How can we protect ourselves from the danger of falling stock markets? The standard way is to diversify among assets, to hold cash or bonds as well as shares. But we can also use another form of diversification – time diversification.
Its power rests upon an important tendency – for prices to recover, often strongly, after falling, with the result that falls provide great buying opportunities. For example, if you’d bought the All-Share index in December 1999 (the peak of the tech bubble) you would have made a total return after inflation of just 1.8 per cent a year since then. But if you had bought at the low point of the tech crash in March 2003 you would have made 6.1 per cent per year. Similarly, if you had bought at the pre-crisis peak in October 2007 you would have made 2.4 per cent a year since then, but you’d have made 8.3 per cent if you’d bought at the low point in February 2009.
This means that some of us can diversify across time as well as across space. Just as good returns on one asset might offset losses on another, so good returns in one period can offset losses in another.
This does not happen simply because stock markets overreact to bad news. It can do so even if markets are wholly rational. If investors become more risk averse, prices fall so that expected returns rise to compensate us for the market’s additional perceived riskiness. We’ll then see prices rise after a fall even if investors are rational. Plus, of course, big falls in prices cause policy responses to support the market, such as cuts in interest rates, bank bail-outs or quantitative easing.
Whatever the reason, the fact is the same: market falls lead to rises and so time diversification works.
You might object that this is a counsel of perfection. We only know in hindsight when the market is at its cheapest. And when prices are low we are often too nervous to buy; in fact it is such nervousness that causes prices to be low.
We have two possible solutions here. One is to focus on the dividend yield. My chart shows why. It plots the dividend yield against the total annualised real return on the All-Share index since that date. So for example, the dividend yield in December 2000 was 2.2 per cent and returns since then have been 2.4 per cent a year: my chart stops in 2016 because the yield is less good at predicting shorter-term returns. It’s clear that high yields predict high returns and low yields low returns.
A second solution is to use the 10-month moving average rule. It says that we should buy when prices are above their 10-month (or 200-day) average and sell when they are below it.
Imagine you’d applied this rule during the tech crash. Every £1,000 you’d had in the All-Share index in December 1999 would have been worth less than £540 in real terms in January 2003, and you wouldn’t have recouped your £1,000 until December 2006. If, however, you had put £500 into the market in March 2002, when the 10-month rule told you to buy, it would have grown to £632 by December 2006, a real return of over 26 per cent.
Similarly, if you’d followed the rule during the 2007-09 crisis and put £500 into equities in May 2009 it would have made £232 by March 2013 – the time when total real returns returned to their pre-crisis peak. High returns during the post-crisis recovery thus mitigated the crisis losses.
Time diversification, therefore, works. To apply it reasonably successfully, however, requires three things, not one of which is intelligence.
One is the discipline to apply these rules, to buy when you don’t feel like doing so.
A second one is time. It takes years of good returns to offset the 30-40 per cent losses we can suffer in bear markets – although of course these losses can themselves be mitigated by applying the 10-month rule when shares fall.
Thirdly, we need cash. If you are retired or not saving you can only implement time diversification by having a large cash holding in the first place.
If you are working and saving, however, things are easier. You can implement time diversification with your new savings. Regular direct debit payments into funds in fact do exactly this.
It’s in this sense that stock markets are safer investments for workers (and especially younger ones) than for retired people, especially older ones.
Or are they? History tells us that time diversification works because shares have bounced back after every bear market. But there’s no guarantee history will repeat itself.
The danger here is not so much of Japan-style lost decades: the Nikkei 225 is even now far below 1989’s peak. For one thing, the UK market is nothing like as overvalued as Japan’s was then; in fact, dividend yields suggest it is cheap. And for another, even the Japanese market has enjoyed some huge rallies over the past 30 years, which time diversifiers using the 10-month rule could have profited from.
Instead, the danger is of a near-permanent fall. US academics Will Goetzmann and Philippe Jorion have shown that of the 25 national stock markets that existed in the 1920s, 11 suffered a catastrophic failure because of war, revolution or hyper-inflation. Of course, the chances of that happening here are minuscule. But it is this tiny chance that really matters. Anything else is manageable with time, discipline and cash.
