New tricks

PORTFOLIO POINT: With correlation and price movements exceeding the bounds of rationality, this will be yet another year of risk-on and risk-off trading. Your asset allocation strategy should become a risk allocation strategy.

While the world boozed on in those strange, unstructured days between Christmas and New Year, the dense middle pages of the Wall Street Journal published a short piece on movements in investment theory, specifically from a conference held by the Norwegian Ministry of Finance some two months earlier.

As far as journalism goes, it must have been a slow news day, but I'm glad I had nothing better to do than read it. For despite it being a relatively arcane subject – the portfolio strategy of the Government Pension Fund of Norway – it contained an important message that any Australian investor could learn from.

Norway's $US500 billion sovereign wealth fund – either the largest or the second largest in the world after Abu Dhabi's, depending on how you calculate it – has been a model for other government investment vehicles (including our own Future Fund) since its inception several decades ago.

As a “rainy day account” for when Norway inevitably depletes its North Sea oil reserves, the fund owns an astonishing 1% of world equity on issue. Yet despite its preeminence in the marketplace – and its leadership on ethics and shareholder activism (it famously banned all investment in Rio Tinto three years ago, citing the company's environmental record) – the fund was not immune to the credit crunch of 2008-09, suffering the same negative returns as nearly everyone else.

To avoid such calamities in future, and to be a better fiduciary of the Norwegian people's birthright, Norges Bank Investment Management, the fund's manager, has commissioned academic research into how it could better run its vast riches. And among these studies – the subject of the Journal's article – was the recommendation to employ “factor investing” strategies, whereby a portfolio is balanced by risk factor rather than asset class. In other words, this new approach ignores the traditional breakup of investing activities into stocks, bonds, etc, and looks instead at the inherent risks in any activity such as holding cash for special opportunities or having exposure to certain types of companies.

This idea is not new, having existed on the margins of financial academia for several decades, but it is gaining a new prominence with index fund managers such as BlackRock and Russell, among others, planning to launch exchange traded funds (ETFs) based on the concept (for more traditional ETF offerings however, see my previous story here). Institutional managers like the California Public Employees Retirement Scheme (CalPERS) have already incorporated a factor investing approach into their core operations. And, in a sign that surely heralds the strategy’s entrance into the financial zeitgeist, there’s now a blog.

At its heart, factor investing organises portfolio construction around a quantification of volatility, momentum and Beta, the variable that measures a financial instrument's correlation with a broader index (say, Telstra versus the ASX 200). And since both volatility and correlation have noticeably increased between popular asset classes since the financial crisis (see Correlation's causation), finding ways to diversify your exposure to different volatility, momentum and Beta is easier said than done using traditional portfolio management methods.

Enter factor investing.

Despite the scale of Norges Bank Investment Management’s portfolio, and its expectation that it will double to $US1 trillion by 2020, that doesn't mean its ongoing research isn't useful to Australian retail investors. And although market-weighted asset allocation strategies will remain important for many, if not most investors, factor investing can still play a part in at least some portion of your portfolio's construction.

Of course, some investors will instinctively recoil from such an approach – with the memory of risk-tranching in collateralised debt obligations (CDOs) still fresh. Here, investment banks, wanting to sugar-coat relatively unappetising products such as subprime mortgages, were able to artificially slice and dice risk, with the help of the major credit ratings agencies, so that portions of these products magically became AAA.

Yet at a more basic level it doesn't have to be this way for there are plenty of genuinely safe and stable investment options available without the use of legally dubious financial engineering. And although many of those options (eg, cash, term deposits, government bonds) will already sit in a typical portfolio thanks to traditional asset allocation strategies, by reversing the logic through a factor investment approach, a more precise, and sometimes surprising, risk-adjusted portfolio mix can be achieved, especially when used in sub-asset class management: ie, the selection of shares within your equities portfolio.

As we've discussed many times before in Eureka Report, big doesn't necessarily mean better, but when it comes to share investing many investors can't seem to shake the belief that blue-chip companies equal safety and small-cap stocks equal risk. And that's despite the numerous examples of even the mightiest blue-chips going bust over the past few years, while smaller players have survived, whether due to lower gearing, better management or plain good luck.

Portfolio allocation can thus suffer from a traditional approach, as a seminal study by Eugene Fama and Kenneth French published in the Journal of Finance, showed almost 20 years ago. Fama and French's “three-factor model”, which goes beyond traditional market Beta models, is still controversial in some segments of the investment community, yet it supports what's empirically known by many of our readers already: that small-cap stocks and stocks with a high book-to-market ratio (ie, value stocks), tend to outperform other share classes in the long term.

A variety of software products can assist investors to screen and analyse stocks by Beta, volatility and momentum – among other risk factors such as price-to-book, internal leverage or return on equity – otherwise online services like Bloomberg and Yahoo! Finance present this data for free if investors are prepared to enter their stocks one code at a time.

And although factor investing ETFs are yet to hit Australian shores, US-based factor ETFs already exist, though currency risk, liquidity risk and counter-party risk should be at the forefront of any investor's mind. And most of all: your legal recourse is limited here in Australia should something go wrong.

Finally, as an alternative to this alternative, what other investors are doing is weighting allocation not to mathematical measures of market risk, whether endorsed by the ratings agencies or not, but to more general (and hence subjective) measure of economic risk. You are probably already taking such an approach in an informal gut-feel sense, but it's easy to lose sight of this from a portfolio-allocation perspective as you trade and turn-over your investments. Your portfolio, when dispassionately and objectively observed, may be more concentrated than you think.

In the pie charts above, the chart on the left displays a traditional asset allocation model with activity split into
conventional categories, the chart on the right is a reflection of how a chart might reflect a “factor investing” approach.

Either way, whether one takes a DIY approach to factor investing, or seeks a ready-made solution from a third-party provider, this alternative method of portfolio construction is likely to only increase in prominence, especially if pronounced market volatility and market correlation continue into 2012, which I believe will almost certainly be the case. And while it may not be necessary for all investors to readjust their holdings based on a risk rather than asset class approach, it's worth the investigation.

Maybe a project for the next period you're killing time between public holidays.

Follow Michael Feller on Twitter @MFellerEureka