Skip to main content
rss feedour twitterour facebook page linkdin
The Right Portfolio - Eureka Report

November 20, 2006
The right portfolio
By Scott Francis

PORTFOLIO POINT: Investors too often try to pick stocks and time the market, usually disastrously, where they would enjoy stronger returns by constructing a portfolio and sticking with it.

Asset allocation is a key driver in building a successful portfolio, a point that was stressed by Eureka Report's superannuation editor, Trish Power in a recent column, Art of allocation.

Today I examine some recently updated research that highlights the inability of investors to pick and choose asset classes, again pointing to the importance of a sound long-term asset allocation. I will then explain how to build an asset allocation for your portfolio.

Let's start quickly by reviewing three pieces of academic support for the notion that asset allocation is a key driver of investment returns.

In the Financial Analysts Journal in 1991, leading investment analysts Gary Brinson, Brian Singer and Gilbert Beebower updated the 1986 article written by Brinson, Beebower and Randolph Hood, Determinants of Portfolio Performance, which examined 91 US pension funds from 1974 to 1983. The results were conclusive, with over 90% of the variation in returns explained by asset allocation; 4.6% by security selection and 1.8% by market timing.

More recent studies by Roger Ibbotson and Paul Kaplan of Ibbotson Associates Does Asset Allocation Policy Explain 40, 90, 100 Percent of Performance? published in the Financial Analysts Journal in 2001; and Another Look at the Determinants of Portfolio Performance by Craig French (available at, found that asset allocation policy explains more than 90% of the variation in total portfolio return. This supports the original study by Brinson, Hood and Beebower.

How well do investors choose asset classes?

Dalbar, a US investment firm, produces an annual report that outlines how successful the average managed fund investor in the United States has been. The following graph shows that between 1985 and 2006 the average managed fund investor received a return of 3.90%, against a return of 11.90% from the index. Why is this return so low? Primarily because investors do a terrible job of deciding when to buy into and sell out of asset classes.

When share values are falling, there is a strong tendency for investors to want to sell assets. When share values have risen strongly, the tendency is to want to buy more assets. The results, with the average investor receiving less than one third of the returns on offer from the market, are spectacularly bad! This is another plank in the argument that investors should thoughtfully build an asset allocation and then stick to it!

mIndex a clear leader

Bad asset allocation in practice

At an anecdotal level, I know of many people who thought that after the strong sharemarket performance in 2003-04 and 2004-05, it was time to take a break from the stockmarket. They had received a total return during this time of about 60% and felt it was time to lock in these strong gains.

Since then markets have continued to be strong, underpinned by above average growth in corporate earnings. Those people who sold out 15 months ago are faced with two problems. First, they had their actual returns reduced because of the capital gains tax costs associated with selling shares. Second, when and how do they buy back into a market that has shown returns of more than 25% since then?

Jeremy Siegel is a US-based finance professor at the Wharton School and author of the book Stocks for the Long Run. In a paper entitled The Shrinking Equity Premium, published in 1999 in the Journal of Portfolio Management, he estimated that the long term price/earnings (P/E) multiple of the stockmarket should be about 14, equating to a real (after inflation) return of 7%. On this valuation metric, the Australian sharemarket seems to be fairly valued, which will make it tough for the people who tried to time the market in 2005 to buy back in again at similar values.

If you can't time markets, how else do you manage volatility?

The evidence presented earlier in the article saw that investors struggle in their bid to time investment markets. Given that this ability does not exist, then how should investors cope with volatility?

The first answer is to accept volatility as a reality of investing in growth assets such as shares and listed property trusts. Volatility is what is rewarded with higher investment returns. You can choose lower investment returns with no volatility (eg, cash) or higher investment returns with volatility (eg, share-based investments).

The second answer is that volatility can be moderated, but not eliminated, through diversifying across asset classes.

Building an asset allocation

Step 1: Fixed interest and cash exposure. For most investors the returns from fixed interest and cash are not particularly attractive at the moment. The best cash management trusts and online cash accounts are returning 5.5-6%. This return is completely taxable: a 6% return for an investor on a top marginal tax rate of 46.5% becomes 3.2%, only just beating inflation of around 3%. The returns from fixed interest investments, assuming that you are not indulging in the risky world of unsecured notes, debentures and non-investment grade bonds, is not much more attractive. Similar to cash, the return is tax-ineffective.

So why would we ever incorporate cash and fixed interest returns into a portfolio?

There are two main reasons: to achieve liquidity and to dampen volatility in a portfolio. Liquidity refers to the ability of an investor to access cash from their portfolio: having enough cash and high-quality fixed interest

Fixed interest and cash also "dampen" the volatility in a portfolio. Let's compare two portfolios, one made up of 100% growth assets and one made up of 50% growth assets.

In the 1987 sharemarket crash, growth assets (Australian shares, listed property, international shares) fell in value by about 35%. For people who had 100% of their investments in growth assets, their portfolio fell in value by about 35%. For those people with 50% of their investments in growth assets, their portfolio fell in value by 17.5% when growth assets fell in value by 35%.

Reducing volatility sounds great, however there is a risk involved in putting too great a proportion of your assets in defensive assets. The risk is that the investment earnings from defensive assets are so low, that they returns from your portfolio don't effectively keep pace with inflation.

As you can see, the balance comes down to the inevitable investment tradeoff between risk (volatility of growth assets) and reward (higher investment returns). Once you have decided how much of your portfolio to apportion between defensive and growth assets, the next step is to decide which proportion of growth assets to hold in your portfolio.

Step 2: Growth Asset Classes. The three main growth asset classes I use in investor portfolios are: Australian shares, international shares and listed property trusts.

Direct residential property, also a legitimate growth asset class, tends not to be incorporated into the portfolios I use for clients because:
  • Most clients already have direct property exposure through their residence.
  • Exposure usually involves borrowing, which is not suitable for many of the clients I see who are close to retirement.
  • The current yield on direct residential property is fairly low, and therefore doesn't provide an attractive income stream for funding retirement.

So how does a portfolio get the correct balance between Australian shares, international shares and listed property trusts? The following graph shows the average annual returns from each growth asset class in the 26 years between July 1, 1980, and June 30, 2006.

mGrowth asset classes, average annual returns, 1980-2006

Australian shares returns are based on the All Ordinaries Accumulation Index; international shares on the MSCI World Index Accumulation Index; and listed property trusts on the ASX Listed Property Trust Accumulation Index.

The reason we start at 1980 is because listed property trusts only began in the 1970s, and the listed property trust index is reported from 1980.

What is clear is that the returns from the three growth asset classes are very similar, only a 1.06% difference between them over a long time period. It is this similarity of returns that suggests the basis of the growth asset allocation of a portfolio should be exposure to all three growth asset classes, because they produce similar investment returns. Being exposed to all three will reduce the overall volatility of the portfolio, because as one asset class does poorly the other two might be able to compensate. Of course, there will still be times (such as 1987-88) when all three growth asset classed produced negative returns. Interestingly, this is the only year since July 1980 where all three growth asset classes have produced a negative return.

So should the growth assets be simply allocated 33% to each asset class?

There is a case to favour Australian shares, as the income paid by Australian shares includes the tax benefit of franking credits. There is research that suggests this benefit is not priced by the market, providing a "bonus" return. On this basis, an allocation of 40% growth assets to Australian shares, 30% to international shares and 30% to listed property trusts is reasonable.

Asset allocation in practice

Let us take the example of a couple, about to retire, with $700,000 in assets.

They decide that they need $40,000 a year to live off, and want to put seven years of income ($280,000) aside in defensive assets such as cash and high-quality fixed interest investments, knowing that they these assets will be secure. This will let them sleep at night without having to worry about where the next few years' income will come from.

This equates to having 40% of the assets in defensive investments. It also means that if there were some sort of economic shock and growth asset fell in value by 30%, their portfolio would fall in value by about 18%, and they are comfortable with this level of volatility.

Within the defensive assets they want to keep $70,000 in cash, preferring the immediate liquidity that it offers, along with reasonable current returns of 5.5-6% a year. This equates to 25% of their fixed-interest investments. The remaining 75% of their defensive assets ($210,000) is invested in high-quality fixed interest investments, targeting a slightly higher earning rate than the cash rate.

Within the growth assets, they agree with allocating 40% ($168,000) to Australian shares, 30% to both international shares and listed property trusts ($126,000 each).

mThe case study portfolio
Asset Class
% of Portfolio
$ Value 
mFixed Interest
mAustralian Shares
mInternational Shares
mListed Property Trusts

It is interesting to compare this asset allocation side by side with the asset allocation proposed for long-term investors in the Eureka Team's book Making Money. Its allocation is for a superannuation portfolio, and therefore a slightly longer time horizon. Therefore, it has a slightly higher exposure to growth assets.

mAllocation strategies compared

Lower volatility, better compounding

There are two reasons why an investor would want to reduce the overall volatility of their investment portfolio. The first is that for a given level of return, people would generally prefer to have less volatility. For example, if an investor had the choice between a 10% return each year, or an average return of 10% that was made up by a 30% return one year and then a minus 10% return the next, the choice would be the less volatile return of 10% a year.

Second,  this is important, there is a greater compounding effect with portfolios with lower volatility. This means a lower standard deviation of returns over time leads to a higher ending investment balance given the same average return.

An example of this is to compare $10,000 invested in an international share portfolio that mirrored the index return between July 1970 and June 2005. The portfolio had an average return of 13.58% a year and grew to $461,000 over the period. A diversified portfolio that was invested in 50% Australian shares and 50% international shares through to June 1980, and was then invested in 33% Australian shares, 33% international shares and 33% listed property trusts (once the listed property trust index was started) had an average return of 13.37% a year. Here is the twist: even though the average return was lower, this portfolio had a lower level of volatility, which led to a greater compounding of returns over time and the portfolio had an ending balance of $578,000, more than $100,000 greater than the portfolio with higher average returns but with more volatility.


As investors, we spend a lot of time trying to find underlying investments that might provide an additional 1% or 2% of returns here or there. This is a worthwhile pursuit but might distract us from the main game of choosing an asset allocation that meets our needs. Choosing an asset allocation to meet your needs is a powerful tool in building a successful financial situation.