Let’s debunk this myth about SMSFs and global shares


A week rarely passes without a market commentator criticising SMSFs for holding only 0.5% of their portfolios in global shares. Shame on all those trustees. Apparently, SMSFs are not diversified enough, they have insufficient exposure to great technology and consumer companies listed overseas, there is too much home bias. A typical institutional investor holds 20% to 25% of a default investment strategy in global shares (see, for example, APRA’s Annual Superannuation Bulletin).

At the recent launch of his new global listed investment company, Geoff Wilson of Wilson Asset Management said, “About 65% of them [his investors] are SMSFs, which are grossly underweight international equities.” Well-known broker Marcus Padley told his readers, “the biggest difference is that rather amazingly, considering the fall in the Australian dollar, only 0.5% of SMSF money is invested in international shares.” And this week, high profile adviser Sam Henderson wrote in the AFR, “a quick glance at the ATO’s asset allocation tables will clearly illustrate that SMSFs typically invest in Australian shares and cash and have very little exposure to bonds, international shares and property.” It’s a common assertion, but it’s based on poor data.

The tiny number comes from a source that the industry should be able to rely on, the Australian Taxation Office (ATO). The latest reported statistics for SMSFs for March 2015 shows ‘overseas shares’ worth only $2.7 billion, while total assets were $595 billion, as shown in Table 1. That’s 0.5%. Unfortunately, the data is misleading and counterproductive.

GH Figure1 030715How does the ATO collect the SMSF data?

The ATO collects data on SMSFs via annual tax returns, but an SMSF can lodge its return up to a year or more after the end of the financial year. The ATO says its ‘estimates’ for March 2015 are extrapolated from 2012-2013 data, so the data is now two years old. Plus the ATO guesses at some allocations. For example, it advises, “Assets in trusts are treated as though half were invested in equities and half in property.” And all Australian.

There are obvious problems with old data, especially when the falling Australian dollar has increased the appeal of global equities since 2013.

However, the major problem is not the late data, but the categorisations. There is a wide range of global equity investments held by SMSFs which are categorised into listed trusts, unlisted trusts, other managed investments and even listed shares, and analysts are assuming these are all Australian equity investments.

Global equities are disguised in ATO data

It is obvious that SMSFs worth $595 billion must hold more than $2.7 billion in global equities, and even without knowing the exact numbers, global equities must make up a large proportion of many of the above categories. For example:

1. Managed investments or trusts

Consider the popularity of just two global equity managers, Platinum (funds under management $29 billion, mainly Australian retail) and Magellan (funds under management $37 billion, of which Australian retail is $10 billion). Both these fund managers attract significant support from SMSF trustees. The global funds of Schroders, Lazard, Fidelity, Vanguard, BT, Colonial First State, AMP Capital, Henderson, Aberdeen, Ibbotson and dozens of other popular managers have large SMSF support, not only in broad markets but sectors like infrastructure and resources.

2. Listed Investment Companies

Again, many popular LICs are global, such as Hunter Hall, Perpetual, Templeton, Platinum, AMP Capital China, Global Masters and Magellan. The new global fund from Wilson is targeting $550 million and Wilson says 65% of his clients are SMSFs.

3. Exchange Traded Funds

ETFs are increasingly popular with SMSFs as they are easy to transact on the ASX, and match the desire of many trustees to reduce costs. In May 2015, there were 129 ETFs trading on the ASX with a market capitalisation of $18.6 billion. Flows into global equities are among the top few categories. In 2014, net inflows into developed market global equities ranked first at $1.4 billion.

According to the BetaShares/Investment Trends October 2014 ETF Report, the third most common reason for investors using ETFs (after ‘diversification’ and ‘low cost’) was ‘to access overseas markets’, and an estimated 63,000 SMSFs held ETFs at that date.

What’s a more accurate number?

There is potential for ‘sample bias’ using any other source, because SMSF administration is highly fragmented among the 550,000 SMSFs. The best place to look is among the SMSF administrators which can delve ‘real time’ directly into the portfolios of the funds they administer.

Multiport releases a quarterly analysis of SMSF Investment Patterns, based on the 2,500 funds it administers. They assigned 14.4% of SMSF assets to ‘international shares’ for March 2015, a significant increase on the 10.7% from a year earlier. This is predominantly managed funds, plus ETFs and direct shares, as shown in Table 2.

GH Figure2 030715In fact, Multiport believes the global exposure may be higher, because it does not include the global equity allocation in multi sector balanced funds. On the other hand, Multiport has a large proportion of ‘advised’ SMSFs, and advisers are inclined to use managed funds. A study of the Top 10 investments by dollars shows Magellan sixth and Platinum eighth, above Wesfarmers and Woolworths.

However, another leading administrator, SuperIQ, estimates that across its 11,000 funds, only about 5% is invested in global equities, although it rises with fund size to about 9% for larger funds.

In another survey, AMP Capital’s ‘Blue Sky Report’ on SMSF opportunities, among the SMSFs which invest in managed funds, 36% say they invest in actively-managed international equities and 19% in index international equities. In July 2014, a Vanguard/Investment Trends report stated that the intention to invest in international shares by SMSFs almost doubled in the year to April 2014 from 12% to 22%.

Global equities in SMSFs much higher

SMSFs do hold more Australian shares and cash than balanced institutional portfolios, but the weaknesses in the ATO data mean there is no definitive source on the exact proportions. SMSF allocation to global shares is likely to at least 10 to 20 times the level in the ATO data. Maybe more.

In fact, the official statistics are measuring in the wrong area, because few SMSFs actually invest in global shares directly. SMSF trustees are eager to use managed funds, LICs and ETFs to gain exposure to global companies because they are far less familiar with transacting on foreign exchanges than they are on the ASX.

Given the importance of SMSFs in holding one-third of all superannuation and the retirement savings of over one million Australians, and the design of superannuation policy, the knowledge about what they invest in needs significant improvement. This applies to much of the official data produced on SMSFs.

The ATO needs to run up a few red flags about using the data. SMSFs are not as badly diversified as most claim.


Graham Hand is Editor of Cuffelinks.


, , , , , ,

10 Responses to Let’s debunk this myth about SMSFs and global shares

  1. Chris S July 2, 2015 at 8:20 PM #

    About time we nailed this one. This is one of many examples of how the ATO data on SMSFs is unreliable and inaccurate.

    The expectation is that the ATO statistics provide reliable information on the status, health and trends of the SMSF market, however the reality is that it is outdated and uninformative.

  2. GregO July 2, 2015 at 10:00 PM #

    Great article.

    As a case study, my/our SMSF is 0% Global by ATO reporting categories.

    The fund specific investment strategy recognises a Global strategy with global investments including MFG, RHC, PTM, QBE, TGG. Dollar weighted Global of 30%, actual v ATO reported number of 0%.

    Bottom line – it is time to repair ATO reports.

  3. Dugald Higgins July 3, 2015 at 8:10 AM #

    While Graham makes a valid point, I think it is important to recognise that the shift has been very recent. Using LICs and ETFs as an example, if you look at these sectors 12 months ago (June 2014), they were materially smaller than they are now. As a case in point, International ETFs were 23% of that sector and international LICs made up 17% (by value). This compares to 37% and 35% respectively today. I thinks it’s correct to highlight that the shift is occurring, and the ATO data of increasingly little use and the same old tired commentary no-longer applies in the same way. But in fairness, given these articles were appearing more than a year ago (and I’ve written them myself!), its probably having the desired effect.

    • Chris July 29, 2015 at 1:51 PM #

      I think Dugald is right. I remember back in 2009 when iShares (international ETFs) had just listed, and many of the people in my shareholder discussion group (most of whom are over 60) didn’t want to know about them and didn’t care.

      The reason ? It was ‘international’ so they couldn’t ‘touch and see’ the company, they thought withholding tax was going to be an issue, but most of all, they also were addicted to their fully franked dividends (and franking credit refunds). My experience of retirees is that most want the income stream from fully franked dividends, not necessarily capital growth (which, if you ask any American, is why you buy equities).

      The shift to international is occurring, but at the risk of being smug, the time to do it was when the dollar was around parity and the international markets were much lower; and yes, that was when I was investing to be overweight global equities. The boat has already sailed.

  4. Liam July 3, 2015 at 10:01 AM #

    I was so glad to see this myth debunked as time and time again I have had to argue with new clients that international exposure is much more widespread than reported. Typically the engineer and teacher clients are the ones who read up on asset allocations and question why my 15-35% allocation to international in warranted. On the flip side the last 3 years has made me look like a genius to them but I have had to remind them that it’s good portfolio theory not my stock picking!

    Disraeli and Mark Twain put it far better than I

    Twain :”Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: “There are three kinds of lies: lies, damned lies and statistics.”
    – Mark Twain’s Own Autobiography: The Chapters from the North American Review

  5. Peter Worn July 3, 2015 at 11:07 AM #

    Great article Graham. They never let the facts get in the way of a good story!

  6. John July 3, 2015 at 1:33 PM #

    Thanks for this clarification and a bit of clarity; and I would add that Amcor, BHP, CSL and many others are fundamentally international companies, domiciled here (and local success stories).

    Owning these can be a conscious decision to “gain international exposure”, hopefully without tax complications and sometimes even franking credits.

  7. Andrew Wakeling July 4, 2015 at 2:41 PM #

    Most of the SMSF money must be held by retirees or those close to retirement?. I guess most of the money is in relatively small funds? So the liabilities are on average relatively short and there isn’t much buffer to absorb losses or time to wait for recoveries. Aggregate international equity exposure of more than 3% would at first sight seem excessively ‘brave’. The average international equity exposure by member might reasonably be much higher, but I don’t think available statistics give us such fugures.

  8. Douglas Morris July 15, 2015 at 9:57 PM #

    Great point Graham, I agree. According to our data, for SMSF portfolios running on the Sharesight platform, the international exposure is more like 20%. Take into consideration that people using Sharesight are self-directed individuals, tending to be a bit more sophisticated.

  9. Graham Hand July 15, 2015 at 9:58 PM #

    Thanks, Douglas, I’ve been looking for more proof points. I was interviewed by The Australian yesterday on this article (they expect to run a story next Tuesday) and said I thought overall it was about 10%, so good to know I was probably conservative. 20% is at the top end of studies I have seen.

Leave a Comment:


Register for our free weekly newsletter

New registrations receive a free copy of our ebook, Cuffelinks Showcase 2015.