Question:

I've been thinking about whether it makes sense to hold on to shares and wait for the upturn, or sell now. 

If you look at the last century, the true long-run cycles in stocks last about 30 to 50 years. If you had kept your money in the Dow Jones index in 1929, you would have had to wait until 1953 to get back to the SAME 1929 level. Quite a while to wait for the "upturn".

There are serious commentators now calling for the Dow Jones index to fall below 5000. Per capita debt levels globally, along with other imbalances, are far worse than they were in 1929.

It is quite conceivable that it will take another 20-plus years until stocks get back to 2007 levels. Investors should see the simple rationale behind this - this is hardly wild and unfounded speculation.

Advising people to "hold on" now seems to be reckless, in my opinion. Stocks have a long way to go down before we reach the bottom.

If you don't need the money or returns and you can wait 20 years or so, then by all means hold on to your stocks.

Answer:

Sorry, but I feel I must contest a few of your notions. The first is to judge what happens to shares by what happens to the Dow Jones index.

That index includes just 30 of America's biggest companies - hardly representative. It also weights the different companies in an illogical way. But most importantly, it doesn't include dividends.

If you were considering buying a rental property, would you forget about rent in your calculations? Ignoring dividends is just as silly.

This is particularly true with New Zealand shares, because our companies tend to pay higher dividends than in most other countries.

Since 1962, if you invested $1000 in New Zealand shares excluding dividends, it would have grown to $15,000, says Russell Investments. But if you included reinvested dividends - which are included in KiwiSaver funds, super schemes and many individuals' savings - it would have grown to $210,000 to the end of February this year, before tax and ignoring imputation credits. That's 14 times as much.

Even in the US, dividends make a huge difference. Looking at the S&P 500 index, which includes 500 US shares and is calculated more logically than the Dow Jones, our top graph shows that since 1962 you would end up with more than four times as much if we include dividends - again before tax.

Going further back, the Seigel index, which also includes dividends, had recovered to its 1928 level by 1943, not your 1953 - as shown in our bottom graph. In fact, in 1928 to 1953, the gross returns on shares averaged 5.3 per cent a year. That's not great, but nor is it zero.

Some other difficulties I have with what you say:

Using 1929 as your example is like looking at a terrible week in which 100 people died on the New Zealand roads and saying that means our annual road death toll is 5200.

The year 1929 was hardly an average 20th century year.

The 1929 sharemarket crash was the start of the worst US share slump in two centuries, which was followed by the Great Depression and then World War II.

Our bottom graph shows the magnitude of the 1929 crash. Unlike the top graph, the bottom one uses a log scale. What does that mean?

In the top graph the ups and downs of recent years look much bigger than in the past. But - while recent volatility has certainly been high - we've been there before. The graph is distorted by compounding.

Let's look at an example. Double a dollar and you get $2, double that to get $4, then $8, $16, $32, $64, $128, $256, $512, $1024 and so on. The rate of growth - doubling - is the same all the way through. But the last numbers are so big they dwarf the earlier ones.

Log scales get rid of that distortion. In our bottom graph, if two downturns look the same they have the same magnitude. And the 1929 crash is clearly the worst in the graph.

We've learnt a lot about economies since 1929 and the Great Depression. While nobody is certain about how to fix things now, we can be confident governments won't make as big mistakes as they made then.

You start your comparison from the peak of the late 1920s sharemarket boom. From there, a big fall was pretty likely. The same could be said for 2007 - a fall was quite likely too. And it happened.

Now that the markets have plunged, it's much less likely that they will fall a long way further. And that's what is relevant now.

Another reader (thanks, John) forwarded me a relevant Bloomberg article. Just 17 months ago, it says, an investment in US shares since 1979 would have grown to more than twice as much as an investment in US government bonds.

But since then, bond returns have leapt and shares have plunged to the point that the investments since 1979 would now be equal.

Does that mean shares have had their day? History suggests the opposite. Over the longish run, shares must return more than bonds. If they didn't, few people would buy shares, given that they are riskier than bonds.

That lack of demand would push share prices down. And the lower the prices, the bigger the chance of strong future gains - making shares attractive to buy again.

You seem confident that "stocks have a long way to go down". But as I said recently, Warren Buffett isn't predicting further big falls. While he doesn't always get it right, he is one of the world's richest people, having built his wealth on share investing. His guess - and all stock predictions are guesses, really - is as good as any.

Nobody's doubting that world economies will take a while to come right. But - precisely because we all know that - it may already have been fully reflected in share prices, and possibly over-reflected. That's certainly happened before.

Another important point: While there's risk in holding on to shares, there's also risk in selling now.

Those who bail out of share investments are making their loss real. They lose the chance to regain what they've lost.

And that leads to another question: What else should people do with their savings?

Property may not be much less risky. And if you borrow lots to invest in property - which most people do - it's quite likely to be riskier than shares.

Going into bonds is risky in another way. There's been a fair bit of talk about the possibility of inflation soaring. If that happens, bond holders can be left way behind, in terms of what their investments will buy.

And that's what matters.

You might turn out to be right - shares might go down much further, and take ages to recover. But - after weighing it all up -

I'm not only keeping most of my retirement savings in share funds but continuing to feed money into those funds every month. And I'm not quite in my 20s or 30s!

I wouldn't do that - or suggest others do it - with money needed within the next 10 years or so. But I'm comfortable putting into share funds the money I expect to spend from about 2020 on.

By the way, I can't say I understand your reference to 30 to 50 year cycles. The graphs don't seem to show any sign of that.

Mary Holm is the author of bestselling books on KiwiSaver and personal finance. She is also a highly praised seminar presenter. Her written advice is of a general nature, and she is not responsible for any loss that any reader may suffer from following that advice.