Corporate bonds: Finding alpha in an efficient market

Daniel Morris: Hello, and welcome to the BNP Paribas Asset Management Talking Heads podcast. Every week, Talking Heads will bring you in-depth insights and analysis on the topics that really matter to investors. In this episode, we’ll be discussing enhancing returns in corporate bond markets. I’m Daniel Morris, Chief Market Strategist, and I’m joined today by Charles Cresteil, Investment Specialist and Anthony Desrousseaux, Portfolio Manager. Welcome, gentlemen, and thanks for joining me.

Charles Cresteil:

Happy to be here.

Anthony Desrousseaux:

Hello, Daniel.

DM: When we think about investing in corporate bonds, initially we think about what’s going on with central bank policy rates, with spreads, economic outlook and so on. But once you’ve taken that decision to invest in corporate bonds, what you’re really going to be looking for is how to generate the most alpha out of that investment. Where are the relative opportunities? What parts of the market do you want to invest in? And that is what you guys focus on.

If I could start with you, Charles, we think of corporate bond markets as fairly efficient. That makes the challenge of finding alpha all the more difficult. Where, for you, does the alpha come from?

CC: At first sight, corporate bond markets look highly efficient. They are widely researched, institutional investors dominate trading, and a great deal of information about issuers is available publicly. But that doesn’t mean that every bond is always perfectly priced. In fact, the structure of the market creates many small recurring opportunities. There are four main reasons for that.

The first is breadth. A large company may have five, 10 or even more bonds outstanding. Those securities were all issued by the same company but can differ in terms of maturity, seniority, currency liquidity, or even some ESG (environmental, social & governance) characteristics like green or social use of proceeds.

As a result, two bonds from the same issuer can offer different compensation for broadly the same fundamental risk. That gives us a much broader opportunity set than a market in which each company only has one listed security.

The second reason is fragmentation. Corporate bonds do not trade on a single central exchange as large-cap equities do. Liquidity varies significantly from one issue to another, and different investor groups may value the same bond differently. Insurance companies, pension funds, banks, ETFs (exchange-traded funds) and total return managers all operate with their own constraints, and their buying and selling can create temporary dislocations that are not necessarily linked to a change in the issuer[s[ long-term risk.

A third source of opportunity comes from the benchmark itself. A traditional corporate bond index is usually weighted by the amount of debt outstanding. In other words, the companies that borrow the most receive the largest index weights. That makes sense when the objective is to be representative of the overall market, but it doesn’t mean the index is optimal from an investor point of view in terms of expected risk-adjusted returns.

An active process can therefore be more selective about which issuers and which individual bonds provide good reward for their risks. And finally, prices do not always adjust at the same speed. A company’s fundamentals may improve before all its bonds fully reflect the change. A new issue may come with an attractive concession on the primary market. Or a bond may become temporarily cheap because a large investor needs to sell. These opportunities can occur repeatedly over a  large investment universe.

Our objective is not to make calls on the direction of interest rates or credit markets as a whole; it is to identify a large number of small, relative value opportunities, referring one bond to another with comparable characteristics, avoiding securities where the reward looks poor compared to the credit risk and diversifying those decisions across the portfolio.

Each individual position should contribute only modestly, but together they can create a consistent source of excess return. That is the central idea: alpha in corporate bonds doesn’t have to come from big bets. It can come from a disciplined security selection approach applied across a broad market that remains fragmented and imperfectly efficient.

DM: Anthony, even though theoretically markets are supposed to be efficient, we know the reality is somewhat different. It’s nice to know there are these discrepancies in the market, but finding them is the real challenge. What are the signals you use to identify those opportunities?

AD: We use a multi-factor framework because there is no single indicator that can reliably identify attractive bonds in every market environment. A bond may look cheap, but it may be because the issuer is deteriorating. Another bond may have strong fundamentals but offer too little spread to compensate investors. The purpose of combining signals is to obtain a more balanced view of each security.

The first family of signals is valuation. How much compensation are we receiving for the risk we are taking? We compare bonds spread with those of relevant peers while adjusting for characteristics such as sector, rating, maturity and duration.

We may also compare different bonds from the same issuer. The objective is not simply to buy the bond with the wider spread, but to identify securities that appear attractively priced relative to their underlying risk.

The second family is quality. We examine the financial strength and resilience of the issuer using measures linked to profitability, leverage, cash flow, generation, balance sheet stability and the ability to service debt.

Quality is particularly important in credit because the return distribution is asymmetric. The upside on a bond is limited while the downside from a downgrade or default can be significant. A quality signal helps us avoid being paid a little more spread in exchange for a disproportionate deterioration risk.

The third family is momentum. Credit markets often adjust gradually as information is absorbed by different investors. Improving spread trends, supportive price behaviour or positive changes in market perception can indicate that a bond is moving in the right direction.

Conversely, a bond that looks statistically cheap but continues to underperform may be signalling that the market sees a problem not yet fully visible in accounting data. We also use technical and low risk indicators. Technical signals can capture elements such as issuance, liquidity, investors flows and supply demand imbalances.

Low-risk indicators help distinguish securities whose apparent return potential depends on taking excessive credit or market risk. These signals matter because a theoretically attractive valuation may not be investable. Once trading cost, liquidity and portfolio constraints are considered, we then combine the different signals into an overall assessment for each bond.

Importantly, the comparison is made within appropriate peer groups. We are generally not trying to forecast whether the entire credit market will rally or sell off. We are asking whether one bond is more attractive than another with broadly comparable characteristics.

The advantage of a systematic process is consistency and scale. The same questions can be asked across thousands of securities every time the data is updated. That does not eliminate judgement. The model, data, constraints and portfolio construction all require careful design and oversight. But it prevents the process from depending on a small number of subjective convictions.

In practice, the strongest opportunities are usually those where several independent signals agree [a] bond is reasonably valued, issued by a sound company, supported by market momentum and sufficiently liquid to trade. That combination is more robust than relying on any single factor alone.

DM: Anthony, you’ve talked about the different ways you can evaluate the markets to look for opportunities. But when it comes to portfolio management, there’s risk and return. How do you manage keeping tracking error low when you’re taking many active positions?

AD: The key is to separate active security selection from broad market exposure. We want the portfolio to remain close to its benchmark in the dimensions that investors expect, while using a limited risk budget to favour the bonds that our process finds more attractive.

Let us start with duration. If the portfolio has much more or much less interest-rate sensitivity than the benchmark, performance can be dominated by a view on government bond yields that may produce a positive result, but it’s not the type of alpha we are trying to deliver. We therefore keep duration tightly aligned with the benchmark, both at the portfolio level and, where relevant, across maturity segments.

We apply the same discipline to credit exposure. A portfolio could outperform simply by holding systematically lower-rated bonds or by taking more exposure to volatile sectors. Again, that would be a broad beta decision rather than evidence of successful security selection. We therefore control rating spread, duration, sector allocation and other measures of credit sensitivity. The objective is that the portfolio does not need a strong market rally to succeed.

Issuer concentration is another important control. Even when the model identifies a very attractive bond, we do not want the result to depend excessively on one company.

Corporate credit has an asymmetric risk profile. A single adverse event can have a meaningful impact, while the upside from any individual bond is naturally capped. Position limits help ensure that no single issuer or security dominates the portfolio.

Active risk portfolio construction then brings all these elements together. The process looks for the best combination of expected attractiveness and controlled risk. For example, if we overweight one bond in one sector, we may underweight a less attractive bond with similar duration and credit characteristics. The broad exposures largely offset one another, leaving the relative value preference as the main active decision. This is how we try to isolate alpha from beta.

Diversification is crucial. Instead of making a few large calls, we spread the risk budget across many modest positions. Each preference may have only a small impact on tracking error but the portfolio can contain a large number of such decisions. If those signals contain useful information and are sufficiently independent, their combined contribution can become meaningful while the effect of any individual error remains limited.

We also account for liquidity and turnover. A portfolio that looks attractive before cost may be much less compelling after bid/offer spreads and market impact. Trading therefore needs to be selective. We do not want to react to every small change. The expected benefit of a trade should exceed its likely cost. Low tracking error is not achieved by avoiding active positions. It is achieved by being precise about which risk we want to take and which we want to neutralise.

We accept diversified research-based security selection risk. We tightly constrain unintended duration. Market beta sector and concentration risk. The result is a portfolio designed to preserve the benchmark characteristics that matter to investors while seeking a modest but repeatable improvement in return.

DM: Well, Charles, Anthony, thank you very much for joining me.

CC: Thanks for having us.

AD: Thank you, Daniel.

DM: That’s it for this week’s episode of Talking Heads. If you would like more information about our capabilities in investing in corporate bonds, please reach out to your asset management contact or check out Viewpoint, our website, for investment insights at viewpoint.bnpparibas-am.com. The Talking Heads podcast is available on Spotify and on YouTube. For YouTube, visit YouTube.com bnppam playlist and tap or click on Talking Heads.

You’ve been listening to the BNP Paribas Asset Management Talking Heads podcast with me, Daniel Morris, Charles Cresteil, Investment Specialist, and Anthony Desrousseaux, Portfolio Manager.

Please do join me next week. Until then, take care.