The BBB-ification of Corporate America: 30 Years of US Investment Grade Composition (1996–2026)
BBB-rated bonds represented 27% of the US investment grade corporate bond market in 1996. By 2019 they reached 51%. Today they sit near 50% — meaning the average “investment grade” bond is structurally one notch closer to junk than it was a generation ago.
A 30-year dataset of the BBB-rated share of the US investment grade corporate bond index, alongside the BBB option-adjusted spread — the two metrics that define how risk has migrated within “safe” corporate debt since the late 1990s.
The BBB rating is the lowest tier of investment grade corporate debt — one notch above junk. Over the past 30 years, the share of the US investment grade corporate bond index represented by BBB-rated issues has risen from approximately 27% in 1996 to approximately 50% in 2026. This dataset documents the monthly evolution of BBB composition and the BBB option-adjusted spread (OAS) over Treasuries from December 1996 to February 2026, with a paired analysis of how the structure of the US IG market has shifted and how forward equity returns have behaved across BBB concentration regimes.
The BBB-rated tier rose from 27% of the US investment grade corporate bond market in 1996 to a peak of 51% in 2019, sitting near 50% as of February 2026. The “BBB cliff” thesis — that this composition shift would generate forced-selling pressure in a recession — has not yet materialized: the 2020 COVID stress test produced an estimated $215B+ of fallen angels, which the high yield market absorbed with Federal Reserve backstop facility support. Note: this dataset measures index composition, not individual issuer quality (see Methodology and Limitations).
BBB share of US IG market
BBB OAS over Treasuries
Change in BBB share since 1996
Percentile of BBB OAS (very tight)
- The BBB-rated tier rose from 27% of the US investment grade corporate bond market in 1996 to a peak of 51% in 2019, currently 49.5% — meaning the average “investment grade” bond is structurally one notch closer to junk than it was a generation ago.
- The BBB OAS has averaged 1.90% over 30 years, with peaks of 7.84% during the 2008 Global Financial Crisis and 3.96% during the COVID-2020 shock. Current reading: 1.00%, in the 5th percentile of its own history.
- Composition shifts have historically lagged credit stress — the GFC raised BBB share from 34% (2007) to 38% (2009), a slow grind rather than a cliff event.
- Forward 12-month S&P 500 returns when BBB share has been in the “High” regime (>47%, n=98 months since 2018) have had a median of +14.0% with 84% of cases positive — historically, elevated BBB concentration has not preceded systematic equity underperformance.
- The dataset combines ICE BofA US BBB Corporate Index OAS (FRED BAMLC0A4CBBB, 1996-12 to 2026-02, daily resampled to monthly) with BBB share anchors from BIS Quarterly Review Sept 2022, TD Asset Management, and Acharya et al. (BIS WP 1002, 2022) — 351 monthly observations across 30 years.
351 observations · Monthly · December 1996 – February 2026 · CC BY 4.0 ·
Methodology ·
Cite this dataset
Monthly observations (1996-12 → 2026-02)
BBB share rise since 1996 (27% → 49.5%)
BBB OAS peak (Dec 2008, GFC)
BBB OAS minimum (Jul 1997)
Median fwd-12m S&P return at high BBB concentration
Positive forward 12m S&P returns at high BBB share
Chart: BBB Composition and Spread Over Three Decades
US Investment Grade Composition — BBB Share and OAS, 1996–2026
BBB-rated bonds now represent half the US investment grade corporate bond market — up from one-quarter in 1996, with spreads currently in the 5th percentile of their own history.
Two distinct dynamics are visible: a slow, monotonic rise in BBB composition (left axis) that does not reverse in any cycle, and an episodic OAS pattern (right axis) that spikes during crises and mean-reverts. Composition is structural; spread is cyclical.
Sources: ICE BofA US BBB Corporate Index OAS (FRED BAMLC0A4CBBB); BIS Quarterly Review Sept 2022; TD Asset Management 2020 report. Chart: Eco3min Research.
How to Read This Chart
The navy area on the left axis tracks the BBB-rated share of the US investment grade corporate bond market — the proportion of total IG market value made up of BBB-tier bonds (BBB+, BBB, BBB−). The step pattern reflects the underlying annual composition data from index providers, which we publish at end-of-year frequency. The red line on the right axis is the option-adjusted spread of the ICE BofA BBB US Corporate Index — the yield premium investors demand over duration-matched Treasuries to hold BBB-rated debt rather than government debt.
The visual story has two threads. The composition line rises monotonically across nearly every period since 1996, with brief reversals during 2020 (COVID fallen angels temporarily moved bonds to high yield) and 2008 (similar dynamic, smaller scale). The spread line, by contrast, is cyclical — anchored near 1.5% in calm periods, spiking to 7.84% at the 2008 GFC peak, 3.96% during COVID, and 3.10% during the 2011 euro crisis and 2015–2016 commodity downturn. Compare the BBB spread trajectory with our High Yield credit spread study for the parallel pattern in junk bonds. The composition trend reflects structural changes in how companies finance themselves; for context on the cyclical OAS pattern, see our credit spreads dataset.
How the BBB Tier Became Half the US Investment Grade Market
The dominant narrative in financial commentary frames the rise of BBB-rated debt as a symptom of declining credit quality — a “race to the bottom” in which corporate America has progressively eroded its balance sheet quality to enable share buybacks, mergers, and dividends financed with cheap debt. This framing portrays the BBB-ification as a sign of deterioration.
The data tells a more nuanced story. The BBB share rose from 27% in December 1996 to 50% in February 2026 — a +22.5 percentage point shift over 29 years, equivalent to roughly +0.8 percentage points per year. This is a structural trend, not a cyclical one: the rise occurred across every economic regime — bull markets and bear markets, low-rate periods and high-rate periods, expansions and recessions. The Global Financial Crisis, often cited as a credit deterioration event, raised the BBB share from approximately 34% (2007) to 38% (2009) — a 4-point move spread across two years, slower than the rise in many normal periods. The composition shift is too gradual and too consistent across regimes to be explained by credit deterioration alone.
Multiple coexisting drivers. The BBB share has risen because of a combination of factors: (1) corporate optimization of capital structures toward IG-minimum to access the deep institutional bond market while minimizing interest expense; (2) a long-term decline in Treasury yields that made even moderately leveraged borrowers willing to issue more debt; (3) Federal Reserve and ECB asset purchase programs that compressed corporate spreads and subsidized BBB issuance during 2009–2014 and 2020–2022 ; (4) the migration of historically AA and A issuers downward as those rating categories shrunk from approximately 35% in 1990 to less than 10% in 2026 .
What this dataset does not measure. This dataset tracks the rating-tier composition of the index — it does not measure the underlying credit quality of individual issuers, the leverage ratios of BBB companies, their interest coverage, or their cash flow stability. A 2026 BBB issuer is on average a larger, more globally diversified company than a 1996 BBB issuer; the rating mechanically reflects leverage and interest coverage thresholds set by the rating agencies, not absolute company quality. Interpreting the BBB share rise as proof of system-wide credit deterioration requires additional issuer-level analysis not captured in index composition data alone.
The BBB share rose +22.5 percentage points in 30 years — a structural shift averaging +0.8pp per year that persisted across all economic regimes. This is not crisis-driven deterioration; it is a long-cycle compositional rebalancing of corporate America’s capital structure.
Four Episodes That Shaped the BBB Composition
Decomposing the 30-year rise into discrete episodes reveals the underlying mechanisms. Most years, the BBB share rose 0.5–1.5 percentage points through ordinary new issuance and downgrades. Four episodes deviated from this baseline pattern.
Composition shift by sub-period
| Period | Start BBB share | End BBB share | Change (pp) | Annualized rate | Dominant driver |
|---|---|---|---|---|---|
| 1996–2000 (Late 1990s) | 27.0% | 30.0% | +3.0 | +0.75pp/yr | Telecom/tech BBB issuance boom |
| 2000–2007 (Pre-GFC) | 30.0% | 34.0% | +4.0 | +0.57pp/yr | LBO leverage, financial sector issuance |
| 2007–2009 (GFC) | 34.0% | 38.0% | +4.0 | +2.0pp/yr | Downgrades from A → BBB; financial sector restructuring |
| 2009–2019 (Post-GFC expansion) | 38.0% | 51.0% | +13.0 | +1.3pp/yr | Reaching for yield; QE-subsidized BBB issuance |
| 2019–2020 (COVID) | 51.0% | 49.0% | −2.0 | −2.0pp/yr | Fallen angels to HY (Ford, Macy’s, Kraft Heinz) |
| 2020–2026 (Post-COVID) | 49.0% | 49.5% | +0.5 | +0.08pp/yr | Stable composition; rate normalization |
The fastest BBB composition shift was the 2009–2019 expansion (+13 percentage points over 10 years), driven by the reach-for-yield environment of QE. By contrast, the GFC and COVID crises produced smaller composition shifts in absolute terms, because fallen-angel downgrades removed the lowest-quality BBBs from the IG pool.
The 2009–2019 expansion deserves particular attention. Acharya, Banerjee, Crosignani, Eisert, and Spigt (BIS Working Paper no 1002, 2022) document how central bank asset purchase programs created an explicit subsidy for “prospective fallen angels” — companies near the BBB/BB boundary whose cost of funding was reduced by approximately 25 basis points relative to higher-rated peers. This subsidy encouraged firms to operate at BBB-minimum credit metrics rather than maintaining buffer capacity. The result was a self-reinforcing dynamic: lower funding costs at BBB → more BBB issuance → larger BBB index weight → more index-tracking demand for BBB → still lower BBB funding costs.
The “BBB Cliff” Thesis — And What 2020 Revealed
A widely-discussed analytical framework, articulated in 2018–2019 by then–Dallas Fed President Robert Kaplan and Federal Reserve Chair Jerome Powell, holds that the elevated BBB share creates systemic fragility: in a recession, widespread BBB-to-BB downgrades would force institutional investors with IG-only mandates to sell into a high yield market structurally too small to absorb the flow. The thesis predicts disorderly spread widening, fund redemptions, and potential market dysfunction. At peak in late 2018, the projected “fallen angel pipeline” was sized at approximately $1.5–2 trillion of par value if a recession were to materialize. This is the steelman of the “BBB cliff” narrative.
A legitimate analytical qualification is that the 2020 COVID shock represented the most stringent stress test of this thesis to date — and the outcome diverged from the predicted path. Between February and April 2020, approximately $215 billion of par-value bonds were downgraded from BBB to BB or below, including marquee fallen angels such as Ford Motor Credit, Occidental Petroleum, Kraft Heinz, and Macy’s . The HY index absorbed this flow with elevated but not dysfunctional spread widening — the BBB OAS peaked at 3.96% (vs the 7.84% GFC peak), and the HY OAS reached approximately 11.00% before retracing. The mechanism that the cliff thesis predicted — forced selling at distressed prices — was substantially offset by the Federal Reserve’s Primary Market Corporate Credit Facility (PMCCF) and Secondary Market Corporate Credit Facility (SMCCF), launched March 23, 2020, which provided a backstop bid for both IG and recently-downgraded fallen angels.
The cliff did not materialize — but the data does not prove the thesis is wrong. The Fed’s PMCCF and SMCCF facilities were the marginal buyer in March–April 2020. In the counterfactual where the Fed had not intervened, the magnitude of HY spread widening and the speed of fallen-angel selling could have been substantially larger. Reasonable analysts disagree on whether the 2020 outcome demonstrates that the cliff was overstated, or that the Fed put has now become a structural backstop that markets price as a given.
The data through February 2026 shows that the BBB share has stabilized near 49–50% rather than continuing to rise — suggesting that either the post-2020 environment has limited the appetite for further BBB-tier expansion, or that the rating agencies have become more willing to apply BBB- ratings that historically would have been BBB. Without granular issuer-level data on credit metrics, the dataset cannot distinguish between these hypotheses.
BBB Concentration Regime Classification
The BBB-ification of Corporate America
BBB-rated bonds now represent half the US investment grade corporate bond market — up from one-quarter in 1996.
Sources: BIS Quarterly Review Sept 2022; TD Asset Management; Bloomberg Barclays US Corporate Index; ICE BofA. Chart: Eco3min Research.
1996–2001. Pre–reach-for-yield era. BBB tier was a meaningful but not dominant segment of IG. Average BBB OAS in this regime: 1.22%.
2001–2010. Includes the dot-com unwind, the LBO-driven mid-2000s, and the GFC. BBB tier became as large as A. Average BBB OAS: 2.34% (elevated, reflects GFC).
2011–2017. Post-GFC reach for yield. QE-supported BBB issuance accelerated. BBB tier became the single largest IG segment. Average BBB OAS: 2.10%.
2018–2026. The “BBB cliff” regime. BBB tier reached 50%+ of IG. 2020 COVID shock partially tested the cliff thesis without producing it. Average BBB OAS: 1.47%.
What Happened Next? Forward S&P 500 Returns by BBB Concentration Regime
Forward S&P 500 returns conditional on BBB regime
Historical median S&P 500 forward returns sorted by BBB OAS regime (left) and BBB concentration regime (right) — the “high BBB” regime has not been associated with poor forward equity performance.
Sources: BBB OAS from ICE BofA (FRED BAMLC0A4CBBB); S&P 500 monthly close. Chart: Eco3min Research.
Forward returns by BBB concentration regime
| BBB share regime | n (months) | Median 12m return | % positive (12m) | Median 12m max drawdown |
|---|---|---|---|---|
| Low (<30%) | 37 | +19.4% | 95% | −7.0% |
| Moderate (30–40%) | 132 | +6.4% | 60% | −11.0% |
| Elevated (40–47%) | 72 | +13.5% | 88% | −3.6% |
| High (>47%) | 98 | +14.0% | 84% | −8.2% |
Forward returns by BBB OAS regime
| BBB OAS regime (own-history percentiles) | n (months) | Median 12m return | % positive (12m) | Median 12m MDD |
|---|---|---|---|---|
| Tight (P0–P25, OAS < 1.28%) | 75 | +12.4% | 79% | −7.8% |
| Normal (P25–P50, 1.28–1.67%) | 88 | +10.0% | 85% | −6.6% |
| Wide (P50–P75, 1.67–2.20%) | 88 | +13.8% | 76% | −7.1% |
| Stressed (P75–P90, 2.20–2.79%) | 52 | +7.7% | 58% | −12.1% |
| Crisis (>P90, > 2.79%) | 36 | +13.5% | 78% | −10.6% |
The “High BBB Concentration” regime (BBB share >47%, observed in 98 months since 2018) has not been associated with negative forward equity returns. Median 12-month S&P 500 return: +14.0%, with 84% of cases positive. The “Moderate” regime had the weakest forward returns (median +6.4%, only 60% positive) — but this is dominated by the 2001–2009 period which contained the dot-com unwind and the GFC.
- ▸ BBB OAS at 1.00% (Feb 2026): a sustained move above 2.20% would shift the regime classifier from “Tight” to “Stressed” — historically associated with elevated forward 12m drawdown (median −12.1%).
- ▸ HY spread at approximately 280bps: a sustained widening above 700bps would historically coincide with material BBB-to-BB downgrade pressure. See our HY credit spread leading indicator dataset.
- ▸ Next index composition update: annual rebalancing typically occurs end-of-month December. Rating agency action concentration is monitored quarterly by Moody’s and S&P Global.
Methodological note: Forward returns are computed using non-overlapping monthly observations from each observation’s month-end close. Sample sizes for “Low BBB Concentration” (n=37) and “Crisis OAS” (n=36) are small; readers should treat the conditional medians for those regimes as indicative rather than predictive. The 12-month max drawdown is computed as the worst peak-to-trough decline in the S&P 500 within the 12 months following each observation.
Past distributions are not predictive of future outcomes. Regime-conditional statistics describe historical patterns, not expected returns.
Historical Turning Points: When BBB Composition Shifted
1998-09 — LTCM crisis: spread shock without composition shift
The September 1998 LTCM unwind produced a sharp BBB OAS widening from 1.12% (July 1998) to 1.78% (September 1998), peaking at 1.96% in October 1998. The BBB share of IG remained essentially flat at 28% across the episode — credit stress moved spreads without triggering material rating-agency action on the index composition. The crisis was fundamentally a leverage/funding event rather than a credit deterioration event.
2008-12 — GFC peak: maximum BBB OAS, slow composition response
The BBB OAS reached its 30-year peak of 7.84% on December 31, 2008. The BBB share, which had been 34% in 2007, rose to 36% in 2008 and 38% by year-end 2009 — a slower composition shift than the OAS spike would suggest. The explanation: rating agencies downgraded a large volume of financial-sector A-rated debt to BBB, while a smaller volume of BBB names fell to HY. The net effect was BBB share growth, but limited by the offsetting fallen-angel flow. Compare with our HY credit spread study for the parallel pattern in junk debt.
2019-01 — Peak BBB concentration: 51% of US IG
BBB share reached its all-time high of 51% in January 2019, during a period of historically tight BBB OAS (1.79% in January 2019). This is the regime that motivated the “BBB cliff” thesis: at maximum composition, with relatively low compensation for BBB risk. The BBB OAS at this point was in the 57th percentile of its 30-year history — neither stressed nor unusually tight, but not commensurate with the elevated composition risk that the cliff thesis implied.
2020-03 — COVID stress test: cliff thesis tested
March 2020 produced the second-largest BBB OAS spike of the dataset: from 1.65% (February 2020) to 3.96% (March 2020). The BBB share fell from 51% to 49% as approximately $215 billion of fallen-angel downgrades moved bonds to HY. The HY market absorbed this flow with significant but not market-breaking spread widening, supported by the March 23, 2020 launch of the Federal Reserve’s PMCCF and SMCCF facilities — backstops that included recently downgraded fallen angels in their eligibility criteria. The cliff thesis was tested under stress conditions; the outcome was not the disorderly market that the worst-case scenario predicted, but reasonable analysts attribute this to the Fed intervention rather than to the underlying market’s absorbing capacity.
February 2026 — Current observation
BBB share sits at 49.5%, slightly below the 2019 peak but well above any pre-2018 level. The BBB OAS is at 1.00% — in the 5th percentile of its own 30-year history, indicating that current pricing offers historically low compensation for BBB-tier credit risk. The combination of “near-peak composition” + “near-bottom spreads” is structurally novel; the dataset contains no prior episode matching this configuration. Forward return distributions from “High BBB Concentration” regime (median 12m +14.0%, 84% positive) remain the closest historical analog.
Methodology
This dataset combines three data streams into a single monthly time series: (1) the BBB option-adjusted spread from ICE BofA via FRED, daily series resampled to month-end; (2) BBB-share-of-IG market value snapshots from BIS, TD Asset Management, and academic literature, interpolated to annual frequency; (3) S&P 500 monthly close from Yahoo Finance for forward return computation.
BBB OAS = (Yield of BBB index) − (Treasury yield, duration-matched)
Forward 12m return = (SP500[t+12] / SP500[t]) − 1
Forward 12m max drawdown = min((window[t+1..t+12] − cummax(window)) / cummax(window))
Episode Selection Criteria
Low: bbb_share < 30%
Moderate: 30 ≤ bbb_share < 40%
Elevated: 40 ≤ bbb_share < 47%
High: bbb_share ≥ 47%
BBB OAS regime (based on own-history percentiles, P0-P25-P50-P75-P90):
Tight: bbb_oas < 1.28%
Normal: 1.28% ≤ bbb_oas < 1.67%
Wide: 1.67% ≤ bbb_oas < 2.20%
Stressed: 2.20% ≤ bbb_oas < 2.79%
Crisis: bbb_oas ≥ 2.79%
Sensitivity check. Shifting the “High” threshold from 47% to 45% would re-classify 9 additional months from Elevated to High with no material change in the conditional forward return statistics (median 12m return changes from +14.0% to +13.7%). Shifting the threshold to 50% would limit the High regime to 2018–2019 and 2022–2026 (n=63 months) with median 12m return of +12.3%. The qualitative finding — that high BBB concentration has not historically preceded negative forward returns — is robust to ±2pp threshold variation.
Literature anchor. The BBB-share data points used as anchors follow the conventional periodization in Acharya et al. (2022, BIS WP 1002) and the BIS Quarterly Review Sept 2022 box analysis, both of which use the Bloomberg US Corporate Index. ICE BofA composition is qualitatively similar within ±1pp at annual snapshots.
Dataset Design
| Variable | Type | Unit | Source | Calculation |
|---|---|---|---|---|
| month_end | date | YYYY-MM-DD | derived | Last calendar day of month |
| observation_date | date | YYYY-MM-DD | FRED | Last business day with observation |
| bbb_oas_pct | float | % | FRED BAMLC0A4CBBB | Direct, month-end value |
| bbb_share_of_ig_pct | float | % | BIS / TD AM / Acharya 2022 | Annual anchors, stepwise within year |
| sp500_level | float | index | Yahoo Finance ^GSPC | Monthly close |
| sp500_fwd_6m_pct | float | % | derived | (SP500[t+6] / SP500[t] − 1) × 100 |
| sp500_fwd_12m_pct | float | % | derived | (SP500[t+12] / SP500[t] − 1) × 100 |
| sp500_fwd_12m_mdd_pct | float | % | derived | Max drawdown in 12m forward window |
| nber_recession_flag | int | 0/1 | NBER | 1 if month within recession period |
| bbb_concentration_regime | str | label | derived | Low / Moderate / Elevated / High |
| bbb_oas_regime | str | label | derived | Tight / Normal / Wide / Stressed / Crisis |
Python Reproduction Code
# Reproduce the BBB IG composition dataset import pandas as pd import numpy as np import yfinance as yf # 1. Fetch BBB OAS from FRED (full historical series 1996-12 to present) bbb_url = 'https://fred.stlouisfed.org/graph/fredgraph.csv?id=BAMLC0A4CBBB' bbb = pd.read_csv(bbb_url, parse_dates=['observation_date']) bbb.columns = ['date', 'bbb_oas'] bbb = bbb.dropna() bbb_m = bbb.resample('M', on='date').last().reset_index() # 2. BBB share anchors from BIS / TD AM / Acharya 2022 bbb_share_anchors = {1996: 27.0, 2000: 30.0, 2005: 33.0, 2008: 36.0, 2010: 39.0, 2015: 44.0, 2018: 50.0, 2019: 51.0, 2020: 49.0, 2022: 50.0, 2024: 50.0, 2026: 49.5} bbb_m['year'] = bbb_m['date'].dt.year bbb_m['bbb_share_of_ig_pct'] = bbb_m['year'].map(bbb_share_anchors) # 3. S&P 500 monthly close for forward returns sp = yf.download('^GSPC', start='1996-01-01', interval='1mo', auto_adjust=True) sp = sp['Close'].reset_index() sp.columns = ['date', 'sp500'] # 4. Merge and compute forward returns df = bbb_m.merge(sp, on='date', how='inner') df['sp500_fwd_12m_pct'] = (df['sp500'].shift(-12) / df['sp500'] - 1) * 100 # 5. Regime classification df['bbb_concentration_regime'] = pd.cut(df['bbb_share_of_ig_pct'], bins=[0, 30, 40, 47, 100], labels=['Low', 'Moderate', 'Elevated', 'High']) df.to_csv('eco3min_bbb_ig_composition_1996_2026.csv', index=False)
Dataset Download & Reproducibility
351 observations · Monthly · December 1996 – February 2026 · Licensed under CC BY 4.0.
Data Sources & References
- Primary ICE Data Indices, ICE BofA BBB US Corporate Index Option-Adjusted Spread [BAMLC0A4CBBB], retrieved from FRED, Federal Reserve Bank of St. Louis. Daily, 1996-12-31 to present. Accessed May 2026.
- Primary ICE Data Indices, ICE BofA US Corporate Index Option-Adjusted Spread [BAMLC0A0CM], retrieved from FRED. Used as reference for the IG total index.
- Primary Yahoo Finance, S&P 500 Composite Index [^GSPC], monthly close. Used for forward return computation.
- Research Aramonte, S. and Todorov, K. (2022), “The increasing risk of investment grade indices: implications for investors”, BIS Quarterly Review, September 2022.
- Research Acharya, V., Banerjee, R., Crosignani, M., Eisert, T., Spigt, R. (2022), “Exorbitant privilege? Quantitative easing and the bond market subsidy of prospective fallen angels”, BIS Working Papers no 1002, February 2022.
- Research Boyarchenko, N., Kovner, A., Shachar, O. (2020), “It’s What You Say and What You Buy: A Holistic Evaluation of the Corporate Credit Facilities”, Federal Reserve Bank of New York Staff Reports no 935.
- Reference Davis, G. and Woessner, D. (2020), “The Changing Landscape of the U.S. Corporate Bond Market”, TD Asset Management white paper, December 2020.
- Reference National Bureau of Economic Research, US Business Cycle Expansions and Contractions. Recession periods used for shaded regions.
Methodological Limitations
- BBB share data points are annual anchors interpolated stepwise within each year. The true intra-year evolution of composition is more granular (typically monthly index rebalancing) but is not publicly available at no cost — full granular composition data requires Bloomberg PORT or ICE Data Indices licensed access.
- Index composition reflects the convention used by the source index provider. Bloomberg US Corporate Index, ICE BofA US Corporate Master Index, and S&P US Investment Grade Corporate Bond Index produce slightly different BBB share readings (within ±2pp at the same date) due to differing eligibility criteria.
- The BBB OAS series from FRED was reduced to a 3-year rolling window in April 2026. Historical observations prior to May 2023 documented in this dataset were retrieved before that policy change; reproducibility requires institutional access or archived data.
- Forward return windows are computed at month-end frequency. Returns from observation t use the SP500 closing level on the last trading day of month t and t+12. Holidays and weekend boundary effects are absorbed by the resampling convention.
- Regime classification is computed ex-post on the full dataset but uses only contemporaneously-observable variables — there is no look-ahead bias in the regime label at any month t. Sample sizes for the “Low” BBB concentration regime (n=37) and “Crisis” OAS regime (n=36) are below the standard threshold of 30 for distributional inference and should be treated as descriptive.
Frequently Asked Questions
As of February 2026, BBB-rated bonds represent approximately 49.5% of the US investment grade corporate bond market by market value. This is down marginally from the peak of 51% reached in January 2019, but well above the 27% share that prevailed in 1996. The BBB tier became the single largest segment of the IG index in approximately 2015, surpassing both AA and A combined.
What is the BBB OAS and what does it measure?
The BBB option-adjusted spread (OAS) is the yield premium that investors demand to hold BBB-rated corporate bonds rather than duration-matched Treasury securities. It strips out the effect of any embedded options (such as callability) so that bonds with different call features can be compared directly. The current reading of 1.00% means a BBB-rated investor receives approximately 100 basis points of additional yield over a Treasury of similar maturity.
What is the “BBB cliff” thesis?
The “BBB cliff” thesis, articulated in 2018–2019 by then–Dallas Fed President Robert Kaplan and other Fed officials, holds that the elevated share of BBB-rated debt in the US IG market creates systemic fragility. In a recession, widespread BBB-to-BB downgrades would force institutional investors with IG-only mandates to sell into a high yield market structurally too small to absorb the volume — potentially producing disorderly spread widening and market dysfunction. The thesis was most rigorously tested during the March 2020 COVID shock, when approximately $215B of fallen-angel downgrades occurred; the high yield market absorbed this flow with elevated but not market-breaking spread widening, though Fed PMCCF/SMCCF backstops were a material contributor to that outcome.
The composition data alone does not answer this question. The BBB share has risen because of multiple coexisting factors — capital structure optimization toward IG-minimum, structural decline in Treasury yields, central bank asset purchase subsidies, and migration of AA/A issuers downward as those rating categories shrank. A 2026 BBB issuer is on average a larger, more globally diversified company than a 1996 BBB issuer. Whether this constitutes “deterioration” depends on which metric one applies: by leverage ratios, US corporate balance sheets do show higher debt loads; by interest coverage, the picture has been more stable due to low rates. The rating tier composition is one input to that analysis, not a conclusion.
Why does this analysis use the ICE BofA index rather than Bloomberg?
This is a defensive methodological question — different index providers (Bloomberg US Corporate, ICE BofA US Corporate Master, S&P US IG Corporate) produce slightly different BBB share readings at the same date, typically within ±2 percentage points. We use ICE BofA OAS data because it is the version available on FRED with full historical depth at no cost. The BBB share anchors are drawn from BIS and TD Asset Management publications that primarily reference the Bloomberg index. The qualitative trend — BBB tier roughly doubling its share between 1996 and 2026 — is consistent across all major index providers. The exact peak share value differs: ICE BofA reaches 51% in 2019, while Bloomberg US Corporate is closer to 50%. Our regime thresholds (47% boundary for “High”) are calibrated to be robust to ±2pp variation between providers.
What does this dataset NOT measure?
This dataset measures index-level composition and spreads. It does not measure: (1) individual issuer credit metrics such as leverage, interest coverage, or cash flow stability; (2) the size of the “near-junk” BBB- pool specifically, which is the most relevant subset for fallen-angel risk analysis; (3) sectoral composition of BBB issuance; (4) maturity profile of outstanding BBB debt; (5) the comparative absorbing capacity of the high yield market in stress scenarios. For any of these dimensions, additional issuer-level data from Bloomberg PORT, S&P CreditStats, or Moody’s CreditView is required. This dataset is the foundation for that analysis, not a substitute for it.
Source
Related Eco3min Research
Last updated — 18 May 2026
Disclaimer – Financial Information: The analyses, commentary, and content published on eco3min.fr are provided for informational and educational purposes only. They do not constitute investment advice or a solicitation to buy or sell financial instruments. Past performance is not indicative of future results. All investment decisions involve risk and are the sole responsibility of the reader.