Would it kill you to own fewer shares?
By Scott Francis
August 13, 2010
PORTFOLIO POINT: Cutting your exposure to the stockmarket would have little effect on overall returns, and might let you sleep easier.
When markets are trending down it’s a natural reaction to reassess your exposure to higher-risk instruments such as equities, however many investors don’t act on this because they assume the returns more than compensate for the volatility. But are they?
A high exposure to equities can be a beautiful thing in a bull market although there are times in our lives – such as when we approach retirement – where we could benefit from taking on a little less risk and consider more defensive assets.
This suggestion is radical in a country where many people own shares, but it is well worth considering, as I will illustrate.
In a recent column, Robert Gottliebsen spoke about the peace of mind that could be gained from constructing an asset allocation that relied less heavily on equities. But what you may not know is that significant reductions in your allocations to equity will only have a minor impact on your returns.
For the sake of this article, let's simplify the investment world into two possible investments – shares and cash. This can become a proxy for growth assets and defensive assets.
There’s no question that there is place for fixed interest investments, international shares, property and a handful of other assets in your portfolio, but since March direct shares and cash have been by far the biggest holdings in self-managed super funds.
Let's look at the basic characteristics of both and how we might combine them into a portfolio with attractive returns but much less risk.
An article by Paul Kerin, of the Melbourne Business School, showed that the average total return from Australian shares in the period from 1925 to 2009 had been 11.9% a year. This is consistent with other estimations by other respected academics (click here) who found that returns were perhaps as much as two to three percentage points higher over different periods.
The question, then, is whether a figure of 12% a year is a reasonable expectation for the future. I suspect that it is slightly high, assuming that inflation stays in the Reserve Bank’s 2–3% target band. The US academic Jeremy Seigel has shown that long-term returns tend to be about 7% above inflation.
From this we may deduce that in an environment of 3% inflation, the long-run returns will tend to be 10% a year. This is consistent with many commentators suggesting investors might have to temper their expectations.
Given the tax structure in Australia, where franking credits are refunded by the tax office to most (but not all) people in retirement who don't pay tax, I think it is reasonable to calculate expected returns from Australian shares as being 10% a year, plus a 1% tax return.
We all know that shares are volatile – however there are a lot more periods of very poor returns than most investors would suspect. Events like the terrible returns from the first half of the 1970s, the 1987 crash and the 50%-plus downfall recently experienced show those periods of really terrible returns happen more frequently than might be expected.
So how do we best capture this volatility in planning our asset allocation? My suggestion is that we use the worst periods in Australian stockmarket history to define downside risk; that is, we keep in mind that the value of shares can fall by 55%.
Our next investment option in the portfolio is cash. Long-run estimates of the cash return are somewhat simpler. A 6% return fits in with long-run estimates of returns from cash in Australia, and seems reasonable given the current cash rates available to investors.
The ugly part of the cash return is inflation. A historical rate of inflation a little above 4% a year means that the after-inflation return from cash is generally estimated to be about 2%.
The following table shows the trade off between expected long run investment returns and potential downside risk.
|-The risk-return tradeoff
||Expected portfolio return
||Exposure to fall in portfolio value
Let’s start by looking at the portfolios of two investors who each have portfolios valued at $1 million.
The first is extremely risk-tolerant and has 100% of their portfolio in shares. The average long-term return on their portfolio is 11% before inflation. Not bad, you say … especially when you consider our second investor, who has 100% of their portfolio in cash and is generating a return of just 6% before inflation.
It’s when you begin to consider what’s at stake that the risk-reward trade-off begins to make sense. Using the low-tide mark set during the recent global financial crisis, the first investor risks losing as much as 55% of his capital stake or going from $1 million to $450,000. The second investor has no such risk.
Let’s look at another two investors in retirement who – being eminently reasonable people – recognise a need for both cash and equities in their portfolios. Our third investor has 80% of their portfolio in shares and 20% in cash, while the fourth investor has 50% in shares and 50% in cash.
The third investor should be able to generate a return before inflation of about 10%, while the fourth would be looking at 8%. That’s a different of just two percentage points, which you could argue is almost negligible at this point of our investing career.
However the third investor is taking on markedly more risk for those two percentage points.
By returning to our low-water mark, if equity markets unleashed another round of pain then the third investor would stand to lose as much as 44% of their capital stake or a fall from $1 million $560,000. The fourth investor is only exposed to a 28% loss, or a fall from $1 million to $720,000.
I think that it is worth considering the base case of an asset allocation of 50% to shares and 50% to cash. Let's say that a person or couple has a $1 million portfolio and wants to draw on their assets at the rate of 5% a year.
Initially they would have $500,000 invested in cash. This effectively provides them with drawings for at least the first 10 years of their retirement, keeping in mind that their portfolio will be earning both interest and dividends.
This in itself is a good safety net against stockmarket volatility: having enough cash in the bank for the next 10 years, while interest and dividends accrue in your cash account, give plenty of flexibility to wait for any significant capital losses (a “long tail or returns” event) from the stockmarket to recover.
The average return from the portfolio is 8.5% a year. This means that if the drawing rate is 5% a year, the value of the portfolio should still roughly increase in line with inflation. This assumes that there is no tax payable, which would be the case for most, but not all retirees.
Investors behaving badly
There have been any number of academic papers written over the years that support the notion that asset allocation is the most important aspect of an investor’s strategy.
In 1995 Brinson, Hood and Beebower published Determinants of portfolio performance, in which they found that the total return that investors receive is dominated by decisions made on asset allocation. In 2000 a paper published by Ibbotson and Kaplan, Does asset allocation policy explain 40, 90 or 100 per cent of performance, found that “on average about 100 percent of the return level is explained by the policy (asset allocation) return level".
Combine this knowledge with the wide body of evidence that suggests investors have a remarkable capacity to “buy high” and “sell low”. Even with the encouragement of great investors like Warren Buffett, we seem to struggle to be “greedy when others are fearful” and vice versa.
Given this behaviour, it might make sense to have slightly more of your assets in cash and slightly less in shares. It will help you take a more disciplined approach to inevitable market downturns and, just as importantly, it won’t kill you.
Scott Francis' articles in the Eureka Report