The Importance of Asset Mix Decisions

Note that this is an intellectual exercise based on historical data.  It does not constitute advice as that would require knowledge of an investor's individual circumstances, such as income commitments, risk tolerance, and so on.  However, this article does highlight the importance of carefully choosing an asset class mix.

The Australian Financial Review of 17 September 2008 (page 32) published a table of historical performance by asset class from 1987 to 2007.  We have analysed this data to find some of the better performing mixes and then checked how each of a range of mixes would have performed in 2008, the year that the Global Financial Crisis (GFC) hit.

The compound return for each asset class, based on $1,000 invested at the end of 1986, is shown in the chart below.

 

Up until 1998, the best performing asset class was Australian fixed interest.  Interest rates were unusually high during the late 1980ís, helping cash to perform better than Australian shares until 1996.  International shares became the best performing asset class in 1999 and 2000 but then the tech wreck of 2000, followed by a stronger Australian dollar brought this asset class back into line with cash.  Between 2003 and 2007, Australian listed property was the star performer, although for a long time it was out-performed by Australian fixed interest.

Returns over this period illustrate the need for an investment portfolio to have a mix of asset classes.  This point was reinforced by the collapse of Australian listed property in 2008, when a 54% slump changed Australian fixed interest from the best performer to be worse than cash.  Better to have a portfolio which may slightly under-perform the maximum possible for quite a while but which can withstand a bubble burst when it inevitably happens.

Is there an optimum mix of asset classes which will maximize returns while minimizing the damage of another 2008-style panic?  Of course there is no such optimum because, while we can optimize the mix based on past performance, we canít accurately predict future returns by asset class.

However, assuming that future patterns of returns will not be vastly different to those between 1987 to 2008 we can identify some potentially good asset mixes.  The table below shows the value of per $1,000 invested at the end of 1986 for a range of asset mixes.

Name Cash % Australian Fixed Interest % Australian Shares % International Shares % Australian Listed Property % Number of negative returns 1987 to 2007 Value at 31/12/2007 Value at 31/12/2008
Typical Fund Manager

6.0

23.0

39.0

25.0

7.0

3

8,007

6,026

Optimum 1978 to 2003

6.0

0.0

32.0

7.0

55.0

3

10,111

5,448

Capital Stable

55.2

15.8

14.0

0.0

15.0

0

7,002

5,924

Growth

0.0

13.0

40.2

0.0

46.8

1

10,292

6,020

Potential 1987 to 2007

0.0

0.0

0.0

0.0

100.0

3

11,302

5,199

The Growth mix has been optimized over the period 1987 to 2007 to provide a good return and to limit the number of years with negative returns to only one over that period.  It provided the best return out of all the mixes shown (except the Potential comprising 100% allocation to property which then collapsed in 2008).  This outperformed the Typical Fund Manager Mix over the period 1987 to 2007 but fell behind slightly by the end of 2008 due to its relatively high allocation to property.

The Capital Stable mix has been optimized over the period 1987 to 2007 to provide a good return while having no years where returns were negative.  Returns were relatively low over the period 1987 to 2007 and it suffered its first fall in 2008, but by the end of 2008 it had performed nearly as well as the Typical Fund Manager mix.

The Optimum 1978 to 2003 was based on similar principles to the Growth mix but was optimized on the basis of returns over the period 1978 to 2003.  It performed quite well over the period 1987 to 2007 but suffered in 2008 due to a relatively high allocation to property.

This analysis shows how widely different asset class mixes perform over different periods of time.  One could perform very well for many years and then suffer badly in one year.

Our next report on this subject is based on our search for an asset mix which performed better than those above over the 1987 to 2008 period and we evaluated the 1987 to 2009 returns for a range of mixes.

This analysis has focused on performance for a one-time investment.  We have also analysed the asset mix performances for other situations including regular saving and drawing down an investment.

The updated analysis can be purchased online at www.foreseechange.com.au.   

 

 

Charlie Nelson, Director Foreseechange, May 2010

Home