These days, many investors are familiar with incorporating factor investing or smart beta into their equity portfolios. However, equities are by no means the only asset class in which factors can improve the risk-return profile of a portfolio. Empirical studies show that the concept of factor investing can also be applied to many other markets, in particular to corporate bonds. “I see a growing number of research papers, seminars, products and indices for fixed income smart beta and more specifically for investment grade and high yield credits,” says Patrick Houweling, a quantitative researcher at Robeco and portfolio manager of the QI Global Multi-Factor Credits fund.
According to Houweling, the reason why ‘quants’ started paying closer attention to credits only recently is “plain and simple”. To conduct empirical analysis, quants need data, and historical datasets on individual corporate bonds were hard to find for a long time. Nowadays, such datasets have become more widely available and, as a result, the number of research articles has risen substantially. “We counted as many papers on factor investing in credit markets that were published over the last 18 months as in the five years before,” Houweling says.
‘We counted as many papers on factor investing in credit markets that were published over the last 18 months as in the five years before’
The graphic below illustrates the rapid rise in the number of research papers and broker reports on factor investing in credit markets (per factor) published over the past ten years.
This trend in the number of articles and reports has occurred alongside the rapidly growing interest on the part of institutional investors in so-called ‘smart beta’ fixed income funds. Various recent surveys confirm this. For example, an annual study by index provider FTSE Russell found that 27% of the participants have either already invested in fixed income smart beta products, or are considering doing so in the next 18 months. “We clearly see rising interest in factor credit strategies,” says Houweling. “Over the past two-and-a-half years, we have held about 250 client meetings and presentations on this topic.”
According to another study by market intelligence consultant Spence Johnson, European respondents anticipated that for fixed income, the annual growth rate in factor investing would be 19%. This is the highest rate of all asset classes. In particular, the consultant expects the popularity of non-equity smart beta products to grow faster than that of equity products. Of the total of EUR 48bn predicted to be invested in ‘other’ smart beta asset classes by European investors in 2019, EUR 42bn is expected to be invested in fixed income strategies.
The asset management industry has also spotted the potential of fixed income smart beta. For example, a recent newspaper article in the Financial Times indicated that 25 smart beta fixed income ETFs are already available, and another 25 are being registered. “The fact that more and more fixed income smart beta products are available in the market is another major trend”, says Houweling.
There are many ways in which credit portfolios can benefit from factor investing strategies. Different real-life case studies help illustrate the very diverse situations in which this kind of investment approach is worth considering. These case studies also improve our understanding of the implications of applying factors to credit markets.
For example, a large sovereign wealth fund, which had almost its entire credit holdings invested in passive portfolios, was looking for an alternative because passive strategies typically underperform after costs. Moreover, index-based portfolios invest more in companies with a high debt load, which is not comforting for an investor. “This client liked the rules-based and transparent investment process of factor investing, as well as its modest turnover and fees compared to traditional active management,” says Jeroen van Zundert, a quantitative researcher at Robeco. “In the past, these had been key arguments in favor of passive over active investing.”
Similarly, an Italian private bank wanted to replace an underperforming active asset manager. This bank was looking for an investment strategy that would provide distinctive, diversifying alpha. However, it also wanted to achieve alpha using an evidence-based investment philosophy. “Factor investing, being an evidence-based approach to active investing, was therefore a natural fit for this bank,” Van Zundert says.
A third example concerns an Australian insurer which, after implementing factor investing in its equity portfolio, was looking for a factor-based approach for other asset classes too. “This client was attracted to credit factor investing, as the evidence for credit factor investing is mounting, and so is the number of investment products available,” Van Zundert says. “The investor felt the time was right to implement factors in its credit portfolios.”
According to Houweling, these different case studies show that there are three main reasons why clients may choose a factor-based strategy for credit markets. First, it acts as a style diversifier for those who already invest in classic credit products. The quantitative, rules-based strategy offers diversification by generating alpha in a different way.
Second, for passive investors, factor investing can be an attractive alternative, as it does not have the structural weaknesses and inefficiencies associated with traditional market cap-weighted benchmarks. On the other hand, like passive index-based investing, factor investing also offers transparency, low turnover and modest fees. Third, it provides an opportunity to implement the factor investing approach in a client's portfolio for multiple asset classes. Applying multi-factor investing to both equities and credits enhances returns, without increasing risk.
Now that the use of factor investing in the credit market is coming of age, asset owners have started to think beyond regular off-the-shelf products. Houweling and Van Zundert underscore that investors increasingly ask for tailor-made solutions. For example, clients may have specific requirements regarding the investment universe, the factor mix or the level of turnover. They may also have particular risk constraints or ESG targets. “In those cases, we can definitely customize the standard solution,” says Van Zundert.
One way to adapt a strategy to a client-specific requirement is by adjusting the investment universe. “For instance, a German pension fund did not want to have exposure to financials, so we developed an ex-financials solution,” says Van Zundert. Another example of possible customization has to do with the way bonds should be dealt with if they have been downgraded from investment grade to high yield status. Most investors allow some high yield bonds in their investment grade portfolio, but some do not want to invest in high yield at all. Constraining the time-to-maturity of the investable bonds, in order to better match a client’s liability structure for example, is another customization option.
In addition to these investment universe adjustments, factor investing can also be combined with other concepts. For example, a UK-based pension fund wanted to combine factor investing with the ‘buy-and-maintain’ approach. Buy-and-maintain means that after a bond is bought, the intention is to hold the bond until maturity, unless it becomes too risky. The low risk and quality characteristics, which also imply low turnover, are useful for determining which bonds to buy and, if necessary, to sell. “This is why we proposed adjusting the factor mix in order to make it more suitable for a buy-and-maintain approach,” says Van Zundert.
‘Since factor investing is rules-based, the incorporation of additional sustainability rules, is relatively straightforward’
Last but not least, improving the sustainability profile of a factor portfolio is another important area of possible customization. In some cases, for example, clients want to reduce the CO2 emissions, water use, energy consumption and waste generation of the portfolio compared to the benchmark. Since a factor strategy is already rules-based, the incorporation of additional rules, such as environmental footprint reductions or other ESG dimensions, is relatively straightforward.
But despite the growing number of products and research papers available, investors should always look at new strategies with a critical eye, says Houweling. “Asset managers have to do their own thorough research,” he warns. “Especially when it comes to the practical implementation of factor strategies in the credit market and dealing adequately with the illiquidity of corporate bonds. After all, we think that some fixed income smart beta strategies are ‘smarter’ than others.”
Robeco has long been a pioneer in quantitative research on factor-based investing in credit markets. We started researching factor-based credit selection models at the end of the 1990s and have been managing internal factor credit portfolios since 2005. We have also been offering standalone factor credit strategies since 2012.
In 2014, Houweling and Van Zundert wrote the research paper Factor investing in the corporate bond market1. In this academic study, the first of its kind, they argued that factor strategies can also be implemented in credit markets. They concluded that a multi-factor credit portfolio would have generated attractive Sharpe and information ratios over the 1994-2014 research period. This paper was later published in the well-known Financial Analysts Journal, the academic journal of the CFA Institute.
Following the research on the application of factors to corporate bonds, Robeco turned theory into practice and launched the Robeco QI Global Multi-Factor Credits fund in 2015, which provides global exposure to investment grade corporate bonds which score well in terms of the factors value, momentum, low-risk, quality, and size. This fund uses a disciplined, research driven investment process to efficiently capture factor premiums, avoiding unrewarded risks and unnecessary turnover.
At the heart of the investment process lies Robeco’s quantitative multi-factor model, which ranks bonds in terms of their factor characteristics. The strategy is designed to deliver long-term outperformance throughout the credit cycle. This makes it possible to avoid the losers while also picking the winners.
1‘Factor Investing in the Corporate Bond Market’, Patrick Houweling and Jeroen van Zundert, Financial Analysts Journal, 2017, Vol. 73, No. 2.