ATO Self-Managed Superannuation Funds: A Statistical Overview

Transcription

1 ATO Self-Managed Superannuation Funds: A Statistical Overview BRIEFING NOTE Issue date 15 March 2018

2 ABOUT INDUSTRY SUPER AUSTRALIA Industry Super Australia is a research and advocacy body for Industry Super Funds. ISA manages collective projects on behalf of a number of industry super funds with the objective of maximising the retirement savings of over five million industry super members. Please direct questions and comments to: Stephen Anthony Chief Economist Phone: santhony@industrysuper.com Gary Lu Financial Analyst Phone: glu@industrysuper.com The opinions above are those of the author in their capacity as spokesperson for Industry Super Australia (ISA). ISA, the authors and all other persons involved in the preparation of this information are thereby not giving legal, financial or professional advice for individual persons or organisations. Consider your own objectives, financial situation and needs before making a decision about superannuation because they are not taken into account in this information. You should consider the Product Disclosure Statement available from individual funds before making an investment decision. Past performance is not a reliable indicator of future performance. ISA Pty Ltd ABN Corporate Authorised Representative No of Industry Fund Services Ltd, ABN , AFSL ATO SMSFs: A Statistical Overview

3 Executive Summary This briefing presents analysis on the latest ATO Annual report Self-managed superannuation funds: A statistical overview and explores key issues and trends arising from this release. The statistics reveal a sector with widespread underperformance problems, contributing to financial stability risks by adding fuel to the boom/bust property cycle through leverage, which undermines the efficiency of the national economy. The key take-outs from the statistics are: The self-managed superannuation fund (SMSF) sector achieved an average return-on-asset (ROA) of 2.9 per cent in financial year 2016 with small funds continuing to experience persistent negative returns. The average ROA for SMSF members was on par with APRA regulated funds but below the 4.1 per cent achieved by industry funds during the same period. Negative returns were widespread amongst smaller funds. SMSFs with assets less than $50,000 performed the worst (-16.7 per cent), and such fund size category failed to achieve any positive returns between 2009 and SMSF members on lower balances also faced higher average total expense ratio. SMSF membership growth slowing. Total SMSF membership grew by 3.7 per cent, down from 4.2 per cent achieved in As of June 2016, there were more than 1.07 million SMSF members and 568,315 SMSFs. Growth in net establishments slowed to 1.5 per cent with 22,643 new SMSFs a lower growth rate compared with 3.1 per cent in Average female SMSF member fund balance is 21 per cent lower than their male counterparts. Female SMSF members aged between 25 and 64 on average, have 21 per cent less assets in their accounts compared with their male counterparts in the same age range. 1 The under-25s is the only female age group with higher balances than their male counterparts. The gender pay gap and broken work patterns associated with unpaid primary care roles looking after children and elderly parents are leaving women s superannuation balances significantly worse-off over the long-term. SMSF asset allocations are still lacking in diversification. The asset allocation of SMSFs varies significantly from fund to fund. Annual statistics has shown for example, 60 per cent of SMSFs with balances under $100k have 80 per cent or more invested in a single asset class. These asset classes are more likely to be cash and term deposits, domestic listed shares and non-residential real property. This level of concentration is noticeably reduced for funds with assets greater than $1 million. Limited recourse borrowing arrangements (LRBAs) continue to grow unabated. Total assets under LRBAs grew 17.8 per cent in financial year 2016 to $25.4 billion a staggering 18-fold increase since June LRBA enables direct borrowing to acquire an investment property. This investment vehicle has been heavily promoted by commission driven real-estate promoters and tax agents as a tax-advantageous way to invest in property. Surging interests in LRBAs has raised serious questions regarding their role in inflating property prices and exposing members savings to the volatile property market. 1 Female member average balance deficit (percentage) weighted by average asset amounts across age brackets. ATO SMSFs: A Statistical Overview

4 Table SMSF statistics at a glance Total SMSFs Total members Total assets (million) Average assets per SMSF Average assets per member Average return 568,315 1,070,796 $634,683 $1,126,863 $599, % SMSF gender gap Female member balance deficit (age 25-64) Top 5 SMSF asset classes according to size Average amount Average percentage -$92,600-21% Listed shares ($187b) Cash and term deposits ($157b) Non-residential real property ($66b) Unlisted trusts ($63b) Other managed investments ($32b) Source: ATO SMSFs: A statistical overview Note: SMSF gender gap averages are weighted by fund balances across age brackets. Top 5 SMSF asset classes applies to funds in accumulation phase only. Average gender gap amount is weighted by female account balance by age bracket. Key policy implications from this brief are: SMSFs with assets less than $2 million are not viable retirement savings vehicles. SMSF members with assets less than $2 million are on average, worse off compared with members of an APRA regulated fund. A SMSF with balance between $1-2 million earned on average 2.2 per cent return on asset, while those on lower balances achieved even lower returns (SMSFs with balances less than $200k had negative returns). In contrast, industry super fund members regardless of their fund balance, saw average ROA of 4.1 per cent during the same period. Female SMSF member balance lagging their male counterparts. For the first time, ATO SMSF statistics have provided insight into the extent of the gender super gap amongst SMSF members. Maintaining an adequate superannuation balance in early to mid-stages of life is essential to allow the compounding effects of interest and investment returns to work to maximise living standards in retirement. Therefore, measures by both employers and the government are needed to address the savings gap for women, particularly for those on maternity leave and those with family caring duties who are not in a position to save for retirement. Surge in LRBAs are adding unwarranted risks to SMSF members saving and are a source of speculative demand contributing to the boom/bust property cycle. LRBAs have the potential to cause significant losses for member s retirement savings and threaten the nation s financial stability should property prices collapse in our major capital cities. Therefore, in the interest of securing member s retirement savings, such borrowing arrangements within SMSFs must be abolished or curtailed. In early January 2018, the Federal Government announced draft legislation to add LRBA loan balances (member s portion) to a SMSF member s total superannuation balance (TSB). It intends to prevent members circumventing the $1.6 million balance cap by withdrawing and re-contributing the funds as loans under LRBAs back into the SMSF. This does not prevent the use of LRBAs, it is a step in the right direction in terms of curbing LRBA activities by taxing contributions at a higher rate should the balance cap be exceeded with the addition of the loan balance. ATO SMSFs: A Statistical Overview

5 Proportion of SMSFs in asset classes Introduction SMSF membership continues to grow despite 40 per cent of funds receiving zero or negative ROA in SMSFs remain the most numerous type of fund and the sector has the greatest share of assets in the superannuation system. According to the latest ATO release Self-managed superannuation funds: A statistical overview , as of June 2017, there are 596,516 SMSFs with combined asset over $696.7 billion, representing 30 per cent of the total superannuation sector in Australia. 2 In comparison, industry and retail funds account for 23 and 25 per cent respectively. 3 The latest ATO annual SMSF statistical overview has again raised questions whether such retirement savings arrangement is good for its members given the significant problem of historical underperformance, especially compared with industry funds. Analysis of the statistics has raised doubts if the typical SMSF member received the reported average return of 2.9 per cent. Figure 1 shows approximately 86 per cent of SMSFs annual return is lower than the reported average. Underperformance and negative returns present huge risks to SMSF members, and casts a cloud of uncertainty over their retirement outcomes. Figure 1 Return on asset by SMSF size (2016) 3 25% 2 15% 1 5% Annual return for each fund asset bracket -16.7% -7.3% -3.3% 1.4% 2.2% 4.3% > $0-$50k >$50k-$100k >$100k-$200k >$200k-$500k >$500k-$1m >$1m-$2m >$2m Proportion of SMSFs receiving less than 2.9% return Source: ATO SMSFs: A statistical overview Note: SMSF members in funds with balance lower than $2 million on average, receive returns below 2.9 per cent. Approximately 40 per cent of SMSFs received zero or negative return in During the same period, the industry fund sector average return was 4.1 per cent. Examining the data closely, persistent poor performance, particularly for SMSFs with balances less than $200k, reflects a combination of factors including high level of asset concentration, high administration and investment costs. Whilst larger SMSFs benefit from scale, smaller and newly established SMSFs continue to suffer from high upfront cost and high expense ratios. Double-digit negative return appears to be the norm for SMSFs less than $50k balance (which likely contains young members with low account balances). If this 2 ATO, (2018). Self-managed super funds: A statistical overview Canberra: ATO, ATO/Research-and-statistics/In-detail/Super-statistics/SMSF/Self-managed-superannuation-funds--A-statistical-overview / 3 The remaining sectors are corporate funds with 2.5 per cent, public funds with 18 per cent. ATO SMSFs: A Statistical Overview

6 is the case, then the persistent poor performance will be detrimental to young members long-term retirement outcomes. The latest ATO statistics on female SMSF members showed those aged between 25 and 64 on average have 21 per cent less savings than their male counterparts. Alarmingly, the SMSF savings deficit for women between 25 and 34 actually increased in the five years to 2016 (Figure 15). Such a savings gap is likely to be caused by factors including women pausing their career to act as unpaid primary carers for their family (for example, raising children, looking after the elderly) as well as the issue of ongoing gender wage gap of around per cent. 4 Considering the fact that on average, women retire with 47 per cent less super than men 5, the gender super gap needs to be urgently addressed and policies need to be in place to maintain superannuation contributions during periods where women are unable to work full-time due to family care obligations. The data also exposed an alarming growth of LRBAs that magnifies the portfolio risk for members. Approximately 7 per cent of the SMSF member population held assets under an LRBA, up from 6 per cent in the previous year. In 2016, assets in LRBAs continued to grow strongly by 17.8 per cent to $25.4 billion (4 per cent of total SMSF assets), with most of the borrowing used to fund investments in Australian property. The former head of the Financial System Inquiry, David Murray, has recently weighed in on this issue and warned that high levels of SMSF borrowing under LRBAs could leave its members vulnerable to a housing downturn. 6 Murray also highlighted the risk these borrowing vehicles pose to the stability of the financial system, we didn t want Australia to have a highly leveraged banking system and a leveraged superannuant system as well, for reasons of systemic stability. Such assessment not only reflects direct concerns over SMSF borrowing, but also whether the concept of limited recourse effectively quarantines other assets from claims made by lenders in the event of loan default. This stems from the fact that LRBA loans have higher costs and often require personal guarantees from trustees. Such guarantees may prompt trustees selling other assets inside or outside the fund - to repay the lenders and so circumvents the original intention of the borrowing arrangement. 7 Therefore given the risks associated with LRBAs, we welcome the Federal Government s attempt to curb LRBAs by adding the outstanding balance of a LRBA to SMSF members TSB. 8 This is a positive step towards 4 Industry Super Australia 2017, The Gender Super Gap, URL: 5 Ibid 6 Murray warned market risk in self-managed retirement savings funds is higher than ever before and the high gearing of individual funds that have bought investment property leaves thousands of SMSFs vulnerable to a housing downturn. Roddan, M., Warning over SMSF borrowing Savings at risk as house prices fall: Murray, The Australian, 20 January The Financial System Inquiry 2014 believed LRBAs are unlikely to be effective in limiting losses on one asset form flow through to other assets, either inside or outside the fund. Commonwealth of Australia 2014, Financial System Inquiry Final Report, Canberra. 8 The Federal Government proposed legislation (consultation paper released 11 January 2018) stating any outstanding balance of LRBAs should be added to the total superannuation balance (TSB) of SMSF members. This is significant because contributions above the $1.6 million balance cap is non-concessional. Current LRBA rules allow members to withdraw lump-sum amount and lend it back for property purchases, reducing their TSB, and service the borrowing by making concessional contributions. Over time if the TSB approaches the $1.6 million cap, members can again elect to withdraw and lend lump-sum amount to fund another investment (hence the distorting effect on property market), reducing the TSB. The proposed legislation prevents circumventing the $1.6 million cap. ATO SMSFs: A Statistical Overview

7 % Return on Assets controlling the use of LRBAs, however we urge the Federal Government to take the bold step of eliminating the use of LRBAs within SMSFs as recommended by the Financial System Inquiry in Finally, the ATO annual review announced revisions to the dataset used to determine SMSF trustee structure. As of June 2017, 57 per cent of the SMSFs had a corporate trustee. 1. SMSF Performance Average ROA in financial year 2016 was 2.9 per cent, lower than the 6.2 per cent achieved in This compares with average ROA of 2.9 per cent and 8.8 per cent recorded by APRA regulated funds respectively during the two periods. Annual SMSF performance measures continue to vary across fund size. Figure 3 shows returns increase with the rise in fund asset balance. SMSF with assets lower than $2 million continue to exhibit significant underperformance. In contrast, industry funds on average outperformed all SMSFs and retail funds with a return 10 of 4.1 per cent. In particular, funds with assets less than $50,000 achieved on average per cent return compared with a return of 4.3 per cent for funds with assets more than $2 million. Historical comparison in Figure 4 showed the distinct variation in return between the smallest and the largest SMSF funds. Note since 2013, ATO data collection included new data on returns relating to non-deductible expense items and assets held under LRBA, which meant adjustments were made on subsequent annual return data. Figure 2 Historical average SMSF fund and average APRA fund returns 2 15% 1 5% 9.3% 7.7% % 0.4% 0.4% % 10.2% 9.8% 8.8% 6.2% 2.9% 2.9% -5% -1-15% % APRA Average SMSFs Average Source: ATO SMSFs: A statistical overview APRA Annual Superannuation Bulletin, June The Financial System Inquiry 2014 recommended to maintain the general prohibition on direct borrowing for LRBAs. However, Treasurer Scott Morrison rejected the ban on borrowings, citing insufficient evidence. 10 ROA calculation for APRA regulated funds involve dividing the Net earnings after tax by Cash flow adjusted net assets. The numerator has administration, investment and advice expenses deducted from investment and operating income. APRA s methodology of calculating the denominator involves calculation of net insurance flows (insurance inflows insurance outflows). Insurance outflows include premiums debited from member accounts, reinsurance premiums charged, expenses incurred for insurance claims and changes in insurance liabilities or reinsurance assets. In other words, insurance related expenses have been deducted from members assets prior to calculating the ROA. ATO SMSFs: A Statistical Overview

8 % Return % Return Figure 3 SMSF fund returns across asset ranges (2016) 1 5% -5% ,000 1,500 2,000 2,500 3,000 Industry Fund, 4.1% APRA Average, 2.9% Retail Fund, 1.6% -1-15% -2 Fund Asset Range ($,000) $1 - $50k >$50k - $100k >$100k - $200k >$200k - $500k >$500k - $1m >$1m - $2m >$2m Industry Fund Retail Fund APRA Average Source: ATO SMSFs: A statistical overview APRA Annual Superannuation Bulletin, June Note returns for APRA-Regulated funds have investment and operating expenses already deducted from investment income. Note that fund asset size ($,000) axis terminates at $3m to keep the chart within a reasonable range for formatting purposes. Figure 4 SMSF historical annual return by asset range ( ) $1 - $50k >$50k - $100k >$100k - $200k >$200k - $500k >$500k - $1m >$1m - $2m >$2m $2m & $50k spread Source: ATO SMSFs: A statistical overview ATO SMSFs: A Statistical Overview

9 % Return Figure 5 Five-year annualised return to June % 8.3% 6.4% 2.5% 4.6% 5.7% 7.2% -5% -1.4% % -15% % $1 - $50k >$50k - $100k >$100k - $200k >$200k - $500k >$500k - $1m >$1m - $2m >$2m SMSF Returns Industry Funds Retail Funds Source: ATO SMSFs: A statistical overview APRA Annual Superannuation Bulletin, June 2016 Comparing annualised 5-year SMSF returns with sectors regulated by APRA, as shown in Figure 5, industry funds maintained outperformance over the largest SMSFs. The chart showed industry funds consistently outperformed all SMSF size categories on an annualised basis. It is also important to note the sustained negative performances of smaller SMSFs during the 5-year period. Table 2 - APRA-regulated funds & SMSFs annualised return by asset range 3-years 5-years 8-years % of entities by fund size % of assets by fund size APRAregulated funds All 7.7% 7.4% 5.1% Corporate 7.6% 7.4% 5.4% Industry 8.8% 8.2% 5.7% Public Sector 8.5% 8.2% 5.7% Retail 6.6% 6.4% 4.3% SMSFs $1-$50k -15.1% -16.3% % 0.1% >$50k-$100k -5.4% -6.2% -6.2% 3.7% 0.3% >$100k-$200k -0.7% -1.4% -1.9% 8.6% 1.1% >$200k-$500k 2.9% 2.5% 1.5% % >$500k-$1m 4.8% 4.6% 3.5% 24.8% 15.8% >$1m-$2m 5.8% 5.7% 4.6% % > 2m 7.7% 7.2% % 51.7% Average 6.3% 5.8% 4.6% Source: ATO SMSF Statistics APRA Statistics - Quarterly Superannuation Performance September 2017 (historical) Table 2 compares a wide-range of annualised returns of APRA-regulated funds and SMSFs during different time-spans. It shows the average SMSF annualised ROA consistently lagged behind both the ATO SMSFs: A Statistical Overview

10 Average Return Fund Performance Index APRA average and industry average for 3, 5 and 8-year periods. Also worth noting is the fact that the majority of assets within the SMSF sector are held by large funds, and this has the effect of giving large fund returns more weight and hence skewing the average SMSF return in their favour. Figure 6 SMSF historical return index by fund sizes ( ) Industry >$2m >$1m - $2m >$500k - $1m >$200k - $500k >$100k - $200k >$50k - $100k $1 - $50k Source: ATO SMSFs: A statistical overview The blue dashed-line represent industry funds. Note this index was constructed based on annual fund return performances and does not incorporate effects of taxation. If contributions and expenses are ignored and the SMSF asset changes only from their investment returns, Figure 6 shows how an initial invested amount would perform through time. SMSFs with more than $2 million in assets performed the best up to 2016 where it was eclipsed by industry fund average while SMSFs under $50,000 will see their assets significantly eroded by consistent poor investment returns. Figure 7 Average SMSF fund risk-return diagram ( ) 15% 1 >$1m - $2m > $2m 5% >$100k - $200k >$200k - $500k -5% -1-15% >$50 - $100k $1 - $50k >$500k - $1m % % % 7. Volatility (Standard Deviation) Source: ATO SMSFs: A statistical overview Circle size indicate relative size of funds (not to be scaled). ATO SMSFs: A Statistical Overview

11 Figure 7 displays the average return and volatility of SMSFs with different fund sizes between 2009 and In the chart, circle sizes are determined by the (relative) SMSF sizes. The simple linear regression line provides a reference for above or below average risk-return performance, with circles above the line indicating more efficient risk-return profile relative to the system. We can see that smaller funds exhibited lower volatility and consistently delivered negative returns throughout the period. As later shown in Figure 17, the average operating expense ratios for smaller funds are significantly higher compared with larger funds as the former does not enjoy the benefit of scale. As previously mentioned in Key Findings, one factor causing negative returns for smaller funds is likely to be a combination of high setup costs and advice fees initially for smaller SMSFs which may exceed that of investment returns and result in a negative ROA for the fund. Table 3 5-years risk-return comparison of SMSFs and APRA-regulated funds 5-yrs average return 5-yrs volatility Industry 8.3% 5.8% Returns outperformance 1 2.4% 8% 0.6% 6% 1 8% 6% Excess volatility 1.6% 1.7% Retail 6.4% 6. 4% 2% 5.8% 5.8% 4% 2% 4.3% 4.3% Industry Retail Industry Retail SMSFs 5.8% 4.3% Return outperformance by APRA funds Excess volatility of APRA funds SMSF 5-yr average return SMSF 5-yr volatility Source: ATO SMSF Statistics APRA Statistics - Quarterly Superannuation Performance September 2017 (historical) Note: Average returns are calculated by annualising the 5-year return. Volatility is the standard deviation of annual returns Whilst initially appealing to conservative investors, the low return and low volatility profile of SMSFs hides the fact that it offers a poor risk-return trade-off compared with APRA-regulated industry funds. In the 5-years to June 2016, SMSF sector annual returns volatility was 1.6 per cent lower than industry funds, but gave up 2.4 per cent in returns on average as shown in Table 3. In contrast SMSF returns volatility was 1.7 per cent lower than retail funds while only giving up 0.6 per cent in returns. This suggests retail funds have a poor risk-return trade-off comparable to the SMSFs. 2. Asset Allocation In the 2016 financial year, SMSFs on average continued to invest a significant portion of their assets in Australian listed shares (30 per cent) and cash and term deposits (25 per cent) as shown in Figure 8. Other significant asset types include non-residential real property (11 per cent) and unlisted trusts (10 per cent). Overall, most asset classes saw their asset value increase in 2016 as shown in Figure 9. Amongst those with larger percentage increases are LRBAs (18 per cent), residential real property (20 per cent), debt securities (27 per cent) and listed trusts (20 per cent). It is not unreasonable to suggest large increases in some of these asset classes is a result of higher exposure to property investments. In the meantime, cash and term deposits, and listed shares rose slightly by 5.1 and 5.7 per cent respectively. ATO SMSFs: A Statistical Overview

12 SMSFs in the pension phase hold majority of their assets in unlisted trusts (10 per cent), cash and term deposits (24 per cent), and listed-shares (33 per cent). While pension phase LRBAs accounts for only 1.5 per cent of assets, lower than 9.4 per cent in accumulation phase. Figure 8 SMSF sector wide asset allocation at June 2016 Australian Listed shares, 3 Unlisted shares, 1% Limited recourse borrowing arrangements, 4% Non-residential real property, 11% Residential real property, 4% Other assets, 5% Cash and term deposits, 25% Other managed investments, 5% Listed trusts, 5% Unlisted trusts, 1 Source: ATO SMSFs: A statistical overview Note that insurance policy, debt securities, loans, collectables and personal use assets have been combined into other assets to simplify the above diagram. ATO SMSFs: A Statistical Overview

13 Figure 9 Percentage change in total SMSF asset by asset type from 2015 to 2016 Listed trusts Unlisted trusts Insurance policy Other managed investments Cash and term deposits Debt securities Loans Listed shares Unlisted shares Limited recourse borrowing arrangements Non-residential real property Residential real property Collectables and personal use assets Other assets Overseas shares Overseas non-residential real property Overseas residential real property Overseas managed investments Other overseas assets Total Australian and overseas assets Borrowings Other Liabilities Net Australian and overseas assets -26.6% -10.9% -9.3% -10.6% -2.3% -5.7% % 8.7% 5.1% 4.9% 5.7% 4.5% 2.8% 7.5% % 11.4% 10.5% 19.5% 17.8% 19.6% 26.9% Source: ATO SMSF statistical report June Note: Applies to assets in accumulation phase only. Five largest asset classes by size are underlined. Figure 10 Comparison of SMSF asset allocation various phases Pension Accumulation Listed trusts Other managed investments Listed shares Non-residential real property Other assets Other overseas assets Unknown Unlisted trusts Cash and term deposits Unlisted shares Residential real property Overseas shares Limited recourse borrowing arrangements Source: ATO SMSFs: A statistical overview Note: Aggregating asset types into liquid and illiquid categories, we found SMSFs in pension phase have 68 per cent liquid assets and 32 per cent illiquid assets, while SMSFs in accumulation phase have 57 per cent and 43 per cent in these categories respectively. ATO SMSFs: A Statistical Overview

14 Portion of SMSFs with 8 or more asset in one asset class High asset concentration Figure 11 Percentage of SMSFs with 80 per cent or more assets in one asset class Source: ATO SMSFs: A statistical overview Note: Most likely asset classes with high asset concentration are (in descending order) cash and term deposits, listed shares, non-residential real property, limited recourse borrowing arrangements and unlisted trusts. A lack of diversification in SMSF asset mix is concerning, especially for smaller funds. As shown in Figure 11, smaller funds are more likely to have one asset class dominating their portfolios. This likelihood decreases as we move up the fund size brackets. SMSFs with a high single asset class concentration are most likely to be holding cash and term deposits, listed shares, non-residential real property and LRBAs. 3. Member Demographics There were slight increases in the proportion of members aged between 25 to 44 while age groups above 50 experienced lower shares of membership compared with 2015 data, see Figure 12. Average income grew 2-3 per cent for SMSF members aged between 25 and 54. The highest income age group is the 45-49s and from there onwards, income steadily declines as we move up the age brackets. Figure 12 Age distribution of SMSF members in 2015 and % 3 25% 2 15% 1 5% < Source: ATO SMSFs: A statistical overview and ATO SMSFs: A Statistical Overview

15 Average Income ($,000) Average Balance ($,000) Figure 13 Average taxable income of SMSF members in 2015 and , < Average Member Balance Average Member Taxable Income 2016 Average Member Taxable Income 2015 Source: ATO SMSFs: A statistical overview and SMSF Asset Gap Between Female and Male Members While assets held by female members under 25 exceeded their male counterparts by an average of $7,610, the reverse was true for those above 25. Figure 14 shows the asset gap widened significantly between the age of 25 and 64, reaching the maximum of $156,000 for those between 60 and 64. The gender asset gap improved slightly compared with the previous year with female members in age groups and above 75s recording lower deficits. ATO SMSFs: A Statistical Overview

16 Female Member Deficit ($,000) Member Balance ($,000) Figure 14 Average SMSF female and male member balances 1,200 1, < Female Members Male Members < Source: ATO SMSFs: A statistical overview Asset Deficit 2015 Asset Deficit ATO SMSFs: A Statistical Overview

17 Member Balance ($,000) Figure 15 SMSF member balance by gender and age Age < 25 Age Age Female Balance Deficit Female Balance Deficit Female Balance Deficit Female Member Balance Female Member Balance Female Member Balance Male Member Balance Male Member Balance Male Member Balance Source: ATO SMSFs: A statistical overview As noted in the introduction, women who become primary carers for the family are often left behind in terms of their retirement savings. Analysing historical trends in gender super gap from the latest ATO statistics reveals the gap has widened in the past five years, particularly for female members in the age group. 5. Fees and Costs Smaller SMSFs continue to record higher operating expenses than funds with bigger assets. The operating expense is calculated by adding various deduction labels from its income tax return. Hence it is likely to be conservative as it does not include other expenses such as establishment costs, direct or indirect management costs and advice costs. The average operating expense ratio has steadily increased for the majority of SMSF asset classes as shown in Figure 16. In particular, ratios for funds with the least assets increased the most, funds under $50,000 recorded a staggering 14 per cent expense ratio in This appears to be in-line with the previous claim that these are newly established funds with high upfront setup costs and advice fees. ATO SMSFs: A Statistical Overview

18 Average Operating Expense Ratio Figure 16 Average operating expense ratio by SMSF fund size ( ) 16% Funds below $500k 2. Funds above $500k 14% 12% 1.5% 1 8% 1. 6% 4% 0.5% 2% $1 - $50k >$50k - $100k >$100k - $200k >$200k - $500k >$500k - $1m >$1m - $2m >$2m Source: ATO SMSFs: A statistical overview Note: Prior to 2013, ATO reporting of SMSF costs did not include SMSF expenses that were not tax deductible. In the year before the inclusion of these costs, SMSF expense ratios were 0.56 per cent on average, but increased to 1 per cent the following year when all expenses were included. This most likely explains the notable jump in expense ratios for funds above $500k. Figure 17 Administration and investment expense ratio by SMSF fund size financial year % 6% 4% 2% $1 - $50k >$50k - $100k >$100k - $200k >$200k - $500k >$500k - $1m >$1m - $2m >$2m Admin & Operating Expense Investment Expense Source: ATO SMSFs: A statistical overview Note: SMSF operating expense ratio was calculated by adding various deduction labels and dividing by the average assets to obtain the ratio. In comparison, APRA-funds record expense ratio as a sum of investment and operating expense ratio. Average operating and investment expense ratios for industry fund in 2016 was 0.4 per cent and 0.4 per cent respectively. According to CIFA Survey (2016) of 505 SMSF members, the annual median annual cost of operating a SMSF was between $2,000 to $3,000, with the range between less than $500 and over $10,000. The majority (43 per cent) of members surveyed reported spending less than one hour per month on running their SMSF Bird et. al. (2016). Who starts a self-managed superannuation fund and why? CIFA Working Paper No.127/2016 / Project no. E220. Sydney: CIFR, page 14. ATO SMSFs: A Statistical Overview

19 6. Limited Recourse Borrowing Arrangements (LRBAs) Asset size of LRBAs continued to grow in By the end of June 2016, the value of LRBA assets totalled $25.4 billion compared to $21.5 billion in This represents 4 per cent of SMSF assets. 12 Australian non-residential and residential property makeup the majority of LRBA assets with share of 45.3 and 48.2 per cent respectively. Figure 18 SMSF LRBA asset allocation Australian nonresidential real property 44.9% Australian residential real property 46. Overseas real property 0.1% Australian shares 3.1% Overseas shares 2.1% Other 3.7% Australian nonresidential real property 45.3% Australian residential real property 48.2% Overseas real property 0.2% Australian shares 2.2% Oversea s shares 1. Other 3.2% Source: ATO SMSFs: A statistical overview and Table 4 Asset allocation within LRBAs LRBA investment by asset type 2014 ($m) 2015 ($m) 2016 ($m) Australian residential real property 6,308 9,372 12,212 Australian non-residential real property 6,519 9,150 11,483 Overseas real property Australian shares Overseas shares Other Total 15,609* 21,523* 25,355 Source: ATO SMSFs: A statistical overview , ATO SMSF statistical report June Note: *Figures represent latest (adjusted) statistics from ATO quarterly report. Amounts from separate asset types may not sum to the figures with asterisks. Strong growth in real property assets under LRBAs continued in As shown in Table 4, residential real estate borrowings increased by 30 per cent in 2016, while non-residential real estate borrowing increased by 26 per cent. Growth in borrowings for overseas real estate accelerated strong in 2016 by 48 per cent. 12 The June 2017 quarterly ATO SMSF statistical report also showed by the end of June 2017, LIBA assets reached $28.6 billion an increase of 12.8 per cent. ATO SMSFs: A Statistical Overview

20 Figure 19 Average value of assets under LRBAs and average SMSF assets ( ) Overseas real property Australian non-residential real property Australian residential real property $0 $200,000 $400,000 $600,000 $800,000 $1,000,000 $1,200,000 Average Asset Value Under LRBAs* $0 $200,000 $400,000 $600,000 $800,000 $1,000,000 $1,200,000 Average Assets per SMSF Source: ATO SMSFs: A statistical overview Note: *Applicable to SMSFs that held these assets. SMSFs that does not have LRBAs are not included in the top chart. According to the annual statistics, in Figure 19, for SMSFs with residential real estate assets under LRBAs, their average or mean value in 2016 was $516,403, while the non-residential real estate average value was $1,139,462. In light of the annual ATO release, David Murray has warned Borrowing magnifies risk irrespective of the asset, and in self-managed funds, given the average size of savings versus median house prices around Australia, it is likely the property will form a significant part of the portfolio in any fund and create a concentration of risk Roddan, M., Warning over SMSF borrowing Savings at risk as house prices fall: Murray, The Australian, 20 January ATO SMSFs: A Statistical Overview

21 LRBAs Asset Size $m Median Price (CoreLogic) $,000 Figure 20 SMSF LRBA asset size and median property values 30, , , , , , Time 300 Limited recourse borrowing arrangements Australia Sydney Melbourne Source: ATO SMSFs: A statistical overview Note: Median property prices are shown by lines scaled to right axis. LRBA asset size over the past five years has grown approximately ten-fold from $2.5 billion in 2012 to $25.4 billion in 2016 (Figure 20). During this period, we also witnessed strong growth in residential property values across major capital cities, in particular Sydney and Melbourne. LRBA background The formation of existing rules on borrowing within an SMSF resulted from a series of events in the 1990s and early 2000s, which led to intense lobbying to amend prior prohibitions on borrowing with a fund. The Superannuation Industry Supervision Act 1993 prevented borrowing within a superannuation fund except in very limited circumstances. This restriction posed a problem to fund trustees wanting to invest in instalment warrant 14 issued by Telstra as part of government s privatisation drive in the 1990s. At that time, the ATO ruled such instalment warrants should not be treated as borrowings arrangements and this prompted investment banks to package a wide variety of managed funds and shares as instalment warrants for trustees to purchase. There was further uncertainty when the ATO reversed its initial ruling in 2006 and deemed there was indeed underlying borrowing within an instalment warrant. This ruling threatened those products offered by investment banks. Faced with intense lobbying, the Howard government amended the superannuation laws in 2007 which resulted in the formation of guidelines (as it stands currently) for SMSFs to borrow. 14 Instalment warrants allowed investors to own shares in a company (Telstra in our case) in the future by making an upfront payment (for a certain amount, and not the value of the shares). Upon the initial payment, the investor had a beneficial interest in the company s shares but the legal title was held by a custodian. The investor received dividends and paid interest under the instalment, with a final instalment amount to be paid for full ownership at a future date. Abbott, G. 2009, The ins and outs of borrowing Strategy, MorningStar, URL: ATO SMSFs: A Statistical Overview

22 LRBAs and financial stability concerns The surge in assets under LRBAs and the level of asset concentration associated with holding a levered investment property in the SMSF portfolio presents challenges in terms of risk management, and it could also be exacerbating the boom-bust property cycle. Leverage will magnify gains and losses and if Figure 19 is any indication of the size of the average borrowing, it also reflects how large such borrowing is in relation to the size of the average SMSF (as shown in bottom of the chart). For those SMSFs where the dominant share of total assets is taken up by LRBAs, such large allocation subjects the fund to higher asset concentration and heightens portfolio volatility. Michael Rice, CEO of Rice Warner shared similar concerns regarding the use of leverage in SMSFs. He warned SMSF members using borrowed funds to buy at the top of the market risks of waiting for decades before the capital gains are realised and the downside of such leveraged investment could wipe everything out. 15 Using debt can increase both risk and therefore return, but if you get it wrong it can wipe everything out Michael Rice Proponents of LRBAs (usually commission driven financial advisers and property spruikers) have encouraged people to move their super savings into SMSFs and take advantage of LRBAs to borrow to invest in property. Such actions are ill-advised if not reckless in the current property market like Sydney and Melbourne after years of double-digit price growth. Those highlighting the protection afforded to LRBAs argued if defaults occurred on the loan, other assets in the fund are protected as only the property involved must be sold. However, the Murray Inquiry (2014) raised concern over such belief given that lenders typically charge higher rate on LRBA loans as they deemed them riskier and require personal guarantees from trustees. Hence in scenario where the asset undergoes significant reduction in value, the trustee may have to sell other assets to repay the lender in cases where personal guarantees are involved. 16 Such action voids the idea of limited recourse and reduces the diversification of the fund. To conclude, LRBAs add additional volatility to SMSF members retirement savings, increase their asset concentration and often its risks are not well understood by members. Such risky strategy is ill-suited to a long-term retirement savings plan and we urge the Federal Government to consider an end to direct borrowing under LRBAs. 15 Roddan, M., Warning over SMSF borrowing Savings at risk as house prices fall: Murray, The Australian, 20 January Commonwealth of Australia 2014, Financial System Inquiry - Final Report, Canberra. ATO SMSFs: A Statistical Overview

23 ATO SMSFs: A Statistical Overview