Crossover bonds: tilting to income and quality

The year 2022 was a historic episode for fixed income amid the largest repricing in fixed-income markets in 70 years. Nominal rates rose by over 3% over a short period, both in the US and the eurozone as seen in figure 1.

Figure 1. 10-year government bond yield moves and drawdowns by asset class

Source: LOIM, Bloomberg. As at end-December 2023. Yields are subject to change and can vary over time. For illustrative purposes only.

The rise in interest rates occurred largely through an increase in real rates, indicating significant tightening from central banks and inflation expectations rising to a lower extent. 

Figure 2. Change in 10-year yields 2020-2023: US and German real yields and breakevens

Source: LOIM, Bloomberg. As at end-December 2023. Yields are subject to change and can vary over time. For illustrative purposes.

The unprecedented rise in yields has made all fixed-income asset valuations attractive, with treasury yields at 4.5% in the US and 2.5% in the eurozone, and IG at 5.5% and 3.5% for US and eurozone corporates, respectively. The crossover segment looks attractive from this standpoint at almost 0.5-0.7% yield in excess of IG bonds. This leads to two questions: 

1. Should investors move lower further down the credit rating spectrum and deeper into high yield?
2. Should investors adopt a conservative approach and remain within investment grade? 

We address both questions in the sections that follow.

Figure 3. Yields by segment and currency as at end-December 2023

Source: LOIM, Bloomberg. As at end-December 2023. Yields are subject to change and can vary over time. For illustrative purposes only.
 

Tightening financial conditions and rising real yields

Financial conditions have tightened significantly in the US and eurozone. The tightening has mirrored an unprecedented rise in real yields which, after abating in the early part of the year, continued in the second half of 2023. The rise of real yields generally reflects more hawkish central banks and ultimately should lead to economic weakening.

Figure 4. Financial conditions and real rates

Source: LOIM, Bloomberg. As at end-December 2023. For illustrative purposes only.

Credit fundamentals began a solid trend on the back of the very low-yield environment in 2020-21. Interest coverage specifically reached highs (figure 5A). This led to a strong upgrade wave starting in 2020, with credit upgrades outnumbering downgrades (figure 5B).

Figure 5. Credit fundamentals

Source: LOIM, Moody’s, Bloomberg. As at end-December 2023. For illustrative purposes only.

However, issuance fell sharply due to the extreme high-rate environment of 2022-23, especially in high-yield, with internal cash flows being used as a funding source wherever possible. Figure 6 shows that the issuance rate in USD and EUR reached its historical lows before reversing marginally in EUR. In general, consistent with historical patterns, we expect high-yield issuance at 20-25% each year to refinance maturing debt. However, we find that high-yield issuers have issued at a rate of less than 10% for almost two years.

Figure 6. 12-month corporate issuance rate: IG vs HY

Source: LOIM, Moody’s, Bloomberg. As at end-December 2023.

This anomaly stems partly from an internal cash flow issue and partly from debt extension, or the terming out of debt maturity. Most high-yield bonds tend to be callable at prices at or above 100. As a rough rule, when bond prices are below 100, there is a limited incentive for issuers to exercise the call on their debt at a higher price. The extension of debt accelerated in 2022 with a significant increase in bonds that are now trading to final maturity. 

Debt extension dynamics can be understood by comparing maturity date with average life. Average life uses the expected maturity based on the probability weighted maturity. This usually biases towards the maturity date which has the lowest yield. In other words, issuers prefer to extend the debt to the maturity at which it is the most economical. The implications are that bonds trading well above par tend to trade to their next call date, while bonds well below par usually trade to their final maturity. 

The current universe of high-yield bonds is therefore trading to its final maturity, thereby making full use of the debt extension available. Historically, this dynamic has created a strong buffer to reduce the refinancing risk of high-yield bonds. However, the bad news for the current market is that this buffer was fully exhausted in 2022 and 2023, as shown by the gap between maturity date and average life in figure 7. The gap is close to zero, signifying no extension buffer remains. Moreover, the gap has remained at historic lows for much longer, indicating a longer period of financing stress.​​​​​​​

Figure 7. Difference between maturity date and average life by ratings category

Source: LOIM, Bloomberg Barclays indices. As at end-December 2023. For illustrative purposes only.

The financing risk is clearly shown by looking at the unavoidable refinancing needs of issuers over the next three years split by ratings. Unavoidable refinancing need is measured by the proportion of bonds with final maturity within the next three years. We find that the unavoidable refinancing need of the segment rated B and below has reached 20-year highs at 40%, implying a significant deterioration of credit fundamentals (figure 8). In the coming years, lower-rated companies can no longer rely on internal cash flows or debt extension. This increased refinancing need arises at much higher rate levels currently, leading to a gradual but significant decline in credit fundamentals.

Figure 8. Unavoidable refinancing need for the next three years by rating category and region

Source: LOIM, Bloomberg Barclays indices. As at end-December 2023. For illustrative purposes.

While higher rated companies can weather the decline in fundamentals, B rated assets, especially in eurozone corporates’ interest coverage¹, are significantly worse than in the past, as shown in figure 9. 

Figure 9. Interest coverage by rating - past, present and future

Source: LOIM, Bloomberg Barclays indices. As at end-December 2023. For illustrative purposes only.

The worsening of fundamentals along with a general rally of credit spreads – particularly in the lower-rated part of the high-yield spectrum – has made us less comfortable with the idiosyncratic parts of the credit market. Therefore, a preference for higher quality, BB rated assets remains our secure spot in the coming years, while still offering access to attractive levels of carry.
 

source:

1 We measure interest coverage as the ratio of earnings before interest and tax (EBIT) to interest paid. This is a conservative measure and assumes that depreciation and amortisation are not avoidable given a need for maintenance cap-ex. Another common usage is the ratio of EBITDA to interest, which generally provides higher numbers (twice as high for lower-rated companies).

Investment-grade issuers are generally expected to weather the rise in funding costs better than high-yield issuers. However, investment-grade bond returns tend to be dominated by rates (duration) while high-yield returns derive more from credit risk. This was illustrated in the  previous insight of this series in the section entitled ‘Better balance of rates and credit’.

Credit return and risk are generally well understood and depend on the level of credit spreads with an adjustment for defaults. In addition, the mean reverting nature of credit spreads enables reliable estimates of long-term volatility and return. However, estimating rate return is far more difficult. One approach followed in the literature is treasury term premium models such as the Adrian, Crump and Moench ​​​​​​​ (ACM) model of the New York Federal Reserve². Term premium largely measures the expected excess return of a treasury bond over cash and is estimated using the historical relationship of excess returns with the shape of the curve (pricing factors) and changes in the shape of the curve (factor innovations). ​​​​​​​

Figure 10.  ACM term premium model

Source: Adrian, Tobias, Richard K. Crump, and Emanuel Moench. "Pricing the term structure with linear regressions." Journal of Financial Economics 110, no. 1 (2013): 110-138. As at December 2023. For illustrative purposes only.

The ACM term premium model shown in figure 10 indicates that the term premium, while having corrected in September and October, is still at relatively low levels below zero for the 5-year point and slightly above for the 10-year point. This indicates that the long-term expected return over cash remains flat or negative for treasuries. Note that interest rates are a key driver of risk and return for IG, indicating a large source of risk with limited long-term performance benefits.

Figure 11 shows the performance of investment-grade and high-yield debt as a function of the 10-year term premium. We calculate the risk-adjusted performance of different fixed income segments since 1983 based on the level of the term premium³. Figure 11 shows that the risk-adjusted performance of IG corporates is significantly worse than HY in periods of low term premia, with treasuries performance showing that rates exposure is a drag in such an environment. The reverse is true in high term premia periods, with extra rates exposure being a benefit to IG over HY. Crossover, however, benefits from its equal balance of rate and credit risk to post the best risk-adjusted performance regardless of risk premia regime.

Figure 11. Sharpe ratio: US IG vs HY conditional on levels of 10y term premia, 1983 – 2023

Source: LOIM, Bloomberg Barclays indices. As at end-December 2023.

The low Sharpe ratio of investment-grade credit in periods of low term premia indicates the need for a more balanced mix of rate and credit risk. While credit fundamentals are likely to worsen, especially for lower-rated issuers, we believe that staying in the intermediate rating segment of BBB and BB can provide a better blend of credit and rate risk going forward.

source:

2 See Adrian, Tobias, Richard K. Crump, and Emanuel Moench. "Pricing the term structure with linear regressions." Journal of Financial Economics 110, no. 1 (2013): 110-138.
3 We calculate performance based on the term premia at the start of the month.

Duration exposure, however, does have an advantage beyond its isolated risk-return benefits, namely diversification. The diversification of rates and credit is clear from figure 12 where lower quality credit substantially outperforms in times of rate sell offs. Even in a real rate sell off, while lower-quality credit also suffers, performance is less negative than in high-quality corporates, making drawdowns less severe as evidenced by the move higher in real rates over the last two years.

Figure 12. Performance of corporate bonds in a rate sell off 1997-2023

Source: LOIM, Bloomberg Barclays indices. As at end-December 2023.

However, a forward-looking perspective suggests a shift in correlations. The scatter plots in figure 13 show that while high current real rates coincide with a positive rate-credit correlation, a fall in real rates is indicative of a negative correlation. Being at cyclically elevated real rate levels would suggest the direction of travel for real rates could be lower over the coming months and years. This indicates to us that credit-rate correlations can return to negative territory through the next stage of the cycle and reignite diversification.

Figure 13. Rate-credit correlations vs level and change in real rates

Source: LOIM, Bloomberg Barclays indices. As at end-Oct 2023.

The arguments presented above suggest investors should be tilting towards both income and quality. In today’s world of high cash rates, income is much easier to come by. However, a negative term premium reduces the attractiveness of duration as a performance driver. 

This fuels our preference for shorter duration and higher spread names within the investment-grade bucket. In general, we prefer greater exposure to credit risk than duration risk within the quality segment. In global BBB to BB, the contribution of credit to risk and return is greater, while remaining similar to IG. While an increased exposure to credit risk is desirable, moving lower down and deep into high yield ratings exposes investors to deteriorating fundamentals and marked idiosyncratic risk. Therefore, we recommend remaining within the BB-rating category for high yield.

There is a robust long-term strategic case for crossover providing the best risk-return trade-off across the credit spectrum, in our view, to generate a credit sweet spot. Crossover tends to substantially outperform investment-grade returns but with comparable risk, both from a mark-to-market and default risk perspective. Mark-to-market risk is further mitigated by the superior credit-rate diversification within this universe, resulting in the highest risk-adjusted returns (Sharpe ratios) across rating categories. Valuation shocks from fallen angels and improving debt fundamentals primarily drive the outperformance of crossover. We believe that improved debt fundamentals in the BBB to BB segment arise from debt holders exercising greater control over a company’s financial policy, especially around the IG-HY threshold. Finally, in our view, the crossover universe could serve as a replacement for investment grade strategies due to its similar risk characteristics, sector distributions and ratings.

Adrian, Tobias, Richard K. Crump, and Emanuel Moench. "Pricing the term structure with linear regressions." Journal of Financial Economics 110, no. 1 (2013): 110-138.

Ambrose, Brent W., Cai, Nianyun, & Helwege, Jean. (2008). Forced selling of fallen angels. Journal of Fixed Income, 18(1), 72-85,4

Baxter, N. (1967). Leverage, Risk of ruin and the cost of capital*. Journal of Finance, 22(3), 395-403

Becker, B., & Ivashina, V. (2015). Reaching for Yield in the Bond Market. Journal of Finance, 70(5), 1863-1902

Dichev, I. (1998). Is the Risk of Bankruptcy a Systematic Risk? Journal of Finance, 53(3), 1131-1147

Dor, A., & Xu, Z. (2011). Fallen Angels: Characteristics, Performance, and Implications for Investors. The Journal of Fixed Income, 20(4), 33-58,4

Ellul, Jotikasthira, & Lundblad. (2011). Regulatory pressure and fire sales in the corporate bond market. Journal of Financial Economics, 101(3), 596-620

Fan, H., & Sundaresan, S. (2000). Debt Valuation, Renegotiation, and Optimal Dividend Policy. The Review of Financial Studies, 13(4), 1057-1099

Ilmanen, A. (2003). Stock-bond correlations. The Journal of Fixed Income, 13(2), 55-66

Jensen, & Meckling. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure.
Journal of Financial Economics, 3(4), 305-360

Kraus, A., & Litzenberger, R. (1973). A State-Preference Model of Optimal Financial Leverage. The Journal of Finance, 28(4), 911-922

Murray, Scott and Nikolova, Stanislava, The Bond Pricing Implications of Rating-Based Capital Requirements (November 19, 2018). 14th Annual Mid-Atlantic Research Conference in Finance (MARC). Available at SSRN: https://ssrn.com/abstract=2993558 or https://dx.doi.org/10.2139/ssrn.2993558

Solomon, E. (1963). Leverage And The Cost Of Capital. Journal of Finance, 18(2), 273-279

Szilagyi, J., Hilscher, J., & Campbell, J. (2008). In Search of Distress Risk. Journal of Finance, 63(6), 2899-2940