A question of timing

By MICHAEL LAUREN*

Whatever happened to "It's time in, rather than timing"?

Like a mantra, this is what small investors with their savings tied up in managed funds which invest overseas are supposed to be chanting right now to soothe their fears.

The saying is, of course, a reference to the standard patter of financial advisers and fund managers. That is, that it is better to stay in a market for as long as possible than it is to pick your entry and exit times. No prizes for working out that it is also very much in their interests for you to hand over your money to them for a long time.

If you have been watching international markets swing backwards and forwards over the past few months, you might be wondering if there is anything to this financial theory.

Let's look at the rationale behind it. Everyone knows that the market fluctuates up and down from day to day, with no particular pattern. For practical purposes (especially from the point of view of a small investor), whether the market moves up or down on a given day is unpredictable and often irrelevant. This being the case, such fluctuations might as well be regarded as random.

In the background to these "random" fluctuations, a given company may be making more or less money than it did in previous years, affecting its share price. This movement in the real value of the company's shares should drive the long-term trend in its share price. Hence the random short-term movements can be seen as having little effect when averaged over time.

That sounds completely reasonable. But history suggests not all is right with this picture. Take a look at the US stock market.

The Dow Jones index has increased from 307 points at the start of 1929, to 10073.4 at the start of this year. That is a capital value increase of 3181 per cent after 73 years. This capital growth equates to about 4.9 per cent per year. You might expect to add a further 1.5 to 2 per cent to this after dividends, giving you a total return of about 6.5 to 7 per cent a year over the long term.

The question is, how long is long term?

At face value, you might think that since the market is increasing on average for the whole period, it would not particularly matter when you invested.

Let's suppose you go to university and become a professional. You graduate at 23, and spend the next few years paying off your student loan. By the time you are in your early 30s, your career is in full flight. You decide to put some money away for the future. You want to get your money back while you are still relatively young. A good period of time to invest, then, might be 25 years, so that you will be somewhere between 55 and 60 years old when you take your money out.

However, when we look at possible 25-year periods, we find something surprising. Twenty-five years is not long enough to average out the up and down motion of the sharemarket. That is, returns after 25 years are quite variable. For example, from 1977 to 2002 the market rose 908 per cent (or nearly 9.7 per cent a year, on average). From 1952 to 1977 the market rose 270 per cent (5.3 per cent a year), and from 1929 to 1952 it actually fell 12.1 per cent.

To understand what is going on here, you need to be introduced to the concept of "intermittency". What this means is that the biggest movements in a financial market are confined to limited periods of time.

THE reason this is so is that the short-term fluctuations in share prices are not disconnected from the long-term movements, as assumed earlier. Rather, the effect of the long-term movements "cascade" down into the short-term fluctuations. That is, the 25-year movements have some effect on the size of the 12-year fluctuations, the 12-year fluctuations affect the 6-year fluctuations, the 6-year fluctuations affect the 3-year fluctuations, and so on, all the way down to movements of less than a day.

When a cascade process like this exists, it leads to high variability. It also means that the greatest changes in value will occur close together in time.

Although US mathematician Benoit Mandelbrot pointed this out several decades ago, this subtle insight into the nature of financial markets is not widely understood. It turns out that intermittency is a nearly universal property of complex systems, that is, systems that depend on a large number of variables. Nearly exactly the same patterns seen in financial markets are also seen in wind and rain data, for example.

In practical terms, intermittency means that "time in" will not necessarily bring you the sort of returns your financial adviser promises. The element of risk remains.

Likewise, you cannot expect the spectacular rises on Wall St of the past two decades to continue.

In fact, there have been only two periods of more than a decade since 1929 when the US sharemarket was rising at a relatively constant rate for an extended period: the first starting in the 1950s, and the second starting in the mid-80s.

The bottom line is that if you missed out on these periods, your return on your investment would look miserable. Only occasionally does the market erupt into action and move greatly one way or the other. It is during these eruptions that the real money is made or lost. This volatility is only averaged out over the very long term (i.e. longer than 25 years).

The great crash of October 1929 is another example. The market initially lost nearly half its value, before briefly recovering. However, five years later the market had lost nearly 90 per cent of its value. That kind of dramatic movement in a relatively short period overwhelms the much quieter years before and after. Intermittency means that when it rains it pours.

People often think that the great crash of '29 was a one-off. But these events are mathematically inherent in the nature of financial markets.

Similar slumps have happened to other sharemarkets around the Pacific Rim during the late 80s and mid-90s. The slump on the Japanese market beginning at the end of the 80s follows a remarkably similar pattern. Before the crash, the Nikkei index reached levels of nearly 40,000. At the start of this year it was around 10,000.

When viewed over a decade, the final levels of the Nikkei and the Dow Jones after both crashes are very similar.

New Zealand is no exception. The New Zealand sharemarket rose dramatically in the three years leading up to 1987, much more so than most international markets. However, the crash that followed was in proportion to these gains. It could be argued that the New Zealand market is yet to get over the aftershock, as the market has barely moved over the past decade.

As difficult as it may be, getting the timing right is important.

The implications for small investors are significant.

Much has been made of the merits of managed fund investment, particularly given the strong performance of Wall St in the 90s. The fall in the value of the New Zealand dollar has made these returns appear even more attractive. But if these strong returns are based on only a decade-and-a-half worth of rises, a more realistic goal might be to hope for the 6 per cent or so (after fees) return that one might expect over the very long term from a managed fund investment.

This leaves small investors to consider if they should manage their share investments themselves, trying to avoid downturns and exploit bull markets. It also makes traditional investments look more attractive. Fixed-interest rates of more than 6 per cent are frequently available in this country, and good returns from property investment.

Given the small impact a fund manager's decisions make when compared with the dramatic effects of a period of extreme volatility, it is hard to see what value a manager adds.

* Dr Michael Lauren is a physicist with a specialist interest in chaos theory and how it applies to financial markets. Email: mkl@ihug.co.nz