Disappearing diversification: can you have a truly uncorrelated portfolio?
The rationale behind diversification is that it gives investors the chance to achieve positive returns in one market when another market is generating negative returns.
A downturn in the property market, for example, may not affect commodities in the same way, so the investor would have some protection against the poorly performing asset class.
However, during the 2008/9 financial crisis, the dogma of asset non-correlation did not apply. When the market collapsed, asset classes moved in step together – downwards – as illustrated by the graph below.
At the time seemingly well-diversified portfolios performed poorly.
Since then, markets have recovered yet many investors still diversify their portfolios on the basis of asset allocation. This is despite the same phenomenon resurfacing in 2010 when the Greek debt crisis saw US, European and international stocks as well as corporate bonds all falling together.
“Diversification has a habit of disappearing when you need it most,” says Sebastian Page of PIMCO.
“Assets are always fairly closely correlated. The problem is, this only tends to reveal itself clearly during extreme market moves. In periods of stability or growth, asset classes may move in slightly different directions or at different speeds, apparently without reference to one another. During downturns, however, it becomes clear that all assets ultimately move the same way.”
Page’s view is supported by Pierre Sarrau, deputy chief investment officer of BlackRock’s Multi-Asset Strategies division. Sarrau notes that assets move in the same direction because they are exposed to similar risks, and the same risks drive returns in each asset class. For example, while equities and corporate bonds can appear uncorrelated, they share common risk factors, such as sensitivity to inflation.
Despite his analysis, Page believes that diversification remains a viable strategy. He believes, however, that portfolios should be structured according to risk factors, rather than asset classes. In essence, this would mean assessing a portfolio’s components according to their exposure to interest rates, currency, commodity or inflation risks, and ensuring the risks associated with each component do not overlap too closely with other holdings within the portfolio.
“Risk factor correlations are lower than asset class correlations, and risk factor correlations tend to be more resistant to wider market movements,” explains Page.
“In an economic downturn your portfolio will go down according to the amount of total risk you hold – so view risk as your fundamental portfolio building block.”
Sarrau believes that while asset class correlation rises significantly during a crisis, this doesn’t necessarily undermine the importance of diversification on the basis of asset allocation.
“Periods of crisis don’t last. In buoyant times, investors have a tendency to chase performance, while flocking to low-risk investment options during a downturn. This can not only lead to missed opportunities during ensuing market recoveries, but blind them to the macro benefits of asset allocation. The benefits of diversification are derived over many years – they reveal themselves in the long-term” Sarrau explains.
“The basic tenet of portfolio construction – that your investments shouldn’t be concentrated in one place – still holds firm, because concentration will only ever produce the best long-term risk/reward trade-off in the most exceptional of circumstances. A well-diversified, global portfolio can always help reduce unnecessary risk.”
He recommends diversifying within individual asset classes – for instance, loading a portfolio with a diverse array of individual securities, based on attributes such as market sector, geographic location, and market capitalisation size, or even all three.
Of course, diverse needn’t mean voluminous. A portfolio comprised of hundreds of stocks would be preclusively expensive – and, Sarrau notes, the more stocks you own, the more likely it is a portfolio will duplicate market results overall.
Ben Seager-Scott, senior research analyst at Tilney Bestinvest, however, strongly supports diversification on the basis of asset allocation, and disputes the notion that diversification failed during the 2008/9 financial crisis.
“It’s true that correlations went up, and many different types of investments lost value simultaneously. It may have felt as though diversification stopped working during the downturn, but it certainly didn’t entirely. Diversification still helped contain portfolio losses,” he explains.
He acknowledges, however, that no matter how diverse a portfolio is, its fate is intimately connected to the overall health of a market.
“Asset allocation can only mitigate losses, not prevent them outright,” he concludes.