Factor investing also works in credit markets. Over its first three years of existence, our QI Global Multi-Factor Credits fund delivered strong performance.
After years of success in equity markets, factor investing is slowly imposing itself in fixed income markets as well. Robeco has been at the forefront of this move, carrying out pioneering research on this topic. This research led to the launch in 2015 of our QI Global Multi-Factor Credits fund which recently celebrated its third anniversary.
The Robeco Global Multi-Factor Credits strategy seeks to offer a balanced exposure to the low-risk, quality, value, momentum and size factors in the global investment grade credits universe. It aims to achieve an attractive Sharpe ratio and information ratio, by obtaining outperformance with a beta close to 1 and therefore market-like volatility.
Of course, there were rougher patches, in particular in 2016, when global credit markets experienced sharp reversals mostly related to strong commodity price moves and central bank interventions. But despite sometimes choppy waters, the strategy has been able to deliver significant and relatively constant outperformance over the years.
From July 2015 to June 2018, the fund returned an average of 2.18% per year, gross of fees, or 0.46% more than its reference index, the Bloomberg Barclays Global Aggregate Corporates Index hedged into euros. This excess return was accompanied by a level of risk that was lower than the index, resulting in a Sharpe ratio of 0.87 compared to the index figure of 0.69.1 The information ratio achieved over the period was 0.73. This has led to a rapid rise in assets under management which stood at EUR 465 million at the end of June 2018.
“Our long-term research experience on factor investing in credit markets, which recently won an award from the CFA Institute’s Financial Analysts Journal, and our proven track records really set us apart in a field where most of our competitors are still in the early stages,” says Patrick Houweling, Head of Quant Credits. “This adds to our extensive experience in bond trading and analyses dating back to the 1970s, while various new competitors in the quant fixed income arena can be still considered as newcomers.”
The selection of bonds and issuers is driven by an in-house multi-factor model, augmented with fundamental checks. The strategy constructs a well-diversified portfolio of bonds across sectors and regions. The selected bonds have both strong model scores and no risks or events beyond the scope of the model, as checked our team of more than 20 fundamental analysts. Sustainability assessments are systematically incorporated in the investment process by requiring that the portfolio’s score in terms of environmental, social and governance (ESG) criteria is higher than that of the reference index.
Our quantitative investment process benefits from behavioral biases
For clients that already invest in actively managed credit portfolios, this factor-based investment approach enables style diversification by using a systematic way to generate alpha. Indeed, while traditional funds invest according to fundamental views on selected companies and can implement significant beta views, our approach is factor-based and features a beta-neutral portfolio construction process. And while investment decisions of fundamentally-managed funds may be affected by portfolio managers’ emotions, our quantitative investment process benefits from behavioral biases.
Moreover, fundamentally-managed strategies may provide exposure to one factor but will usually not avoid factor clashes and may be prone to style drifts over time. Our multi-factor approach, on the contrary, enables a balanced and persistent exposure to the five well-rewarded factors we exploit in our factor credit strategies.
Our research shows that these important differences resulted in low, mostly negative correlations between the performance of our Global Multi-Factor Credits fund and fundamentally managed peers, over the live period of the fund. Therefore, adding a multi-factor manager to an existing pool of fundamental managers will make the overall outperformance generation more stable.
A key feature of our investment process is that it embeds a unique liquidity management workflow in the portfolio construction. Since bonds can differ greatly in their liquidity, and an individual bond’s liquidity varies greatly over time, we measure the liquidity of all the bonds in our investment universe in real-time. We measure both ‘persistent liquidity’, owing to regular two-way flows, and ‘transient liquidity’, owing to short-term dealer positioning.
“This enables us to send only those orders which are highly likely to be executed,” says Houweling. “As a result, we ensure maximum factor exposure, while keeping trading costs low. In contrast, a liquidity-naïve approach would incur higher transaction costs, would take more time to execute, and would select bonds with worse factor scores.”
1 Source: Robeco Performance Measurement. Returns are measured in EUR, based on IH share class an gross of fees. In reality costs (such as management fees and other costs) are charged. These have a negative effect on the returns shown. The value of your investments may fluctuate. Results obtained in the past are no guarantee for the future.