Investment Road Test: AMP Multi-Asset Fund

PORTFOLIO POINT: Using dynamic asset allocation to dodge volatility has helped this fund post an impressive return in its first year. If it can continue, it will be a great addition to portfolios.

One of the hardest GFC lessons for traditional fund managers has been to acknowledge the weaknesses in the asset allocation models that drive their investment mix. The usual approach (derived from a 1960s concept known as Modern Portfolio Theory, or MPT) selects assets when they are “cheap” and sells/takes profits when they are “expensive”.

The problem is that the maths behind MPT assumes that asset prices behave predictably and that value can be anticipated by comparing with long-term averages. Events like the GFC are seen as anomalous by MPT adherents, who tend to ignore the problem of short-term volatility in favour of the “security” of the world of averages.

Thankfully, progressive fund managers are beginning to design investment funds that are flexible and nimble, making them able to significantly increase cash holdings during times of anticipated or actual increased volatility. AMP Capital has launched its own version of this type of fund, which explicitly uses the innovative “dynamic asset allocation” (DAA) approach as part of an armoury of risk-management tools. For details for the AMP Multi Asset Fund, click here.

The problems with MPT were first articulated in Australia by a mainstream investment house in January 2010 when JB Were published a strategy paper calling for “far higher than usual allocations to cash” and “permanently lower allocations to equities”. JB Were also advocated higher allocations to hedge funds to seek the higher equity-style returns, without the usual equity market risk.

It did so as a result of analysing the reasons why MPT held high levels of equities exposure, despite the deepening market volatility during 2008. Although the JB Were paper tried its best to let MPT “wriggle off the hook”, (saying that MPT isn’t actually designed to work during periods of market turmoil), it did sow the seeds of debate that by now has had far-reaching consequences. Analyses like this have now proliferated and led to the emergence of the DAA approach.

To understand the significance of DAA we need to understand what it has replaced. MPT proponents think of the markets like a gigantic casino – but not of the wasteful speculation typical of most punters – where high frequency and repeated activities start to conform to predictable patterns. The maths behind MPT relies on the notion of “reversion to the mean” – where the outcomes of seemingly random events actually express repeatable outcomes.

For example, a game of Two-Up will over the long term see an exactly equal distribution of heads AND tails; even if a punter during the short term sees a long repeat of heads OR tails, eventually the pattern will become exactly 50:50.

Mean reversion adherents expect that asset prices will rise if they are below the long-term average price for the asset, and that they will fall if they are below the long-term average. For example, if a share or market normally trades on a price/earnings (P/E) multiple of 15 times, then MPT proponents will expect that stock to rise in value if it trades below 15, or to fall if it is trading above 15. The model includes a number of filters and factors that determine when and if an asset class is to be bought or sold, but it essentially expects that history will repeat itself, eternally.

Critics of mean reversion are taking on an investment establishment that enshrines it and MPT as the holy grails of investment science. To suggest that mean reversion may not continue to work as strongly as it may once have done so is taken as heretical by traditionalists.

This makes it all the more tantalising to notice that investment gurus such as Jeremy Grantham of GMO is now openly on the record as debunking mean reversion – citing as he does the permanent inclusion of emerging markets like China and India. Grantham doesn’t hold the view that China and India are on one way growth paths, but he doesn’t need to in order to make the argument that investing markets have permanently changed.

In the context of commenting on the permanent and insatiable growth in demand for commodities arising from these countries, Grantham in April 2011 said:
“The world is using up its natural resources at an alarming rate, and this has caused a permanent shift in their value. We all need to adjust our behavior to this new environment. It would help if we did it quickly'¦From now on, price pressure and shortages of resources will be a permanent feature of our lives. This will increasingly slow down the growth rate of the developed and developing world and put a severe burden on poor countries.”

MPT advocates a portfolio diversified across multiple asset classes to insulate the investor from the risk of loss, and it also assumes that different assets won’t behave in sync with each other. It’s clear that game-changing events such as the GFC (or the tech-weck or the September 11 terrorist attacks) hit all asset classes such that they were highly correlated, which was one of the factors that led to creation of DAA.

The frequent bouts of severe market disruption since the bottom of the GFC have become the fillip for widespread adoption of the DAA approach, as retail investors have voted with their feet, by selling out of poorly performing traditional funds and increasingly placing their money into the safe haven of cash.

Without wanting to labour the point, the increasingly fallacious outcomes from traditional MPT adherents are starkly evident in the post GFC world. Typical international fund managers have been increasing their exposure to developed world equities since the middle of 2010. They have been doing so because their models tell them that first world stocks are cheap, because they are trading below their pre GFC valuations.

Mean reversion expects these valuations to increase – because they have done so in the past, they are expected to do so again. But this blind faith ignores the far-reaching and potentially permanent shift in economic power, from the developed nation to the emerging markets. The US, for example, should recover in time, but it may never be the same as it was before the GFC. The damage and indigestion is far deeper in many eurozone countries, and when assessed on their present fundamentals, most of these economies don’t deserve investment.

Enter DAA and funds that embed it, such as the Multi Asset Fund, which contains a number of very important breakthrough concepts. The Multi Asset Fund does not measure its performance against arbitrary benchmarks like the general stockmarket index.

The stated ambition of the Multi Asset Fund is to produce a return of 5.5% above the inflation rate over a “rolling” five year period (we’ll come back to the notion of “rolling” return periods). The aim of doing so is stated by AMP as being “logical because it is inflation that portfolios must outperform in order for an investor’s wealth to have at least the same purchasing power in the future as it does today”.

The Multi Asset Fund relies on three layers of investment science to improve its performance. By setting its objective to provide inflation plus 5.5%, it frees itself from the normal structures of sharemarket-based funds. It can select from a wide range of investment assets, including shares as well as credit market instruments, hedge funds, infrastructure, property and cash.

Some of these may be far riskier than shares, so the skills of the manager are important in this fund. A full list is set out on the product website.

As well as allowing for wider than usual asset class diversification, the fund can also use a range of investment vehicles, including unlisted or direct investments (presumably the appetite for these is tempered by their relative illiquidity). It is likely that last year’s returns were enhanced by inclusion of exposure to developed nation “high-yield debt”, such as corporate and sovereign bonds. And finally, in line with the core idea of DAA, the fund can increase or decrease its exposure to these risky assets, to minimise the volatility and risk of loss.

DAA has spawned its own lexicon including the fashionable concept of “risk-on/risk-off.” This simply means that investors now have the encouragement of moving into and out of assets to avoid the risk of loss arising from structural or broad market disruption. This isn’t seen as using a trading approach to generate revenues, more so it is an acknowledgment that the old adage of “time in the market, not timing the market” is far less viable now than it once was.

And here is the ultimate value provided by funds like the Multi Asset Fund. The problem for normal investors is that they don’t typically have the skills to decide when to increase or decrease exposure to risky assets, nor do they have the capacity to swiftly buy or sell a broad range of assets quickly.

The precise method the Multi Asset Fund uses to guide its DAA isn’t disclosed publicly, and this “black box” may be one of the reasons why inflows to the Multi Asset Fund have been slow. Investment managers right now are debating the utility of measures like volatility targeting and momentum to predict and manage risk, and it’s likely that these or variants are part of the DAA toolset that the Multi Asset Fund is using. Some commentary on the methodology used by AMP for this fund would be welcome.

The fund deploys DAA within the context of a unitised investment fund, where portfolio wide allocation changes are implemented by the manager for the benefit of all unit holders. Since the Multi Asset Fund uses a traditional “managed investment scheme” structure, high levels of turnover can mean that unit holders suffer from a relatively high tax bill, some of which may not receive the benefit of concessional CGT treatment. That may be a small price to pay if the fund performs in line with its stated objectives.

The Multi Asset Fund is little more than 12 months old and so it’s hard to comment on its performance. Fees for retail investors (direct investors, without a financial adviser) are a modest 1.15% including GST. Some of the underlying fund managers in which the Multi Asset Fund invests may charge base as well as additional performance related fees; these are deducted from the performance of those funds so don’t represent an additional impost to Multi Asset Fund investors.

Actual performance data is only available for the “adviser only” version of the Multi Asset Fund, with the slightly lower than retail fee load of 0.98% pa, the adviser version has produced returns of 7.86% for calendar 2011. Since this period included some profoundly volatile periods, if the Multi Asset Fund can continue to generate such returns, it will be a great addition to portfolios.

The score: 4 stars
0.5 Ease of understanding/transparency
1.0 Fees
0.5 Performance/durability/volatility/relevance of underlying asset
1.0 Regulatory profile/risks
1.0 Innovation

Tony Rumble is the founder of the ASX-listed products course LPAC Online. He provides asset consulting and financial product services with Alpha Invest but does not receive any benefit in relation to the product reviewed.