Brent Sheather: Report will turn tables on stock theory

It seems every so often an article, academic paper or book is published that captures the nature and mood of the current investment scene and serves as a guide to investment strategy from that point on.

We have had Triumph of the Optimists from the trio of economists at the London Business School and Dow 36000 by three US journalists, both of which pointed to the huge historic outperformance of shares over the long term.

Investors have tended to embrace these revelations with both hands and, in the case of the articles above, pile into shares. But because historic performance tends to look best at times when prices are high, the results of acting on the lessons learned have frequently been disappointing.

Ironies abound - about 20 years ago another high-profile bit of research showed that small companies outperformed large companies. Investors duly bought small companies and the trend promptly reversed - probably because small companies were expensive at the time the analysis was done which, of course, is why they looked so good.

Last month, Robert Arnott, former editor of the US Financial Analysts Journal and investment manager of some US$31 billion ($53 billion) at Research Affiliates, produced a report on US government bonds which is to be published in the May-June edition of the Journal of Indexes.

The article, "Bonds: why bother", which Arnott has kindly made available to Herald readers, looks like it will be one of the those historic pieces which defines the investing scene at the time.

The article highlights the extraordinary fact that in the past 40 years, from February 1969 to February 2009, super-safe 20-year US government bonds have generated a higher total return than the US stockmarket.

This will be an unwelcome revelation to many as the conventional wisdom is that shares, because they are risky, should beat bonds if you can hang in there for 10 or at most 20 years.

This purported outperformance of shares over bonds is known as the risk premium. It has been generally estimated at 5 per cent a year and up until recently has been thought to be as reliable as a AAA-rated bond.

With the well-documented demise of AAA maybe we should have known the days were numbered for the risk premium as well.

Arnott's article is shocking stuff. The question for the panel, however, is whether, on the back of Arnott's insight for analysis, we should be buying bonds or shares?

Let's look at the article in more detail.

First up, Arnott backgrounds contemporary investment strategy and where bonds fit in. Most institutional investors use bonds in their tool kit for two key reasons: to diversify and to reduce risk.

Additionally retail investors use bonds to produce a higher level of income than they could get in the bank. However, bonds aren't usually the first port of call when one wants to grow one's wealth - stocks are favoured for growth and bonds are for widows and orphans.

But Arnott's analysis shows if you bought 20-year US government bonds in February 1969, reinvested the income and rolled over into 20-year bonds each year, this strategy would have produced a higher return than from owning the S&P 500 index of the largest US companies.

That is a 40-year period when bonds have outperformed stocks. Most people retiring at 60 or so don't have any thing like a 40-year investment horizon.

Nor is this result a statistical aberration - it has happened before and more than once.

According to Arnott's paper, US$1 invested in shares in 1802 would have risen to be worth about US$4 million over 207 years, versus "just" US$27,000 if the US$1 had been invested in US government bonds.

So over that period an investor who had opted for shares would be 150 times better off than someone who had stuck to bonds. That is an impressive multiple but it works out to be a risk premium of just 2.5 per cent a year.

So that is the first revelation and one that Arnott has flagged to Herald readers before, back in 2003. Shares can be expected to beat bonds, but only by 2.5 per cent a year.

This is made up of 1 per cent from real earnings and dividend growth and 1.5 per cent from inflation.

So if shares are only good for 2.5 per cent a year over bonds, and if the average New Zealander with a managed investment portfolio is paying management fees to fund managers of 1-2 per cent a year, other fees including turnover costs of at least 0.5 per cent a year and a monitoring fee to a financial planner of another 1 per cent a year, it seems all they are destined to get from the typical private banking equity-centric portfolio is the extra risk of shares with the additional

risk premium going to their favourite private banker. That is lesson No 2. But there is worse to come.

Two-and-a-half per cent is a thin enough differential even without fees. But its cost in terms of risk gives us, in Arnott's words, "one heck of a ride".

So, ladies and gentlemen, fasten your seatbelts as we go back in time to 1802. The graph above plots the performance of the US stockmarket without dividends and adjusted for inflation so we can see how long it took for real share prices to exceed our original buy-in price.

If you had taken the plunge into equities in 1802, you had to wait 32 years before your share portfolio got back to break even. Furthermore it took 75 years to convincingly exceed 1802's levels. But by 1929 your portfolio was up five-fold.

Pride comes before a fall and we all know what happens next. The crash of 1929-32 was so severe that share prices, expressed in real terms, briefly dipped below 1802 levels. This means that the US stockmarket history exhibits a 130-year span in which real share prices were flat.

The 20th century offers us a similar instance: "From the share price peak in 1905, we saw bull and bear markets aplenty, but the bear market of 1982 [and the accompanying stagflation binge] saw share prices in real terms fall below the levels first reached in 1905. A 77-year span with no price appreciation in US stocks."

That is enough bad news for one Saturday. Arnott concludes on a more optimistic note warning that, while much investment activity relies on false dogma, the tools are there now, or will be soon available, to genuinely improve returns and diversification.

He sees the imminent arrival of equal-weighted, exchange-traded bond index funds as something that will address these needs. But that is a story for another day.

For the record, Arnott picks that stocks will outperform bonds in the next 40 years and adds "a 3 per cent yield on treasuries isn't that hard to beat given that stocks now yield more than that. Furthermore, if our staggering new debt load triggers inflation as we monetise that debt, bonds could cancel their bond yield with falling prices."

* Brent Sheather is an Auckland stockbroker/financial adviser and his adviser/disclosure statement is available on request and free of charge.