News · Credit · USBy Icarus Asia Research · · 10 min readDownload PDF
Credit Research · Credit & Restructuring · April 2026
Cleared for Public Distribution
Icarus Asia · Credit Research · Credit & Restructuring · April 2026

A canary in
a coal mine?
U.S. first-lien creditors are
getting back less than ever

Five primary sources tracked by Icarus Asia point to a credit market that has been systematically repriced against the interests of senior secured lenders. Bankruptcy recoveries in 2025 reached a decade low of 43 cents on the dollar. Moody's forward LGD sits 19 points below its own historical mean. Liability management transactions have wiped out recovery entirely in 14 documented cases.

CoverageFirst-Lien Loans · CLOs · LMTs ClassificationCleared for Public Distribution DateApril 2026
Overview
Key Metrics
First-Lien Recovery — 2025 Snapshot
Four signals. Three agencies. One direction.
43%
Fitch 2025 bankruptcy recovery (1st lien)
Decade low · 10yr avg: 62%
88.4%
S&P loan recoveries YTD 2025 (through Sept)
Above long-term avg of 75.4%
~68%
Moody's forward LGD estimate — 1st lien
vs ~87% historical avg · −19 pts gap
70 pts
Avg recovery cut · most-disadvantaged LMT lenders
Zero recovery in 14 of 38 S&P-tracked cases
Sources: Fitch Ratings (Apr 2026); S&P Global Ratings (Dec 2025); Moody's as cited in Icarus Asia research
Recovery Benchmarks
First-Lien Recovery — All Measures
Scale: 0–100% recovery. Crimson bars = stress signals.
S&P 2025 YTD — Loans 88.4%
Above long-term avg · includes distressed exchanges
Moody's Historical Avg — Loans ~87%
Through-the-cycle benchmark
S&P Historical Avg — Loans 75.4%
Long-term average for all loan defaults
Moody's Fwd LGD Estimate ~68%
Forward-looking · 19 pts below Moody's own historical mean
Fitch 10yr Avg — 1st Lien 62%
Ten-year par-weighted average, bankruptcy cases only
Fitch 2025 ex-Diamond — 1st Lien 52%
Excluding Diamond Sports — still 10 pts below 10yr avg
Fitch 2025 — Bankruptcy (1st Lien) Decade Low 43%
Formal bankruptcies ≥$100M first-lien debt · par-weighted nominal
S&P Bonds 2025 YTD Lowest Since 2001 21.3%
Unsecured bond recovery — lowest in 24 years
0%25%50%75%100%
The gap between S&P's strong loan figure and Fitch's bankruptcy figure is not a contradiction — it's a measurement. Fitch captures what happens when companies actually fail. S&P includes distressed exchanges that resolve faster and higher. Both are true. Together they show first-lien recoveries running below historical norms.
Sources: Fitch Ratings (Apr 2026); S&P Global Ratings (Dec 2025); Moody's as cited in Icarus Asia research
LMT Mechanics
Four Ways to Rewrite Priority
Click each box to expand the mechanics and case examples.
Type 1
Drop-Down
Transfer
Collateral / asset transfer
▼ tap to expand
Type 2
Uptier
Priming
Priming loan exchange
▼ tap to expand
Type 3
Double-Dip
Dual-claim nonobligor structure
▼ tap to expand
Type 4
Pari-Plus
Structural + pari passu seniority
▼ tap to expand
Type 1 — Drop-Down / Asset Transfer
Assets transferred to nonobligor subsidiary strip collateral from existing lenders. New secured debt raised against transferred assets gives structural priority over transferred value. Travelport (−75pts), J.Crew (−25pts), Revlon (−25pts) all used this mechanism.
Type 2 — Uptier Priming
Majority lender group amends credit agreement to insert new superpriority debt. Non-participants left with a position that is "first-lien" in name only. Serta Simmons: 55%→5% (−50pts). Trimark: 55%→0% (−55pts).
Type 3 — Double-Dip
New secured debt at nonobligor affiliate. Proceeds lent back as intercompany loan, also secured pari passu. New lenders hold TWO claims on the same collateral. Home and Wheel Pros examples.
Type 4 — Pari-Plus
New debt at nonobligor with guarantees from nonobligors holding valuable assets. Proceeds lent back as intercompany loan. New lenders get pari passu claim on legacy obligors PLUS structural seniority at new borrower. Sabre and Trinseo examples.
"The growing use of liability management transactions has become an important factor shaping recovery outcomes, particularly for lenders that do not participate in these transactions."
— Joshua Clark, Senior Director, Fitch Ratings, April 6, 2026
Source: S&P Global Ratings; Fitch Ratings; Icarus Asia case compilation
LMT Damage Table
Recovery Cut — Most-Disadvantaged Lenders
Click a row to see post-LMT outcome. Scale: recovery points cut (larger = worse).
Travelport (priming) −75 pts
Subsequently bankrupt
Murray Energy −65 pts
Subsequently bankrupt
Magenta Buyer (multistep) −65 pts
Subsequently bankrupt
TriMark / TMK Hawk −55 pts
Subsequently bankrupt
Serta Simmons −50 pts
Subsequently bankrupt
Boardriders −50 pts
Redefaulted / CCC+
RobertShaw (multistep) −50 pts
Subsequently bankrupt
GoTo Group −45 pts
Redefaulted / CCC+
0 pts20 pts40 pts60 pts80 pts
Source: S&P Global Ratings, Credit FAQ: Demystifying Loan LMTs, October 30, 2024
Structural Shifts
Four Drivers of Weaker Recovery
Structural market changes — not cyclical noise.
Debt Cushion Erosion 5%
Share of first-lien loans with debt cushion ≥ first-lien size since 2021. Down from over one-third historically. More than 83% now carry a cushion of 25% or less.
Covenant-Lite at Scale 9.3 pts
First-lien recovery penalty from cov-lite structure, confirmed across decades of data. Cov-lite: <10% of TLs in 2008 → >90% today. The penalty that was marginal is now portfolio-level.
LMT Frequency 14 of 38
LMT cases where expected recovery for disadvantaged lenders fell to zero. A primed first-lien may recover at second-lien levels (~42¢) not first-lien levels (~79¢).
All-Secured Structures ~0%
Junior debt cushion in 2023–24 LBO financing. Sponsors moving to near-entirely secured structures. With no junior debt to absorb first losses, first-lien is next in line.
Sources: S&P Global Ratings; Fitch Ratings; LCD / PitchBook; Icarus Asia synthesis
Icarus Asia · House View
The Averages Remain. The Market Is Gone.
CLO assumptions vs. empirical reality — a growing gap.
CLO Model Assumption Gap
CLO Model Assumption
70–80%
Through-the-cycle 1st lien recovery
Moody's 2023 Actual
59.6%
Empirical 1st lien recovery
A 10–20 percentage point recovery shortfall has a material effect on expected cash flows and stress-scenario outcomes for CLO equity and mezzanine tranches.
S&P Global Ratings · Oct 2024
"With the threat of these restructurings by distressed entities, lenders can no longer rely on realizing average par recoveries of 75%–80% (and more than 90% on a median basis) by virtue of their position at the top of the capital structure, with liens on substantially all assets."
Icarus Asia Analysis
"In 14 of the 38 LMT cases S&P tracked from 2017 through August 2024, recovery expectations for the most disadvantaged lenders reached zero. Senior secured, first-lien, contractual priority — none of it prevented the outcome. The historical averages that anchor CLO models, risk weightings, and private credit return assumptions were built on a market with junior debt cushions, maintenance covenants, and no out-of-court mechanism for majority lenders to subordinate holdouts. That market is gone. The averages remain."
Sources: Moody's; S&P Global Ratings; Fitch Ratings; Icarus Asia Credit Research, April 2026
Section 1 — Overview

Two numbers that appear to contradict each other.
They don't.

The headline data looks broken. Fitch reports 2025 first-lien bankruptcy recovery at 43% — a decade low. S&P reports 2025 YTD loan recovery at 88.4% — above its own long-term average. Both figures are correct. They measure different things.

Fitch captures only formal bankruptcy cases with at least $100 million in first-lien debt, on a par-weighted nominal basis. S&P covers all default types including distressed exchanges, on a discounted basis. Distressed exchanges now dominate the default mix. That composition shift is doing most of the work in the S&P number.

Moody's forward LGD estimate of ~68% — against its own historical average of ~87% — is the signal that unifies all three. Every agency, read carefully, points the same direction: first-lien recovery expectations have reset lower.

The canary is the Moody's gap. A 19-point delta between historical mean and forward estimate is not noise. It is a structural repricing of senior secured risk that CLO models and bank credit systems have not yet absorbed.
Section 2 — Recovery Benchmarks

Read together, all three agencies tell
the same story.

S&P's 88.4% loan recovery figure is real — but it is pulled up by the dominance of distressed exchanges, which by definition settle before the assets deteriorate further. Remove that methodology difference and the S&P data aligns with the directional signal from Fitch and Moody's.

Fitch's 43% bankruptcy recovery is the cleanest apples-to-apples measure of what senior lenders receive when a company actually fails. The 10-year average is 62%. The 2025 reading is 19 points below that. Even stripping out the Diamond Sports outlier, the ex-Diamond figure lands at 52% — still 10 points below the long-run mean.

The S&P bond recovery reading tells a parallel story at the unsecured layer: 21.3% in 2025, the lowest since 2001. Less junior cushion means enterprise value shortfalls arrive at the first-lien layer sooner.

Recovery ratings are backward-looking averages. Neither S&P nor Fitch can capture prospective LMT risk in issue or recovery ratings. The published recovery rating on a first-lien instrument does not reflect the probability that a majority lender group will use that instrument to subordinate the rating agency's assumed recovery scenario.
Section 3 — LMT Mechanics

How contractual priority
gets rewritten out of court.

S&P identifies four primary liability management transaction structures. Each one achieves a similar outcome through different legal architecture: a subset of lenders — typically those with enough votes to amend — improve their position relative to lenders who do not participate or are excluded.

The drop-down transfer moves assets out of the collateral package. The uptier priming inserts new debt above the existing first lien. The double-dip gives new lenders two claims on the same collateral. The pari-plus adds structural seniority at a new entity on top of the legacy pari passu claim.

What all four share: none require court approval. They are contractual maneuvers executed under existing credit agreement flexibility, often in hours or days. By the time holdout lenders respond, the collateral or seniority has already moved.

"The growing use of liability management transactions has become an important factor shaping recovery outcomes, particularly for lenders that do not participate in these transactions."
— Joshua Clark, Senior Director, Fitch Ratings, April 6, 2026
Section 4 — LMT Recovery Damage

Seventy points. On average.
Zero, in 14 cases.

S&P tracked 38 LMTs from mid-2017 through August 2024. For the most-disadvantaged lenders — those primed, excluded, or left outside the transaction — recovery expectations were cut by an average of nearly 70 percentage points.

In 14 of those 38 cases, the expected recovery for disadvantaged lenders fell to zero. Not to second-lien levels. Not to 20 cents. To nothing. Senior secured, first-lien, with liens on substantially all assets — and nothing.

The majority of these companies subsequently filed for bankruptcy or redefaulted at CCC+ or below. The LMT did not prevent insolvency. It determined who got paid when insolvency arrived.

Icarus Asia — Creditor Risk Note

A primed first-lien position in an LMT context should be modelled at second-lien historical recovery levels (~42 cents on the dollar), not first-lien levels (~79 cents). The contractual label "first-lien" no longer determines economic priority once an uptier or drop-down has been executed. Click any bar above to view the post-LMT outcome for that issuer.

Section 5 — Structural Shifts

Four structural changes.
None of them cyclical.

The current recovery environment is not a temporary product of the rate cycle or a thin deal vintage. It reflects four durable structural changes to the leveraged loan market that have compounded over the past 15 years — each of which independently reduces first-lien recovery, and all of which now operate simultaneously.

Debt cushion erosion removes the junior buffer that historically absorbed first losses before the first-lien layer was reached. Covenant-lite documentation removes the early-warning triggers that gave lenders intervention rights before value deteriorated. LMT proliferation moves assets or seniority out of reach. All-secured structures eliminate the subordinated debt layer entirely.

The 9.3-point cov-lite recovery penalty identified in S&P's multi-decade dataset was marginal when cov-lite represented a small fraction of issuance. At greater than 90% market share, it is a portfolio-level headwind applied to the overwhelming majority of current BSL inventory.

These shifts are not additive — they are multiplicative. A first-lien loan issued into a cov-lite, all-secured, thin-cushion capital structure that subsequently faces an LMT is exposed to all four headwinds at once. The historical average was built on a market where none of these conditions applied at scale.
Section 6 — Implications & House View

The models haven't moved.
The market has.

CLO models and bank credit systems still assume through-the-cycle first-lien recovery rates in the 70–80% range. Against Moody's 59.6% empirical figure for 2023 and forward LGD estimate of ~68%, assumptions at the upper end of that range overstate expected recovery on current-vintage leveraged loan portfolios.

A 10–20 percentage point recovery shortfall is not a rounding error. For CLO equity and mezzanine tranches, it materially affects expected cash flows, credit enhancement adequacy, and stress-scenario breach levels. For private credit funds running concentrated first-lien books, it alters the risk-return relationship that anchors underwriting.

Recovery ratings assigned by S&P and Fitch are not, and cannot be, forward estimates of what a lender will receive in an LMT scenario. An instrument can carry a "1" or "2" recovery rating — implying 90%–100% or 70%–90% recovery — and still be economically subordinated to zero by an uptier executed tomorrow morning, without court intervention and without any change to the rating.

Icarus Asia — Credit Research · April 2026

"In 14 of the 38 LMT cases S&P tracked from 2017 through August 2024, recovery expectations for the most disadvantaged lenders reached zero. Senior secured, first-lien, contractual priority — none of it prevented the outcome. The historical averages that anchor CLO models, risk weightings, and private credit return assumptions were built on a market with junior debt cushions, maintenance covenants, and no out-of-court mechanism for majority lenders to subordinate holdouts. That market is gone. The averages remain."

First published by Icarus Asia · Original publish date:

Icarus Asia Research
Research published by Asian Value Investor. Author →

This piece is editorial analysis and does not constitute investment advice. See the Disclaimer.