Prolepsia forecastYes / no
Will US high-yield bond spreads, as measured by the ICE BofA index, be above 4.0% on 31 December 2027?
As of 15 November 2026, Prolepsia puts the chance of yes at 39%.
39%chance of yes
How Prolepsia reasons
The report Prolepsia wrote with this forecast.
How forecasts are made · Track record
TL;DR
The forecast is 38.8% Yes and 61.2% No. The balance favors the ICE BofA US High Yield Index OAS finishing December 31, 2027 at or below 4.00%, but refinancing pressure and a broader credit repricing remain substantial risks.
Key evidence
- The starting point favors No. The latest verified observation is 3.20% on November 12, 2026, below the description’s November 11 reading of 3.30%. Yes requires more than 80 basis points of widening, with the spread still above the threshold at the endpoint.
- History supports meaningful widening risk, not inevitability. Calculations from the reconstructed historical record show month-end spreads above 4.00% in 218 of 359 observations from December 1996 through October 2026, versus 16 of 70 since January 2021. These are historical frequencies, not the forecast. The contrast shows why neither older high-spread eras nor recent calm should dominate the analysis (historical archive).
- Current activity provides a cushion. October unemployment was 4.1%, and payrolls increased by 165,000 (BLS employment release). The Federal Reserve described strong activity but elevated inflation, while holding its policy-rate range at 3.75–4.00% (October FOMC statement).
- Weak-credit stress has not become broad distress. CCC-and-lower spreads stood at 13.14% on November 12. Yet CCC securities represented only 8% of the index at September 30, and some distress was issuer-specific (MacKay Shields outlook). The Fed also reported weak speculative-grade interest coverage alongside ample corporate-bond liquidity (Financial Stability Report).
Uncertainties
The main uncertainty is whether selective borrower distress spreads to better-rated issuers through refinancing constraints, weaker earnings, or sustained risk aversion. Persistent inflation could limit policy relief even without a recession.
Historical comparisons are imperfect: older observations rely partly on archives, overlapping examples are not independent, and index composition has changed. Timing also matters. A temporary widening above 4.00% does not qualify if it reverses before resolution; exactly 4.00% resolves No. If the specified date has no observation, the applicable preceding published close is used.
Read the full report
TL;DR
The forecast is 38.8% Yes and 61.2% No. Use a year-end spread at or below 4.00% as the central planning case, while retaining protection against a broad credit repricing. The decisive issue is whether borrower stress reaches the wider index and remains there at the measurement date, not whether spreads briefly cross the threshold earlier.
Context
The question concerns the overall ICE BofA US High Yield Index option-adjusted spread, not its yield or a rating-specific segment. Yes requires a published closing observation strictly greater than 4.00% on December 31, 2027, or the applicable preceding observation if that date has none. Exactly 4.00% resolves No. The authoritative series is FRED BAMLH0A0HYM2.
The latest verified starting point is lower than the question description’s last observation. The supplied daily history shows 3.42% on November 6, 3.34% on November 9, 3.30% on November 11, and 3.20% on November 12, 2026. The final reading was published November 13. From that level, Yes requires more than 80 basis points of widening (FRED spread series). That is a meaningful move, but it does not require a systemic crisis.
Evidence
The historical backbone points in two directions. Spreads above the threshold are common across the long record. They are much less common in recent years. Calculations from reconstructed monthly history give the following comparison; these are historical occupancy rates, not probabilities for this question. Older observations come partly from a historical archive, supplemented by FRED-derived history and the current official series.
| Sample | Complete month-ends above 4.00% | Total complete month-ends |
|---|---|---|
| December 1996–October 2026 | 218 | 359 |
| January 2010–October 2026 | 114 | 202 |
| January 2015–October 2026 | 61 | 142 |
| January 2021–October 2026 | 16 | 70 |
The long-run frequency alone is a poor guide because it includes periods that began with spreads already above the threshold. Today’s spread starts below it. Among historical month-end starts between 2.75% and 3.75%, the reconstructed record contains 91 outcomes above 4.00% out of 187 pooled 13- and 14-month comparisons. For starts since 2015, the corresponding count is 44 out of 99. These comparisons establish that a low starting spread does not remove widening risk. But adjacent monthly starts often belong to the same economic episode. They are not independent trials.
Recent year-end history tells a calmer story. The reconstructed record gives December closes of 3.86% in 2020, 3.10% in 2021, 4.81% in 2022, 3.39% in 2023, 2.92% in 2024, and 2.81% in 2025 (historical observations). Only one of those six endpoints exceeded the threshold. That supports the No case, but six year-ends cover too few distinct credit cycles to establish a durable rule.
The distinction between a crossing and an endpoint is concrete. The recent record includes a daily spread of 4.61% on April 7, 2025, but that month ended at 3.94% (spread history). A temporary bout of stress can therefore look decisive in headlines and still fail this question’s test. Widening must survive any subsequent recovery through the resolving observation.
Current economic conditions favor containment. The October employment report showed unemployment of 4.1% and payroll growth of 165,000. Those figures do not indicate an existing broad labor-market contraction (BLS employment release). The October 28 Federal Reserve statement described strong activity, resilient spending, and robust capital investment (FOMC statement). My reading is that the economy currently gives corporate earnings and debt service a cushion.
Inflation makes that cushion less secure. October headline CPI rose 0.4% month over month, and core CPI rose 0.3%, both seasonally adjusted (BLS inflation release). The Fed kept its policy-rate range at 3.75–4.00%, described inflation as elevated, and recorded two members preferring a quarter-point increase (FOMC statement). Higher Treasury yields do not mechanically widen OAS. The risk is indirect: expensive refinancing weakens borrowers, while persistent inflation restricts policy relief.
Credit stress is already visible below the headline index. CCC-and-lower spreads rose from 9.41% in May 2026 to 13.27% in October, before standing at 13.14% on November 12 (CCC spread series). BB spreads moved from 1.64% in May to 2.09% in October and 1.83% on November 12 (BB series). Single-B spreads moved from 2.99% to 3.60% and then 3.39% over the same observations (single-B series). The weakest borrowers are under much greater pressure, while the stronger segments remain comparatively contained.
Composition limits the effect of that distress on the overall index. MacKay Shields reported that CCC securities represented 8% of the index on September 30, 2026, versus 11% five years earlier. It also reported that two Altice issuers accounted for about 12% of the CCC segment, and that excluding them reduced the segment’s spread by roughly 200 basis points (MacKay Shields outlook). This is evidence of both real weakness and concentration. It is not evidence that the broad market has already entered a generalized credit crisis.
Bank lending supplies a possible route from isolated distress to broader pressure. The October lending survey showed only 1.7% net tightening for large and middle-market corporate loans (FRED lending-standards series). But the detailed survey showed 22% net tightening for business credit intermediaries, with 12 of 50 banks tightening and one easing. Private-equity funds also faced tighter standards (October SLOOS). My inference is that weaker access to intermediary finance could eventually narrow refinancing options, even while direct corporate-bank lending looks stable.
Counterevidence matters. BNY reported that recent outflows from broad high-yield and CCC holdings had not persisted (BNY flow analysis). The Fed’s November Financial Stability Report described historically weak interest coverage among speculative-grade public firms, but also ample corporate-bond liquidity and a resilient banking system (Financial Stability Report). Vulnerability is present. A broad funding seizure is not established by these observations.
What's non-obvious
Severe distress among the weakest borrowers need not produce a high broad-index spread. A small, concentrated segment can deteriorate sharply while the larger, stronger segments keep the aggregate below the threshold. Changes in index composition, including the removal of defaulted borrowers, also prevent borrower losses from translating one-for-one into higher published spreads. The key signal is therefore propagation into BB and single-B debt, not the CCC level in isolation.
The opposite mistake is to treat recent calm as proof of safety. Recent observations contain few distinct stress episodes, and many historical comparisons reuse the same episodes through overlapping dates. The information in those counts is smaller than the row totals suggest. The forecast favors No because of the starting level and current conditions, while retaining substantial Yes risk because a change in the credit environment can overwhelm both.
Uncertainties
The largest uncertainty is whether refinancing pressure remains selective or becomes broad and persistent. Sustained widening in better-rated high yield, continued fund outflows, weaker earnings, and tighter intermediary financing would make the adverse case more compelling. The current evidence does not establish that this transmission is under way across the market.
Historical comparability is also limited. FRED restricts the available history of this ICE series to approximately three years; older comparisons rely partly on archived observations (FRED series notes). Reported overlap checks found exact agreement, but they do not independently authenticate every older value. A complete official historical record would strengthen the comparisons. Changes in index quality would still limit how directly older credit cycles apply.
Finally, the timing of stress is unresolved. A sharp widening early in the horizon can reverse before year-end, while a late shock can determine the result. Neither the current spread nor recent calm places a hard bound on the endpoint. These uncertainties leave the saved forecast at 38.8% Yes and 61.2% No.
Forecast history
| Date | Forecast |
|---|---|
| 39% chance of yes |
Sources Prolepsia read
- ICE BofA US High Yield Index Option-Adjusted Spread (BAMLH0A0HYM2) | FRED | St. Louis Fedfred.stlouisfed.org
- historical archiveraw.githubusercontent.com
- ICE BofA US High Yield Spread (BAMLH0A0HYM2): Chart & Data - Equiblesequibles.com
- Employment Situation News Release - 2026 M10 Resultswww.bls.gov
- Federal Reserve Board - Federal Reserve issues FOMC statementwww.federalreserve.gov
- High Yield 4Q 2026 Outlook | MacKay Shieldswww.nylim.com
- iFlow | Special Report | CCC stress isn’t contagious yetwww.bny.com
- # Treasury Spread and Probability of US Recessionwww.newyorkfed.org
The question
This question asks whether the ICE BofA US High Yield Index Option-Adjusted Spread (OAS), published by ICE Data Indices and available through the Federal Reserve Bank of St. Louis's FRED database under series identifier BAMLH0A0HYM2, will exceed 4.00% (400 basis points) at the close of December 31, 2027. The index measures the option-adjusted yield spread between US dollar-denominated, below-investment-grade corporate bonds issued in the US domestic market and the US Treasury spot yield curve. It is a widely followed indicator of credit risk and financial market conditions, with wider spreads generally reflecting greater perceived credit risk or tighter financial conditions. As of November 15, 2026, the latest independently verified observation is 3.30% on November 11, 2026, below the proposed 4.00% threshold. Recent observations include 3.42% on November 6 and 3.34% on November 9, 2026, illustrating fluctuations in credit spreads. The forecast therefore concerns whether spreads will widen sufficiently to exceed 400 basis points by the specified measurement date, rather than whether they will cross that threshold at any point beforehand. The measurement will use the daily closing value of FRED series BAMLH0A0HYM2 for December 31, 2027, or the most recent preceding trading day if no observation is published for that date. The forecast is made as of November 15, 2026, and covers the period through December 31, 2027. Source: Federal Reserve Bank of St. Louis, FRED, ICE BofA US High Yield Index Option-Adjusted Spread (https://fred.stlouisfed.org/series/BAMLH0A0HYM2).
How it resolves
The question resolves YES if the ICE BofA US High Yield Index Option-Adjusted Spread (FRED series BAMLH0A0HYM2) has a published daily closing value strictly greater than 4.00% (400 basis points) for December 31, 2027. It resolves NO if the published value is less than or equal to 4.00%. The primary resolution source is the Federal Reserve Bank of St. Louis's FRED database, using the ICE BofA US High Yield Index Option-Adjusted Spread series BAMLH0A0HYM2, sourced from ICE Data Indices, LLC. If December 31, 2027, has no published observation because of a market holiday, closure, or other routine interruption, the most recent preceding date with an officially published daily closing observation will be used. If the FRED series is temporarily unavailable, publication of the relevant observation will be awaited for up to 30 calendar days after December 31, 2027. If FRED permanently discontinues the series or does not publish the relevant observation within that period, the corresponding historical closing value published directly by ICE Data Indices will be used, provided it represents the same index and OAS methodology. If neither source supplies a verifiable observation for the specified measurement date or applicable preceding trading day within 30 calendar days after December 31, 2027, the question will be annulled rather than assigned an inferred outcome. Values will be evaluated as published in percentage points, without rounding beyond the precision supplied by the authoritative source. A value of exactly 4.00% resolves NO; any reported value greater than 4.00% resolves YES.
Fine print
The question concerns the option-adjusted spread (OAS), not the index's yield to maturity, effective yield, total return, or spread of an individual credit-rating segment. The threshold of 4.00% refers to 400 basis points of credit spread over the Treasury benchmark, not a 4% relative change in spreads. Only the specified end-of-period observation matters: a temporary increase above 4.00% earlier in 2027 does not qualify if the closing observation used for resolution is 4.00% or lower. Conversely, a value above 4.00% on the measurement date qualifies even if spreads were below that level throughout most of the year. The question uses the overall ICE BofA US High Yield Index OAS, not a maturity-restricted, rating-specific, or constrained variant. Ordinary changes in index constituents under ICE's established methodology do not invalidate the question. The original BAMLH0A0HYM2 series takes precedence over any replacement series unless ICE explicitly identifies the replacement as a direct continuation of the same underlying index and metric. The applicable calendar date is determined using the date assigned to the daily closing observation by the index provider, rather than the publication date or the time at which FRED updates its database. Subsequent routine revisions published before the resolution deadline will be incorporated; changes published after resolution will not reopen the outcome.
More public forecasts
Will the spread between French and German 10-year government bond yields close above 200 basis points on any day between 1 January and 31 December 2027?
43%chance of yes
Prolepsia forecast
What will the spread between French and German 10-year government bond yields be at the close on 31 December 2027, in basis points?
143median
Prolepsia forecast
Will global commercial insurance prices fall in the second quarter of 2027, as measured by the Marsh Global Insurance Market Index?
96%chance of yes
Prolepsia forecast
What will Japan's 10-year government bond yield be on 31 December 2027 (Ministry of Finance benchmark), in percent?
3.15median
Prolepsia forecast
Will US casualty insurance rates rise by more than 5% in 2027, as measured by the CIAB market survey?
27%chance of yes
Prolepsia forecast
Will the US average effective tariff rate be above 15% on 31 December 2027, as measured by the Yale Budget Lab?
16%chance of yes
Prolepsia forecast