The art of dynamic asset allocation

Dynamic asset allocation gives greater investment control … and any investor can do it.

PORTFOLIO POINT: Only a few managed funds offer dynamic asset allocation. But any investor can set up their own DAA strategy.

Dynamic asset allocation is rippling through the investment management world, challenging the now-struggling concepts that dominated the western world’s financial advisory and investment industry for the last few decades.

DAA acknowledges that asset prices can move quickly, and that exposure to risky assets when they are experiencing high levels of volatility can be very dangerous for portfolios. DAA uses cash as the “reserve” asset, moving out of assets when they are risky, and back into those assets when their risk returns to normal levels. The slogan “risk-on/risk-off” describes DAA and also highlights the challenges for investors – deciding when to buy and sell assets, and then being able to move quickly to implement those changes, requires a far more hands-on approach than many investors are comfortable or skilled to undertake. Understanding DAA and where to get it can help you get started down this hugely important pathway – and help you respond better to some of the looming threats to global markets.

What’s wrong with modern portfolio theory?

From the early 1960s until the global financial crisis, the accepted wisdom (known as “modern portfolio theory” or MPT) was that a blend of more and less risky assets could be calculated precisely and used to generate good positive returns over the investor’s lifetime. MPT assumed that higher returns from riskier assets like shares and property would add enough value, over time, to compensate for the losses experienced infrequently during periods of market weakness. MPT uses complex calculations involving expected returns and risk to slice up the portfolio into various asset classes (i.e. “strategic asset allocation” or SAA) and then holds these asset exposures for the long term. Minor changes are sometimes made (i.e. “tactical asset allocation” or TAA) to “tweak” the portfolio – but essentially MPT hopes that by “banking” high returns in good years, returns over the longer term will be relatively smooth.

The problem with MPT is that SAA models are built using average historical data, which has been calculated over very long periods of time: and which may often predict very different results than can be experienced when markets crash. MPT is a textbook concept which assumes that markets are efficient, and when we recall the textbook definition of an efficient market, we can see why MPT and SAA struggle in the real world. An “efficient” market is considered to exist when:

  • All investors have access to all information relevant to their decisions (i.e. perfect knowledge);
  • We all react the same in relation to that information;
  • We can all borrow infinite amounts and pay interest at the “risk free” rate (e.g. RBA cash rate);
  • No one pays tax.

No wonder MPT portfolios lost massively during the GFC, given the huge information asymmetries between investors and the handful of investment banks that drove the collateralised debt obligations bubble, and the massive leverage that was available to the banks and hedge funds that profited (and then crashed) from that bubble.

Sequencing risk

The former Secretary of the Department of Treasury, Ken Henry, spoke recently about what he considers to be the “absurdity” of the conventional wisdom. Henry critiqued the idea that the returns from good years would be enough to supplement and support the portfolio during periods of poor returns, and that the diversified and relatively stable asset allocations used by MPT and SAA were therefore a suitable way to invest over the longer term. His concern is that it’s the sequence of crashes that is the problem, not just the frequency and magnitude of them. A 50% crash in the year before retirement is obviously far more devastating than a 50% crash in the first year of superannuation investing. Ken Henry’s sequencing problem has also just been outlined in a paper prepared for FINSIA by Griffith University, which also highlighted the need to better respond to market volatility, in order to reduce the problems of sequencing risk.

Enter dynamic asset allocation

JM Keynes famously said that “the market is what people think it is” – and in its most developed form, DAA recognises and responds to just that point. Instead of the textbook assumptions that MPT relies on, DAA knows that markets will move, often irrationally – driven by greed and fear. DAA definitely isn’t “day trading” but does mean that asset exposures may need to be sold down and then increased rapidly to minimise risks and loss. Some of the most promising DAA models use techniques similar to the most successful Commodity Trading Advisor (CTA) traders – futures and derivatives traders and hedge funds active in markets like the Chicago Board of Trade.

CTAs have to maintain certain amounts of collateral on deposit with central clearing houses, and experience margin calls to post extra collateral when the prices of their investments fall. To fund these margin calls they rely on sophisticated models to sell down investments (to realise cash) when their risk rises (and hopefully before prices fall). The best of these CTA risk management models use price momentum and volatility measurements to trigger asset sales and subsequent re-purchases. Instead of just relying on the expected “fundamentals” of specific assets to build portfolios, the CTA method places much importance on what the market participants think about the value of an asset: after all, that is the market.

Momentum is a simple concept. For example, we can get a good feeling for the strength of an asset’s price by comparing its long-term “moving” average price (e.g. over a one-year period) to its current moving average price (e.g. over a one-month period). If the average price of an asset was (say) $10 in the last year, if its average price in the last month is (say) $9, then we can see that it is trending down in price. (The “moving” average concept is a simple way of daily updating these numbers to cover the one-year or one-month period ending on the date of reckoning). Momentum models respond immediately to negative or positive trends by reducing or increasing exposure – ideally starting to move out of risky assets well before their fundamentals change, and preserving capital by moving funds into the “reserve” asset of cash.

Volatility measures the magnitude of price rises, and falls for particular assets, over a defined period of time (commonly one month). Rising volatility happens when price movements become bigger and more rapid and, since volatility is a measure of risk, it can also be used as a tool to guide the risk-on/risk-off decision.

When a broad range of assets are blended in a diversified portfolio, DAA tools like momentum and volatility can guide the levels of exposure to the risky assets versus the “reserve” asset of cash. For example, if momentum and volatility levels remain low and stable, the portfolio is fully invested into risky assets; but these are sold down when momentum falls and/or volatility rises. This has the effect of reducing capital losses in portfolios during times of stress, with the by-product that the portfolio can rally from a higher level when markets recover.

How can you benefit from DAA?

Investor’s with access to market data services can build simple DAA style models and, as long as their investments are liquid, DAA style re-balancing is feasible. Alternatively, there are a couple of DAA style managed funds available in Australia, including the Schroder’s “Real Return” fund and the AMP “Multi Asset Fund” (both have been reviewed in Eureka Report). Each fund invests across a wide range of assets to produce a stable return with minimal losses. The Schroder’s Real Return Fund has generated a three-year average return of 7.95% pa, and 9.63% pa in the last year. The AMP Multi Asset Fund has been active since December 2010 and has posted a one-year return of 9.74%.

DAA doesn’t aim to generate massive returns in bull markets. It will do well when markets trend consistently up but may not capture all the performance of rapidly rising prices. It aims to preserve wealth when the market risk rises, and so to better provide for long-term retirement.

Ironically, DAA has confounded the Australian financial advisory and investment industry. DAA is getting a lot of airtime in professional circles, but since it isn’t supported by the business systems that most advisors use, it is virtually impossible for financial advisers to make meaningful allocations to DAA-style investments.

In a classic case of the tail wagging the dog, because financial advice systems are built on MPT and SAA concepts, their portfolio construction tools and management systems don’t work with DAA funds or styles. Because the asset allocation in DAA funds can change on a daily basis, financial advice systems don’t know how much to allocate to DAA funds. So for now, most advisers will limit a DAA fund to the “alternative” bucket in a diversified portfolio, with a probable maximum allocation of only 5%. For now, DAA funds are best accessed by the self-directed investor, who can allocate as much of the overall portfolio as seems sensible given age, portfolio size and tolerance for risk.


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