If the low-volatility ‘anomaly’ did not exist, would pension funds still have risk-management uses for these strategies? Rachel Fixsen investigates

The market crash of 2008 almost halved the value of many equities portfolios. Sharply stung, many investors looked around for other ways to get exposure to shares – but more safely – and minimum-variance or low-volatility strategies began to gain increasing acceptance.

The outperformance of low-volatility stocks over the longer term is often referred to as the low-volatility ‘anomaly’ – because classic finance theory suggests that risk and return have a linear relationship. The obvious question from investors is: ‘How do I know this anomaly will persist, and will it not simply be arbitraged away?’

There are a number of competing theories to explain the ‘anomaly’ and defend it against the arbitrage assumption, many of them behavioural. Kelly Young, managing director of the London office of Boston-based Acadian Asset Management, which runs nearly $4bn in managed-volatility equities, argues that investors have a psychological bias towards higher-risk growth stocks because it feels appealing to be looking for the next Google.

“It has some glamour and excitement attached to it,” she says. The sheer volume of money invested in cap-weighted strategies, coupled with the fact this is unlikely to change significantly in the foreseeable future, means the payoff to lower volatility stocks will probably continue, she adds.

“The empirical evidence seems to fly in the face of modern portfolio theory,” concedes Mark Roemer, senior vice president and portfolio manager at Allianz Global Investors Capital (AGIC). “Before we launched our strategies, we came at this with a healthy dose of scepticism and were able to validate the anomaly through our own research.”

The team at AGIC also feel that investors gravitate towards stocks with lottery-like payoffs – a low probability of an extremely high return. This preference leads to such stocks, which have a history of positively skewed returns, becoming overpriced, and then on average earning a negative excess return, the team found. Investors end up with this inferior return not because they want higher volatility explicitly, but because they crave stocks with higher skewness. Sure enough, while stocks with both high skewness and high volatility dramatically underperform, stocks with low skewness behave as expected under standard portfolio theory – higher volatility is compensated with higher return.

“So that finally laid a mathematical framework around this behavioural bias and why this is an effect that will likely remain in the market – why it’s not arbitraged away,” says Roemer.

But even in world in which low-volatility stocks delivered the same risk-adjusted return, or even the same absolute return, are there reasons that might prompt institutional investors to use these strategies?

Regardless of the reality of excess returns, Nicolas Davidson, senior portfolio manager at AllianceBernstein, makes the point that low-volatility stocks bring valuable diversification benefits to broad-based equities portfolios.

“The out-performance from low-volatility equities is largely uncorrelated with positive returns from other styles of active equity management such as value or growth,” he says. “As a result, adding an allocation to low-volatility equities into a diversified equity portfolio can both smooth the pattern of returns and produce a more efficient use of a client’s risk budget relative to a broad equity benchmark, showing up in a better information ratio.”

Low-volatility stocks also suit a pension fund’s primary aims, according to Paul Bouchey, managing director of research at structured portfolio management firm Parametric Portfolio Associates.

“For all pension systems, the main goal is just to pay the liabilities with a high degree of certainty in the future,” he points out. But, he adds, most funds are not fully-funded and thereby able simply to buy bonds and insulate themselves from the future. “So the investing-for-pension-funds game is, what else can you invest in that might earn a better return than this?”

Typically, pension funds mix in equities to boost their returns. But then an event like 2000 or 2008 comes along in which an inflation in the present value of liabilities coincides with a crash in equity markets.

“The argument in favour of minimum-variance portfolios is they’re less of a mismatch [with liabilities],” he suggests. “A cap-weighted index may have 20% volatility, and can have a significant downside risk, whereas a minimum variance portfolio can be quite a lot lower than that – 10 to 15%.”

Similarly, holding low-volatility equities could also potentially reduce the regulatory capital cost of an insurance company’s equity allocation under Solvency II, or the equivalent regime planned for European pension funds, suggests Davidson. “Our research suggests that adding an allocation to low volatility equities can reduce the volatility of insurance companies’ earnings and balance sheets,” he says.

Others see drawbacks with both cap-weighted and minimum-variance strategies, and offer alternatives designed to address these.

For example, PanAgora Asset Management favours the ‘risk parity’ approach, under which index or portfolio constituents are weighted in order to equalise or balance the marginal risk they contribute to the whole portfolio. The firm claims that this results in more stable returns with smaller drawdowns than either cap-weighted or low-volatility, which result in more concentrated risk exposure.

“We think risk parity is the most efficient way to capture risk premium,” says Erik Gosule, head of client solutions and investment strategy at PanAgora Asset Management. “Almost every other product except risk parity requires some view that exposure to a particular attribute is likely to produce outperformance relative to other approaches. A cap-weighted index has very significant sector and country risk concentrations, so in theory by investing in the index you’re implicitly expressing a view that you think financials – for example – will outperform.”

“Market value-based investing has some disadvantages – it is a momentum strategy, and stocks that are performing well get a higher allocation,” agrees Barbara Bleijenbergh, senior investment strategist at the Dutch pension fund PME. As a result, its passive equities portfolio now includes small allocations to two alternative beta strategies – ‘quality’ and fundamental indexation – and will evaluate these and other options in the near future. “We would do so as an addition to – and diversification of – our market value-based passive investments,” says Bleijenbergh.

On the active side of its equity strategy, PME does prefer more stable stocks, with lower than average volatility, but it does not pick them primarily for that attribute.

“PME invests in companies with good quality,” Bleijenbergh explains. “[Stocks with] relatively low leverage, high return on assets and low earnings variability are less sensitive to economic and credit cycles and have lower volatility, especially in downward markets. But low volatility is an output of the stock selection process and not a result of an optimisation model – we do not prefer a black box optimisation approach.”

While investors like PME are finding a place for low-volatility stocks based in various parts of their portfolios and for various reasons, consultancy Cardano finds potential confusion for pension funds in the market for minimum-variance products.

“Being able to generate an equity-like return with lower volatility is attractive to most investors and this concept is being more actively marketed by a number of providers,” observes Steven Catchpole, senior client manager at Cardano UK. “However, the concept is not new and they are many styles and approaches which are trying to achieve the same thing.”

Generally, he believes the average pension fund already relies too heavily on equities to generate returns to remove deficits, and that trustees would be better off accessing a wider range of asset classes and building all-weather portfolios.