HY vs. IG: Quality Spread Differentials and Comparative Returns
We update the running incremental spread differentials and relative returns for HY vs. IG across the tightening cycle.
Quality spread differentials are widely recognized as tight at this point, and we offer some historical context on “how tight” with a look back across the credit cycles.
The OAS differential of “HY OAS minus IG OAS” offers an objective measure of how much you get paid for moving from the investment grade basket into the speculative grade realm, but the BBB and BB tiers offer a way to mix your risk profile with less volatility as we covered in our prior note (2-5-24).
The risk-reward symmetry for HY is on shakier ground than the IG market in terms of absolute spreads while the IG market offers the mitigating risk factor of a material play on the UST curve at a critical time when duration offers more protection against fundamentals heading south.
In this commentary, we follow up on yesterday’s review of quality spread differentials and comparative returns for the BBB vs. BB tier (see BBB vs. BB: Revisiting the Speculative Grade Divide Differentials 2-5-24). This installment looks at the broader IG vs. HY differentials across time from the TMT bubble, across the credit crisis, through COVID, and across the tightening cycle and ongoing inflation war into 2024.
We use the same chart framework as we used with the BB vs. BBB differentials. We look at IG and HY OAS differentials longer term in the chart above, but then zero in on a shorter timeline from Jan 2022 onwards to get better granularity for the spread behavior during the first inflation fighting cycle in over four decades. In that regard, this current mix of macro factors is unique in the life of the HY market. The HY index did not even start until the summer of 1986 and daily HY OAS did not really become “a thing” until 1996.
Inflation spikes and massive tightening are rare in corporate bond market history….
The last time around that we had an all-out inflation war took place in the early 1980s with the Volcker regime at the Fed after he took over in August 1979. The HY market was in its toddler stage in the early 1980s. I was working as a CPA on the audit of one of Milken’s bigger clients at the time (Rapid-American and its various high-risk subsidiaries in retail and consumer products). The hyper-tightening and the double dip recession was over by the end of 1982.
The HY market itself really did not accelerate as an asset class until 1984 when many would argue the inflation war was already over. Some say it was over by the end of 1982, but we tend to say it was over in 1986 just to be conservative since that was when 1% handle CPI metrics ruled. The HY market and the publicly traded bond markets generally soared from 1986 on as disintermediation of the banks and insurance company private placement departments accelerated.
We look back at the fed funds vs. PCE relationship in a separate update we will post after this commentary, but our main point is that this current market has a unique set of factors in recent history with the inflation factor being the main difference from the HY credit cycles of the 1980s, 1990s, and into the new millennium with the credit crisis and COVID.
The multicycle history tells a story of boom and bust…
With the HY OAS minus IG OAS at +245 bps at the latest date posted (2-2-24), we can frame the IG to HY risk premium differential as very tight relative to almost all the historical medians shown above with their +300-handles. The one period with a high +200 bps handle median was COVID to current at +287 bps. That said, we have seen numerous periods more compressed than what we see now.
The chart shows +187 bps lows for Oct 1997 and +147 bps in the June 2007 credit bubble. Both 1997 and 2007 had a much riskier overall profile of issuers yet were still much tighter than today. We can also make the case that the IG markets were higher quality then. Those backdrops were thus worse for relative value.
The lows of June 2014 (+228 bps) and Oct 2018 (+205 bps) were also lower than where we are today while the Dec 2021 low (+203 bps) was even lower. In other words, we are not a danger zone, but we are well below the median mix. As always, it gets back to how you view the risk profile of the market and the key drivers of risk pricing.
The inflation factor that had many of us on edge during 2022 has made remarkable progress. That is the good news. The recent reality check from Powell on the magnitude of cuts creates some mixed emotions, but his views do fall on the side of economic fundamentals being respectable. That is easy enough for asset managers and risk professionals to support in the macro data and earnings season results and guidance.
An asterisk for the timeline above is that the overall mix of the BB tier in the HY index is much higher (48% of current market value is BB tier) than in the TMT years and LBO binges of the credit bubble by 2007. Mix is always an issue in historical index comparisons, but that is a topic for another day.
We lack the OAS data above to capture the end of the 1980s meltdown of HY in the “securities industry vs. banks” dynamic and “loans vs. bonds” in the Glass-Steagall years. That meant less secondary market support and capital supporting the HY market relative to the 1990s. On more than a few occasions during the late 1980s and early 1990s, credit stress at the issuer level saw some cases of “bank on broker” and “loan on bond” violence where the bonds were the losers. The convergence of banks and securities firms changed the range of potential outcomes.
Drexel disappearing into bankruptcy in Feb 1990 did not help the state of affairs until the “Class of 1992” rebuilt the HY market. We cover those times in other commentaries and histories we have published (see UST Moves: 1988-1989 Credit Cycle Swoon 10-20-23, Greenspan’s First Cyclical Ride: 1987-1992 10-24-22).
The post TMT spread differential spikes….
The chart does show the excess of the 1990s and the price paid for the underwriting quality erosion that led to the longest (not the highest) default cycle in history. That default cycle ran from late 1999 into a peak OAS differential in 2001 and 2002 as flagged in the chart.
The +700 and +800 handle HY-IG OAS differentials spanned a timeline from early 2001 to late 2002 in what ended up as a double-dip default wave in a very muted and mild economic downturn. The easing cycle extended well into 2004 even as the expansion was in gear, and the market saw 1% fed funds into 2004 (see Greenspan’s Last Hurrah: His Wild Finish Before the Crisis 10-30-22).
We then see the now old story of the structured credit excess and subprime/housing bubble that witnessed the IG OAS vs. HY OAS differential reach another dimension at +1531 bps in Dec 2008 or almost 2x the 2002 peak.
During the world of ZIRP we had the 2011 spread wave post +638 bps (Eurozone crisis and a near UST default on Tea Party radicalism). The Feb 2016 spike rose back to a satanic +666 differential (2-11-16) that came with the oil crash and ensuing E&P stress. The energy credit crisis brought an infectious fear of HY fund redemption risk driven by the stunning spike in upstream defaults and a plunge in debt recoveries (more than a few single digit unsecured recoveries in the headlines). COVID in March 2023 hit +686 bps ahead of Fed and fiscal relief.
The last stop on the spread wave journey was the risk aversion rising with the tightening cycle and the rapid upward UST curve migration of 2022 that peaked in the chart above at +432 bps. We look at that period below.
The timeline from Jan 2022 through today shows the swing higher from +208 bps for “HY OAS minus IG OAS” in Jan 2022 on the way to the early July 2022 peak of +432 bps as CPI crossed the 9% line. June 2022 was the first of 4 consecutive 75 bps tightening actions by the Fed.
We saw another risk pricing wave in late Sept 2022 as too many in the market were screaming “we are already in a recession!” (see Unemployment, Recessions, and the Potter Stewart Rule 10-7-22). The consumer sector hung tough, and payrolls kept rising.
The 2023 quality spread compression stayed on track after HY OAS closed in 2022 at over +480 bps and the HY rally got underway. March 2023 disrupted the markets briefly with the regional bank crisis as the Fed quickly came to the rescue (see SVB Reprieve: Hail Powell the Merciful 3-12-23). That fueled some fears of credit contraction and bigger problems, but that went away quickly after another earnings season and active jawboning by the Fed.
The spread wave and equity market sell-off in the early fall of 2023 was tied to the UST selloff and steepening before the ensuing UST rally brought a +228 HY-IG differential low in late Dec ahead of the current +245 bps. That recent number includes some adverse mix trends on rebalancing since the Dec 2023 lows.
The above chart breaks out the gross OAS moves we use to derive the differentials across the time period with the peak HY OAS evident in July 2022 just short of +600 bps after starting at +305 bps in Jan 2022. The solid spreads in early Jan 2022 came after a blisteringly bullish credit and equity market in 2021 with plenty of ZIRP support.
The 2023 period was overall a period that shows a compression pattern with disruptions in March with the regional banks and in Sept-Oct 2023 with the brief UST curve turmoil. As we saw this past week with the 30Y mortgage getting back out to the 7% area for many market quotes, we cannot rule out more UST bouts in 2024.
After the New York Community Bancorp (who bought Signature Bank back in the regional bank crisis) headlines this past week, the asset quality question marks around commercial real estate are not going away with more fuses likely smoldering out there.
With the IG OAS level at +102 bps on the latest print in the chart, the IG market is hanging around near the June 2007 levels but only modestly above the 1H04 to 1H07 average and 1997-1998 average. These current IG spreads are in a rarified zone even if we have seen much lower IG OAS in past credit cycle peaks (+53 bps low in Oct 1997). We look at some of those histories for IG and HY spreads in our 2023 yields recap (see Footnotes & Flashbacks: State of Yields 1-1-24).
The bottom line is that we are in a very tight market for spreads with very strong demand for yields the market has not seen since the pre-ZIRP years except for in HY when spread waves spiked yields for high risk credit during steep selloffs since the crisis. The UST curve (as opposed to all-in credit yields) is strictly at pre-crisis levels at this point.
The above chart updates monthly total returns for IG vs. HY across the last 12 months. Overall, the last 12 months was a period of favorable spread compression with a few events such as the regional banks and erratic UST moves that slammed the UST markets in the early fall.
The Sept-Oct price action saw some serious duration pain. That setback then quickly pivoted into a monster rally in Nov-Dec that rewarded all risk (duration and spreads). Both fixed income and stocks came up roses with an equities rally from large caps to small caps that reversed the YTD pattern of narrow tech-centric leadership tied to the Mag 7. The year ended with an expansive breadth of equity subsector recoveries (see Footnotes & Flashbacks: Asset Returns 1-1-24).
The above chart highlights the excess return profile across the year with a very strong move in HY in a few concentrated pockets in the summer and during the wild finish in Nov-Dec. We see the market anxiety of the regional bank crisis in March 2023 and the heightened UST curve jitters that peaked in Oct 2023.
The flows into credit assets from IG to HY come and go, but the IG market is likely to sustain very strong interest from asset allocators as a favorable way to gain yield. In time, corporate bond investors will find more current coupons that are in line with yield in what is a steeply discounted market overall after years of post-crisis ZIRP era coupon generation followed by COVID era refinancing.
US HY will hold much of the same allure with new issue and refi/extension staring at yields above many assumed plan asset returns built into pensions. The massive growth of defined contribution AUM will also be a likely destination for cyclical peak credit yields not seen since before the crisis and only rivaled in down markets when spreads gap wider.
We believe much better opportunities will come later in HY with IG the better value now. Inflows swing around and time horizons vary but high-quality bonds with 5% in the BBB tier and 6% handles in the BB tier will generate a lot of interest.
The new world of private credit also beckons with higher risk and less liquid asset offerings, but those could be ideal for the parameters of massive retirement asset portfolios from the private to public sector. There will be no shortage of debate on what rapid AUM under growth will mean for disciplined lending practices in the battle of the pitchbooks.
Contributors:
Glenn Reynolds, CFA glenn@macro4micro.com
Kevin Chun, CFA kevin@macro4micro.com
See also:
BBB vs. BB: Revisiting the Speculative Grade Divide Differentials 2-5-24
Footnotes & Flashbacks: State of Yields 2-4-24
Footnotes & Flashbacks: Asset Returns 2-4-24
Histories: Asset Returns from 2016 to 2023 1-21-24
Credit Performance: Excess Return Differentials in 2023 1-1-24
Return Quilts: Resilience from the Bottom Up 12-30-23
HY Refi Risks: The Maturity Challenge 12-20-23
Coupon Climb: Phasing into Reality 12-12-23
HY vs. IG Excess and Total Returns Across Cycles: The UST Kicker 12-11-23
HY Multicycle Spreads, Excess Returns, Total Returns 12-5-23