Brent Sheather: Some clues in shares versus bonds lottery

Forecasting share prices one year ahead is just about impossible and thus a waste of time. File photo / Getty Images

In US dollar terms, the world stock market has returned 8.6 per cent a year since 1900. Since 2000, the return is -0.4 per cent a year. In the same period, global bonds have returned 5.4 per cent a year.

Anyone fancy a guess as to the returns of shares and bonds in the next 10 years? It is a lottery, but there are some clues. For instance, as with any purchase, whether you get a bargain or not depends on the price you buy at.

Buy cheap and high returns will follow, buy expensive and you could still be losing money 20 years later.

Financial innovation and the recent popularity of shares has rendered some of the old rules redundant - lower-risk assets can outperform high-risk ones for extended periods and some bonds can be riskier than shares.

The statement of investment policies for a local investment fund reads: "Higher-risk assets will over time normally generate higher returns."

This is a bold claim. Today just about everybody knows that shares have outperformed bonds over the long term.

Because this is a widely held belief, however, there is a risk that, as happened in 1999, share prices may be bid up to such an extent that they underperform bonds, which is exactly what they have done since April 1987.

That is 24 years of low risk beating high risk, which probably meets most people's definition of the long term. Obscuring things further is the investment industry's tendency to forecast returns just a year ahead, usually based on notoriously unreliable bottom-up forecasts for individual company profits.

Added to that is the need for the industry to make a sale and the unfortunate fact of life that the more risky the product, the higher the fees. In this scenario, equities are always good value.

So what have we got so far? Forecasting share prices one year ahead is just about impossible and thus a waste of time, the people who do it most often can't be trusted to give an unbiased answer and future returns depend on the price you bought at.

If you buy expensive, the old "higher risk equals higher return" paradigm is out the window, as has been the case in the 24 years since April 1987.

But retail investors should not despair. There are independent and skilled investment experts out there who have an appropriate long-term perspective, a sense of history and an unbiased and intelligent methodology.

We refer, of course, to the latest editions of the Global Investment Returns Yearbook (GIRY) and the Barclays Capital Equity Gilt Study (EGS).

We have reported the findings of each year's GIRY and EGS for the past 10 years or so and the 2011 versions of each offer some interesting insights, particularly in regard to expected returns from international bonds and equities in the next 10 years.

This column concentrates on those forecasts and, in a fortnight, we will see what they mean for the prospective return from a balanced portfolio, comparing them with the growth assumptions implicit in various local websites' forecasts of how quickly one's savings will grow.

The Credit Suisse Global Investment Returns Yearbook, produced by London Business School professors Elroy Dimson, Mike Staunton and Paul Marsh, uses a database of 21 countries that records performance of stocks, bonds, bills, inflation and currency for most countries back to 1900.

It shows that in 1900 one New Zealand pound was worth US$4.5514; at that time, $1 would have been worth US$2.2757.

At the end of last year, $1 bought only US$0.7812, an average depreciation of -1 per cent a year since 1900. Not a good look for the kiwi.

A highlight of the 2011 edition is the forecast of long-term future returns for global bonds and equities. With global interest rates way below their long-term average, the professors see interest rates rising, which is bad news for holders of 10-year US bonds.

For global shares, they note that, historically, returns have exceeded those of short-term deposits with the US government by an average of 4.5 per cent a year.

However, they don't see this as being a good guide for the next century as the past 111 years of returns have benefited from a number of one-off occurrences, including high dividend yields at the start of the period and shares getting more expensive.

Without going too deeply into the mechanics, they forecast that the future excess return of international shares over short-term government bonds will be between 3 per cent and 3.5 per cent a year. That suggests a future return from shares of 6 per cent to 6.5 per cent a year.

The GIRY does not forecast bond returns, but notes that interest rates are low at present and there have been periods of up to 50 years when bond returns have trailed inflation. The GIRY mentions one such bear market in US government bonds that lasted from December 1940 until September 1991.

The EGS has been published continuously since 1956 and it provides data, analysis and commentary on long-term bond (known as gilts in the UK) and share market returns.

The news is not good, particularly for US bond investors, as the Barclays experts see 10-year US government bond yields rising from the current 3.4 per cent to 7 per cent by 2020.

Rising interest rates mean falling prices for long-term bonds so, taking this into account, they forecast pre-tax, pre-inflation returns of just 3 per cent a year for US 10-year bonds in the next 10 years.

The EGS sees US shares beating bond returns by 3 per cent a year to average 6 per cent a year over the next 10 years.

Understanding how they get there is not critical. However, the numbers are important: just a 3 per cent a year return for bonds and 6 per cent for international shares.

The GIRY and EGS come up with similar numbers for international shares - about 6 per cent a year. Now, of course, global shares could do better than that but, to quote a sarcastic comment from Clifford Asness when he noted that shares were overpriced in 2000 in a paper in the Financial Analysts Journal, "equity returns could exceed this level but share prices would have to go down a lot first".

Next time, we will use these numbers and the typical annual fee structures of financial advisers and the average KiwiSaver fund to estimate what prospective returns we can expect from a diversified portfolio in the next 10 years.

We will also take a critical look at some local savings calculators to see what sort of variation in forecasts occur and whether their assumptions are in line with the latest estimates from the likes of the London Business School professors and Barclays Capital.

Brent Sheather is an Auckland-based authorised financial adviser and his adviser/disclosure statement is available on request and free of charge.