Continue active refreshing of this index's data?

Continue active refreshing of this index's data?

Blog Posts — May 14, 2020

Oil in a multi-asset portfolio: If you’re looking to reduce risk, look elsewhere…

by Christoph Schon, CFA, CIPM

Investing in oil as part of a multi-asset strategy can be risky for two reasons: first, the oil price is very volatile, and, second, because it is usually strongly correlated with the stock market. For example, though the market-value weight of the oil holding in Qontigo’s global multi-asset class model portfolio is only 2%, it has, on average, accounted for almost 5% of the total risk over the past three years. The recent market turmoil has led to large variations in the latter contribution, which dropped from 8% in January to 2% in mid-April, and then bounced back to over 5% more recently. In this blog post, we examine the underlying factors driving these changes, as well as how the current behavior compares to historic periods.

The chart below shows the composition of Qontigo’s model portfolio as of May 8, 2020, both in terms of market-value weights and percentage contributions to portfolio risk[1]. The overall volatility of the portfolio was dominated by equities, which accounted for around 75% of total variation, compared with a monetary weight of 50%. US stocks had a higher contribution relative to their market value, as the returns of non-USD shares were dampened by opposing exchange-rate movements. US Treasuries neither added to nor subtracted from overall risk, due to their recent low volatility and near-zero correlation with stock markets. The risk of non-USD (sovereign) bonds and corporate debt, meanwhile, was mostly driven by FX rates and credit spreads, respectively.

Asset-class weights versus risk contributions in Qontigo’s global multi-asset class portfolio

Source: Axioma Risk

At the moment, oil accounts for around 5% of total portfolio variation, which is in line with its long-term average. However, a closer look at the underlying drivers reveals that this was almost entirely due to recent high volatility, rather than oil’s usually strong correlation with stock markets, which had been the main driver in the past. The chart below shows a breakdown of the percentage-risk contribution from oil to total portfolio risk—signified by the solid dark-blue line—since the start of the year. Though the oil price had already been volatile earlier in the year, the correlation with share prices accounted for at least 85% of the total contribution, while opposing movements for FX rates and fixed-income assets had a risk-reducing effect. As the oil price began to decouple from all other asset classes in April, however, those contributions started to fall, and the recent sharp rebound was entirely driven by increased standalone volatility.

Oil percentage risk contributions from interactions with other asset classes

Source: Axioma Risk

Low correlation of an asset with the rest of a portfolio is normally a desirable property, when seeking to reduce overall risk. And to be fair, there have been periods in the recent past when holding oil actively offset risks from the other assets in the portfolio, as the chart below demonstrates. But such periods typically result when the oil price is driven by geopolitical and/or supply considerations, such as Turkey’s threat to cut off Kurdish crude exports after the independence referendum in Iraq in September 2017, or the tensions between the US and Iran in June 2018.

Oil percentage contribution to total portfolio risk versus market-value weight

Source: Axioma Risk

As the chart shows, instances when holding oil reduced instead of increased risk are short-lived and few and far between. In fact, more than four times out of five, investing in “black gold” added more volatility than warranted by its market-value weight—especially when put alongside equities. Therefore, for those seeking an alternative asset to reduce the overall risk of portfolios, especially in a crisis like the current one, oil is not the one!


[1] The risk numbers shown were calibrated using daily returns over a 3-month period.