29% of all SMSFs at 30 June 2013 to 36% at 30 June 2017).
For SMSFs established in 2016–17, their average assets were $521,000, an increase of 54% over the average assets in the year of establishment of $338,000 for those established in 2012–13. Much of this growth occurred in 2016–17 (38%). Similarly, the median asset size in the year of establishment of SMSFs established in 2016–17 was $320,000, an increase of 35% over the previous year and 57% over 2012–13 (see Graph 9).
SMSF assets by age of fund
For funds established in the 10 years to 30 June 2017, the median asset value in the year of establishment was on average $213,000. Generally, the median asset values of these funds increased by 50% in their second year of operation.
For more recently established funds, the time taken to double their starting asset balance has decreased to approximately four years, compared to five years for funds established in 2007–08 and 2008–09.
SMSF assets by payment phase
For the year to 30 June 2017, 65% of total SMSF assets were reported by funds in pension phase. The assets held by SMSFs in the pension phase show a small downward trend falling by 2% over both the five years to 30 June 2017 and from 2015–16 to 2016–17.Footnote29
Over this period, there were shifts in the distribution of SMSFs by payment phase across asset ranges. There was a decrease in the proportion of SMSFs in accumulation phase with assets of $1 to $200,000 (12%). In contrast, the largest increase of any asset range in the proportion of SMSFs in accumulation phase occurred in those with assets of between $500,001 and $2 million (10%).
There were also shifts for SMSFs in pension phase over the five years to 30 June 2017, with a decrease in the proportion of SMSFs in pension phase with assets up to $1 million (8%), and an increase in SMSFs in pension phase with assets of $1 million to $10 million (7%). From 2016 to 2017 the proportion of SMSFs in pension phase with a balance up to $1 million decreased by 4% while the proportion of funds in pension phase with a balance of $1 million to $10 million increased by 4%.
Graph 15: Change in proportion of SMSFs by asset range and payment phase from 2012–13 to 2016–17
On average, over the five years to 30 June 2017 52% of SMSFs in accumulation phase held assets of between $200,001 and $1 million, and 54% of SMSFs in pension phase held assets of between $500,001 and $2 million.
At 30 June 2017, while 25% of SMSFs in accumulation phase held assets of $200,000 or less, only 6% of SMSFs in pension phase were within this asset range. Conversely, 27% of SMSFs in pension phase held assets of greater than $2 million, compared to only 7% of SMSFs in accumulation phase.
The proportion of SMSFs in accumulation phase with assets of between $200,001 and $2 million in 2017 was 68%. Similarly the proportion of SMSFs in pension phase in that asset range was 66%.
Data on SMSFs by phase and fund asset sizes can be accessed in Data tables (XLXS 503KB).This link will download a file
SMSF assets by range
The following data is taken from Data tables, table 15.
At 30 June 2017, 48% of SMSFs had assets of between $200,001 and $1 million, accounting for 21% of all SMSF assets.
Over the five years to 30 June 2017, 53% of SMSF assets were held by funds with assets between $1 million and $5 million.Footnote30 The proportion of SMSFs in these asset ranges gradually increased from 26% in 2012–13 to 32% in 2016–17.
At 30 June 2017, only 3.3% of funds held assets of more than $5 million, compared with 2.1% five years earlier, but these funds accounted for 24.6% of total SMSF assets. The proportion of SMSFs with assets of greater than $10 million was 0.7%, accounting for 10% of total SMSF assets.
Graph 16 shows a continued shift of SMSFs into higher asset ranges. For the five-year period, the proportion of SMSFs holding more than $500,000 in assets increased.
Graph 16: SMSF asset sizes, 2012-13 to 2016-17
Graph 17 shows that over the five years to 30 June 2017 the asset balances of SMSFs in their first year have increased. Of the SMSFs established in 2012–13, 51% held more than $200,000 in assets, increasing to 67% for those established in 2016–17.
There was a fall of 16% in the proportion of SMSFs established with $200,000 or less in assets from 2012–13 to 2016–17. The proportion of SMSFs established with assets of between $500,000 and $2 million increased by 12% in the same five-year period.
Graph 17: SMSF asset size in establishment year, 2012–13 to 2016–17
For information on asset sizes in establishment year, also see supplementary table 14, This link will download a fileData tables (XLXS 503KB).This link will download a file
SMSF asset allocation
The following data is taken from Data tables, table 16 and table 17.
At 30 June 2017, 53% of all SMSF assets were directly invested in Australian listed shares and cash and term deposits.
Graph 18: SMSF asset allocation, 30 June 2017
Graph 19 shows annual shifts of SMSF asset holdings by asset type as a proportion of total assets.
For the fourth consecutive year, there were decreases in the proportion of total assets held in cash and term deposits and non-residential real property. There were increases in the proportion of assets held in both listed and unlisted trusts, and residential real property over the same period.
From 2012–13, changes were made to ATO data collection methods for assets held under limited recourse borrowing arrangements (LRBAs). In 2016–17 changes were also made to data collection on borrowings specifically related to LRBAs. These changes result in better reporting and data collection for LRBAs and associated borrowings, contributing to increases in LRBA assets reported over the five years to 30 June 2017.
The value of estimated assets held under LRBAs increased to $37.2 billion or 5% of total SMSF assets at 30 June 2017, up from $8.8 billion or 2% of total assets five years earlier. Over the five years, there were annual increases of 1% or less in the proportion of LRBA assets.
In 2016–17 new data collected on the SAR showed that 30% of LRBAs had a personal guarantee.
Graph 19: Change in percentage of total SMSF assets by asset type, 2012–13 to 2016–17
Graph 20 shows the annual shifts in the proportion of SMSFs holding different types of assets.
A comparison of graphs 15 and 16 shows that in 2016–17 cash and term deposits as a proportion of total SMSF assets decreased, while the proportion of SMSFs holding these assets remained stable.
In 2016–17, the only asset class where there was a material decrease in the proportion of funds holding such assets was non-residential real property (2%). The remaining asset types remained stable or saw increases, in particular listed trusts (1.5%), LRBAs (1.5%) and 'all other assets' (1.5%).
In 2016–17, 9% of SMSFs reported assets held under LRBAs, an increase of 7% over the previous year. Most of these funds held LRBA investments in Australian residential real property (6%) and non-residential real property (2%). The remaining 1% of LRBA assets were primarily held in Australian shares. In terms of value, 93% or $34.8 billion of all LRBA investments were attributed to real property assets.
Graph 20: Change in percentage of SMSF population holding assets by asset type, 2012–13 to 2016–17
At 30 June 2017, 80% of SMSF assets were reported as directly invested by SMSFs, with 20% invested in indirect investments such as listed and unlisted trusts and managed investments. This is consistent with each of the previous five years.
In 2016–17, SMSFs with assets up to $500,000 tended to favour cash and term deposits, holding a minimum of 32% of total assets in this type of investment. SMSFs with assets of between $500,000 and $1 million tended to favour both cash and term deposits and listed shares, with these asset classes each accounting for 27% of their total assets. The largest asset class held by SMSFs with more than $1 million in assets was listed shares.
As fund size increased, the proportion of assets held in cash and term deposits and 'all other assets' generally decreased, while the proportion of assets held in trusts, other managed investments and non-residential property increased. This is consistent with the previous year.Footnote31
Larger funds tend to have a greater proportion of their assets in unlisted trusts and non-residential real property than smaller funds. However, the proportion of assets in listed shares and residential real property held by SMSFs across asset ranges is fairly consistent.
In 2017, SMSFs in pension phase had very similar assets to SMSFs in accumulation phase. The only noticeable differences were that SMSFs in pension phase tended to favour listed shares, while accumulation phase funds held a greater proportion in LRBAs.
SMSF borrowings and other liabilities
The following data is taken from Data tables, table 16.
At 30 June 2017, SMSFs had liabilities to non-members of $26.1 billion, equivalent to 4% of total SMSF assets. This consisted of $20.8 billion in borrowings and $5.4 billion in 'other liabilities', 3% and 1% of total assets respectively.
SMSFs are prohibited from borrowing, except in certain limited circumstances permitted under the Superannuation Industry (Supervision) Act 1993. These circumstances include LRBAs that meet certain conditions. 'Other liabilities' are obligations the SMSF has to non-members that do not fall under LRBA borrowings.
Over the five years to 30 June 2017, the proportion of SMSFs with borrowings increased progressively from 5% to 10%. The average amount of borrowings peaked in 2015 at $382,000 and has progressively declined to $368,000 in 2017.
In 2016–17, 9% of SMSFs reported borrowings that related specifically to LRBAs. 89% of the total value of borrowings reported by SMSFs in 2016-17 was for LRBA assets.
On average, 87% of SMSFs with borrowings were in accumulation phase, while 13% were in pension phase in 2016–17. Over the five years to 30 June 2017, the proportion of funds with borrowings in accumulation phase increased by 4%.
In 2016–17, SMSFs with borrowings were more likely to invest in property under an LRBA as well as 'other assets' and less likely to invest in shares and trusts than SMSFs as whole.
Over the five years to 30 June 2017, an average of 62% of SMSFs had 'other liabilities', with the average amount of these increasing to $16,000 at 30 June 2017 from $13,100 five years earlier.
The proportion of SMSFs with reserves fell over the five years to 30 June 2017, from 1.1% to 0.4%. The average amount held in reserves increased over the same period, from $89,000 to $195,000. We attribute this increase to better reporting and data collection on the superannuation annual return.
SMSF asset diversification
The following data is taken from Data tables, table 18, table 19, table 20 and table 21.
At 30 June 2017, 10% of SMSFs held all their investments in one asset class.
Generally, SMSFs with smaller asset holdings were more likely to have high asset concentration, with diversification of assets increasing as fund assets rise. SMSFs with assets of less than $50,000 were more likely to hold only one class of asset (41%), while less than 8% of SMSFs within each assets range over $500,000 held just one asset class.Footnote32
At 30 June 2017, approximately 49% of SMSFs held 50% or more of their assets in cash and term deposits or listed shares. Cash and term deposits were the sole asset held by 9% of SMSFs. While 6% of SMSFs invested 90% or more in listed shares, only 0.2% solely held listed shares.
SMSFs in the accumulation phase were more likely to hold one class of asset than those in the pension phase. This suggests SMSFs in the pension phase tend to hold a more diversified investment portfolio in supporting an income stream.
- Footnote 27
- Based on superannuation annual returns for 2012–13 to 2016–17.
- Footnote 28
- For more detail on income and expenses see taxation statistics at ato.gov.au.
- Footnote 29
- Data collected at 31 March 2019
- Footnote 30
- Based on superannuation annual returns for 2012–13 to 2016–17.
- Footnote 31
- ATO, Self-managed super fund quarterly statistical report – June 2017.
- Footnote 32
- ATO, Self-managed super fund quarterly statistical report – June 2018.