It’s said that one cannot time markets and most people who have tried have discovered this for themselves.
However, does this mean financial planners should never have a view on markets? Should portfolios be identical at all times, no matter what relative value different assets offer? Should asset allocation be the same in high and low inflation environments, or whether markets are at record peaks or troughs?
Surely planners must exercise judgement. This cannot be avoided, even by those who use diversified funds. If growth funds are selected, the planner has chosen a fundamentally stock market-based asset allocation, whereas capital stable funds will favour fixed interest. Fund managers merely play at the edges around their central asset mix mandate.
In order to discuss this subject meaningfully it is necessary to define market timing. I am using the term to describe a strategy based on predicting the direction of markets over periods of less than three years.
In the short-term, markets depart from fundamental value. Predicting such random variations is impossible. Yet, in the long run, reality prevails. Although there have been exceptions, three years is usually long enough for fundamentals to re-assert themselves.
It is not necessary to be able to time short-term market movements in order to give sound investment advice, however, making sound judgements about long-term trends and fundamental values is essential.
For example, in real terms, Australian bonds lost 6.3 per cent per annum during the 1950s and 5.5 per cent per annum during the 1970s. US shares lost money in real terms from 1910 to 1999 and both American and Australian shares did so in the 1970s.
To have been significantly exposed to these assets at these times was not poor short-term timing, but a bad long-term strategy.
It is not always easy to predict changes in economic fundamentals which can cause sustained poor performance. However, there are instances where underperformance is predictable, ie when a market is so over-valued that a reversion to the norm makes declines inevitable.
The most recent case was the US stock market, particularly tech stocks. Most dot coms had no prospect of making a profit, and yet were selling at high prices. Poor investment returns were inevitable.
This was not a question of market timing. No-one knew when the mania would pass from the market, however, that it ultimately would, was inevitable. The long-term outperformance of shares in general doesn’t make it profitable to invest in any stocks or in markets at excessive prices.
This remains relevant today. In my view, investing in the US market with a price earnings (PE) in the mid-20s or above (or investing in a mix of international equity trusts with index-like exposure to the US) makes sustained underperformance a real possibility.
At writing, five-year Australian bonds guarantee were around 5.5 per cent. One would invest in shares only with a reasonable prospect of outperforming bonds.
A tool developed by Jack Bogle (of Vanguard) is useful in assessing this prospect. Bogle established that the return from the US market over decades was consistently similar to a result calculated by adding together three figures: dividend yield at the start; the annual rate of profit growth over the decade; and the return produced by the change in PE over the period.
Macquarie has applied the same method to Australian data. Table 1 shows that, over 30 rolling 10-year periods, the Bogle method averaged within 0.7 per cent of the market return. Thus, if we can estimate these three figures for the next half decade, we may form a realistic expectation for returns.
In the US since World War II, the PE for the S&P 500 has been as low as seven (on trailing earnings) and hit peaks in the low 20s in 1961 and 1987. In the late 1990s, it passed 30, which was unprecedented.
The long-term average is 15. What rationale could there be for believing a superior return could be obtained from buying shares at double their historical average PE?
On May 31, 2002, the PE was 43.22 but, to use moderate figures, let’s assume only 25 times prospective earnings. I’ll apply the Bogle method to three scenarios: PEs remaining at 25, or declining to 20 or 15 over five years.
There is no precedent for PEs remaining at 25 for so long a period. However, if they remained constant this factor would contribute zero to returns.
Assuming some reversion towards the norm, PEs falling to 20 wouldcause a loss of 4.4 per cent per annum. Again, before the current period, there is no precedent for PEs continuing to hold above 20 for so long, so this scenario requires the optimistic hypothesis that history is irrelevant.
Should PEs revert fully to their long-term average of 15, this would cause a fall in prices at the rate of 9.7 per cent each year.
We now have a range of possible figures for one factor in Bogle’s formula. The second factor is initial dividends, which are around two per cent and the third is profit growth. What growth is required to achieve our target of 5.5 per cent per annum?
Table 2 shows that, should PEs remain at 25, 3.1 per cent profit growth is required. This is a modest number, however, this scenario requires great optimism about the sustainability of extreme PEs, that is to say that ‘this time it’s different’.
If PEs fall to 20, profit growth of 7.7 per cent is required. This is possible but not certain, but it again requires optimism about PEs.
If multiples revert to their average, profit growth of 16 per cent per annum is required. This is improbable.
Thus, optimism about PEs or profit growth is required to believe that US equities will exceed Australian five-year bonds. This analysis contrasts sharply with the naive proposition that the US market must recover because it has fallen for two years — ignoring precedents to the contrary.
This raises the fundamental question about whether financial planners should exercise judgement about the relative value of markets.
If you think having a view on value is the same as market timing, consider whether you would review exposure to Australian shares if, overnight, with no change in profit, the All Ords hit 3,500, or 5,000, or 10,000. If the answer is yes, and how could it not be, then this acknowledges that value judgements about markets are required.
Considered judgements need to be taken into account in portfolio construction, along with truisms about the dangers of attempting short-term timing and that, over really long periods, equities outperform.
While there is much more to a financial planner’s service than investment returns, competitive returns are essential. No planner can deliver consistently poor absolute returns or underperformance of competitors without putting client relationships at risk.
Giving quality investment advice is difficult and few of us avoid all mistakes. Inevitably we need to exercise judgement (personal or out-sourced). This requires a knowledge of investment history that covers more than the buoyant last quarter century.
