Small APRA Funds. Wilson Sy Cost, Performance and Portfolio Composition of Small Funds in Australian Superannuation PAGE 0 / 20/07/2007 / VERSION 1

Transcription

1 Small APRA Funds Wilson Sy Cost, Performance and Portfolio Composition of Small Funds in Australian Superannuation PAGE 0 / 20/07/2007 / VERSION 1

2 What, Why and What What are Small Funds Funds with less than five members (about 93% have one or two members) who direct their own investments SMSF: member trustee and ATO regulated SAF: external trustee and APRA regulated Why Small Funds At Jun 2005 hold $175 Billion or 20% of all super assets At Mar 2007 hold $250 Billion or 28%, fastest growing sector What We Contribute Much needed statistical data and research More detailed and revealing APRA annual return data since 2004 to

3 What We Present Data Source Statistics Research APRA annual return SAF database particularly since 2004 Selection from 2005 population of about 6,000 funds (85%) based on 8% limit on component cash flows in the fund General information on the sample and distribution of asset sizes Distribution of expense ratio and its relationship to asset size Distribution of return on asset and with comparison with institutional fund performance Portfolio composition and its relationship with performance A performance attribution hypothesis for the observed differences between small funds and institutional funds 2

4 Sample General Statistics SAF Sample 6,225 funds (85% of notional population) About $3 billion About 9,500 members Funds with one (57%) or two (36%) members Average asset about $310,000 About 48% of funds paying out benefits of $164 million in 2005 Demographic Profile Older workers with high asset accumulation Professional and small business owners Members with confidence and familiarity with managing investments 3

5 Distribution of Total Assets 360 Distribution of SAF Total Assets June Number of Entities Log10(Total Assets) 4

6 Fund Contribution Flows Net Contribution Flows Fund Phase 2004/2005 Average (%) Positive Growing 34 Zero Mature 21 Negative Declining 45 5

7 Cost of Superannuation Why is cost Important Aggregate return is market return minus cost Alpha cancels out, only beta is left Need to measure, understand and reduce cost SAF Data Possibly the Most Accurate SAF portfolios have high proportions of direct shareholding SAF sector has substantially less incentive to under report expenses Non SAF funds have 66% assets invested indirectly through wholesale trusts and other managed funds Current APRA reporting standard does not require funds to track expenses of indirect investments Extensive use of financial intermediaries adds to invisible costs 6

8 Problems with Expense Data Accounting Relationships Starting Total Assets + Net Fund Flow + Net Earnings = Ending Total Assets Net Earnings = Gross Earnings Total Expenses Tax Accounting Uncertainty (Gross Earnings Total Expenses) can be fixed while increasing or decreasing both terms by the same amount Problem Situations Use of financial intermediaries (hedge funds, wholesale trusts etc) Soft dollar deals, free switching, late trading etc. are invisible costs 7

9 Expense Ratio = Total Expenses / Average Assets SAF Distribution of Expense Ratio Year to Jun 2005 Number of Entities Expense Ratio (%) 8

10 Expense Ratio and Total Assets Relationship Expense Ratio and Av Total Assets Relationship Year to Jun 2005 Expense Ratio (%) Log10(Av Total Assets)

11 Return on Assets = Net Earnings/Average Assets SAF Distribution of ROA Year to Jun Number of Entities ROA (% ) 10

12 SAF and Non SAF Performance Comparison Year To June SAF ROA (%) Non SAF ROA (%) Difference ROA (%)

13 SAF Performance Summary 2005 Performance Arithmetic average ROA 16.3% and asset weighted average 17.1% Out perform non SAF sector by 6% on asset weighted basis Standard deviation of ROA 10.5% versus 8.5% for non SAF sector Nine Year Performance Out perform non SAF sector 7/9 years No negative return year versus two for the non SAF sector Average compound out performance of about 2.2% p.a. Volatility is 5.9% versus 5.55% for non sector Differences are non random with 95% probability by t statistics Assuming 5% riskless rate gives a sharpe ratio of 0.64 versus 0.26 for the non SAF sector 12

14 SAF Portfolio Composition Asset allocation is a major determinant of investment performance Possibly explains observed performance differences from the non SAF sector We apply an aggregation method to convert allocations to investment vehicles to more familiar allocations to asset classes Asset Class Jun 2004 Allocation (%) Jun 2005 Allocation (%) Cash Fixed Interest Listed Equity Unlisted Equity Direct Property Diversified Trusts Other Total

15 Impact of Contribution Flows on Asset Allocation Asset Class Growing Funds (%) Mature Funds (%) Declining Funds (%) Funds Paying Benefits (%) Cash Fixed Interest Listed Equity Unlisted Equity Direct Property Diversified Other Total

16 Impact of Growth Assets on Performance 55 Return on Assets and % Growth Assets Relationship Year to Jun ROA (%) Growth Assets (% )

17 Performance Differences from Asset Allocation Small funds appear to have a strong home country bias, but the data probably understate foreign investment We minimize residual errors to find the best correlation between performance differences and differences in Australian/US equity allocation Year To June AU/US Difference SAF/non SAF Difference 15% AU/US Difference

18 Performance Attribution Hypothesis How may we explain the 2.2% p.a. out performance by the SAF sector? Assumptions Constant 10% more growth asset at 5% p.a. out performance over nongrowth assets gives extra 0.5% p.a. Constant 15% more Australian rather than foreign equities gives extra 0.4% p.a. (0.25% in capital gains and 0.15% in dividend imputation) Constant 30% more of SAF earnings are tax free due to fund paying benefits gives extra 0.4% p.a. Visible cost savings from running simpler operations as estimated by Rice Warner Actuaries on a segment asset weighted basis gives extra 0.5% p.a. Invisible costs savings as a balancing item gives the remaining 0.4% p.a. 17

19 Performance Attribution Summary Explanation Factor Return (% pa) Added Idiosyncratic Systematic Higher growth assets 10% 0.5 Home country bias 15% less foreign equities (capital gains and dividend imputation) Lower tax for more funds paying benefits Visible cost savings 0.5 Invisible and other cost savings 0.4 Total All factors

20 Conclusions The SAF sector has out performed non SAF sectors over the nine years to June 2005 due to asset allocation and cost savings Individuals with significant assets (at least $200,000) and adequate experience who are prepared to direct their own investments can save on operating costs born by institutions From summing only visible and known expenses cost is underestimated for institutional funds Poor trustee or corporate governance in the operation of some institutional funds provides opportunities for wealth transfer from the fund beneficiaries to the agents and other financial intermediaries which gives rise to the invisible costs which small funds avoid 19