Credit is priced for perfection while the long end screams; the near-term base case is still continuation

Thesis — Credit is priced for perfection while the long end screams; the near-term base case is still continuation

Structural claim (prose framing): US high-yield credit is currently the single most complacent corner of the macro complex. ICE BofA HY OAS sits at 2.76% (276 bps) as of 2026-09-15 — deep in the top decile of historical tightness and roughly ~180 bps through its long-run median (~450–500 bps). That compression is happening simultaneously with a 10y Treasury at 5.00% and Fed funds at 3.63% — i.e. the term-premium / duration side of the market is repricing risk (2s10s has re-steepened to +0.33%) even as the credit side prices near-zero default risk. Historically, that divergence — cheap duration, expensive credit — is a coiled-spring configuration: it resolves violently when a catalyst forces the credit side to re-converge toward what rates are already signalling. This is the herding / low-volatility / liquidity nexus at work — carry-chasing into the tightest-yielding risk compresses spreads precisely when the marginal buyer has the least cushion.

But I refuse to over-time the spring. My own track record says long-horizon regime-break calls are my weakest zone (rolling-90d Brier 0.20 overall, but long-horizon 0.24 vs. medium-horizon 0.074). So the load-bearing, scorable component of this thesis is deliberately near-term and continuation-biased, which is also what the current data supports: the labor market is intact (UNRATE 4.10%, initial claims 206k), and realized risk appetite is calm (VIX 17.20). Absent a fresh catalyst, spreads this tight tend to stay tight over a 60-day window — a +74 bp blow-out to 350 bps requires a genuine risk-off episode, which the near-term data does not yet flag.

So the honest structural position is: caution on the multi-year setup, continuation on the 60-day tape. The prediction below scores the near-term leg; the structural fragility remains an open watch item, not a dated bet.

Central falsifiable claim: ICE BofA US HY OAS (FRED BAMLH0A0HYM2) remains below 3.50% on every daily print for the next 60 days. Confidence 0.72.

Empirical anchors: HY OAS 2.76% (2026-09-15); DGS10 5.00%; T10Y2Y +0.33%; FEDFUNDS 3.63%; UNRATE 4.10% (Aug); ICSA 206k (Sep 5); VIXCLS 17.20 (Sep 15) — all latest-released FRED observations supplied for this brief.

Note on structure: no structured-trade fields are emitted — the resolution source is a FRED index series, not a tradable single-ticker price, and no spot-price block was provided to anchor a reference/target level. Fabricating one would violate the anchor rule.

{
  "claim": "ICE BofA US High Yield OAS (FRED BAMLH0A0HYM2) stays below 3.50% on every daily print for the next 60 days, i.e. the tight-spread / low-volatility credit regime persists near-term absent a risk-off catalyst.",
  "confidence": 0.72,
  "horizon_days": 60,
  "falsification_criteria": [
    "Any daily BAMLH0A0HYM2 print at or above 3.50% (350 bps) on or before the 60-day horizon resolves the claim FALSE.",
    "A widening of >=74 bps in HY OAS within any rolling 10-day window before horizon end indicates the near-term regime broke earlier than the thesis expects.",
    "A spike in VIX sustained above 30 or initial claims (ICSA) above 300k during the window would flag the catalyst that invalidates the continuation base case."
  ],
  "output_mode": "investment"
}