LIVE EVENT
GCN Investor Conference in Newport Beach, CA
OCT 15 · NEWPORT BEACH, CA
LIVE EVENT
For a decade private credit's central claims have been unfalsifiable in practice, because the conditions that would test them never arrived. That is changing. This report sets out what the test consists of and what evidence would settle each claim.
This is an outlook, not a retrospective. Every statement about 2026 below should be read as a scenario or a monitoring question, not a forecast.
Private credit enters 2026 in an unusual position: a large, established, institutionally-central asset class whose foundational claims have never been tested, and whose test appears to be arriving.
The 2016 report set out three claims: that direct lenders underwrite better because they hold the whole loan, that they achieve better workout outcomes because concentration gives them incentive and authority, and that relationship lending produces better information. All three were plausible. All three required a default cycle to evaluate.
That cycle has been repeatedly deferred rather than avoided. The 2020 shock was removed by policy within weeks. The 2022 rate rise raised borrower costs sharply but the resulting stress was absorbed by cash balances, sponsor equity, amendments and PIK — as the 2022 report describes. Each of those mechanisms defers recognition without changing economics, and each has finite capacity.
By 2026 much of that capacity has been consumed. Cash raised in 2021 has been spent. Sponsors have supported companies for several years and face their own return constraints. Loans have been amended, sometimes repeatedly. PIK balances have grown, which means the eventual obligation is larger than the original.
What follows is not a prediction that the asset class will perform badly. It is an argument that 2026 is when the questions become answerable, and a specification of what evidence would answer them.
Each mechanism the 2022 report identifies has a capacity constraint, and it is worth setting each out because they exhaust in a predictable order.
Cash balances. Borrowers who raised in 2021 held cash that absorbed higher interest costs. Cash is consumed by use. A borrower that has funded a coverage shortfall from the balance sheet for three years has less balance sheet.
Sponsor equity support. A private equity owner can inject capital into a portfolio company facing a coverage problem. The constraint is the sponsor's own economics: the injection reduces the return on an investment that is already impaired, and beyond a point the rational decision is to stop. That point arrives sooner for a fund near the end of its life — which, as the 2026 US venture report notes for a different asset class, describes a large share of 2021-era vehicles.
Amendments. Lenders amend rather than default when they believe the borrower will recover. Repeated amendment on the same credit becomes harder to justify, and lenders' own auditors and valuation committees scrutinise repeatedly-amended loans more closely.
PIK. Converting cash interest to PIK removes the immediate requirement entirely, which is why it is the last mechanism to be exhausted. But it compounds: a borrower on PIK accrues a growing balance against unchanged earnings, so leverage rises every quarter without new borrowing. PIK does not fix a coverage problem — it converts it into a leverage problem and defers it.
Every deferral mechanism trades a present problem for a larger future one. That is the correct trade when recovery is coming. It is a compounding mistake when it is not, and the decision has to be made before you know which.
The practical question for 2026 is how much capacity remains. That is partly observable — cash balances and PIK share are disclosed for listed borrowers and for BDC portfolios — and partly not.
Each foundational claim has a specific evidence trail, and specifying them in advance is the point of an outlook.
The claim: holding the whole loan produces more careful underwriting than distributing it.
What would support it: default rates on direct loans below comparable syndicated loans, controlling for borrower size and sector. Loss-given-default at or below modelled assumptions.
What would undermine it: default rates at or above syndicated comparables, particularly concentrated in the 2021 vintage originated under deployment pressure — which would suggest that capital pressure overwhelmed the structural incentive.
Where to look: BDC non-accrual rates versus the S&P/LSTA leveraged loan index default rate, by vintage where disclosed.
The claim: a concentrated lender achieves better recoveries because it can act decisively.
What would support it: realised recoveries above the historical syndicated average, and above the lender's own modelled assumptions.
What would undermine it: recoveries at or below syndicated averages, which would suggest that the loss of the option to exit outweighs the benefit of control. Also relevant: whether concentrated lenders convert debt to equity and hold, which preserves the mark while deferring realisation — a possibility that makes recovery statistics themselves hard to interpret.
Where to look: disclosed realisations in BDC filings; restructuring outcomes reported in fund communications.
The claim: ongoing relationships produce earlier warning of deterioration.
What would support it: non-accrual designations preceding covenant breaches and payment defaults by a meaningful margin — evidence that lenders identified problems before the mechanical triggers did.
What would undermine it: non-accruals appearing simultaneously with or after payment problems, indicating the information advantage did not translate into earlier action.
Where to look: the sequence of disclosures in BDC filings — when a loan was first marked below cost, when it was placed on non-accrual, when a restructuring occurred.
A note on all three: these are testable, and they will be tested by the cycle rather than by argument. That is unusual and worth valuing. Most claims in asset management are never resolved because the conditions that would resolve them do not recur cleanly.
If a reader takes one operational point from this report, it should be this.
Payment-in-kind income share is disclosed quarterly by listed business development companies, is free to obtain from SEC EDGAR, and is more informative about the state of private credit than any other public series.
Why it matters more than default rates:
What to watch specifically:
Default rates tell you what has been recognised. PIK share tells you what is being deferred. In an asset class with five ways to defer recognition, the second is the more informative number.
The three-way vintage split described in the 2022 and 2024 reports becomes decisive in 2026.
Aggregate manager statistics blend all three. A manager reporting stable portfolio metrics may hold a deteriorating 2021 book offset by a strong 2022–2023 book. Those are different assets with different futures, and the blended number describes neither.
The practical guidance for anyone assessing a manager in 2026: ask for the portfolio split by origination vintage, and ask for non-accrual and PIK by vintage rather than in aggregate. A manager who cannot or will not provide it is a manager whose aggregate number should be discounted.
One distinction is worth holding clearly, because the two questions are routinely merged.
The performance question: will private credit deliver returns commensurate with its risk through a full cycle?
The systemic question: will stress in private credit transmit to the broader financial system?
These can resolve independently. The asset class could deliver disappointing returns without systemic consequence — losses borne by long-dated institutional capital that can absorb them, which is arguably the system working as designed. Or it could deliver acceptable returns while a liquidity event in retail vehicles or a contraction in lending capacity produces broader effects.
The 2024 report identifies three transmission channels — correlated behaviour, funding channel effects, and interconnection with banks. To those, 2026 adds the semi-liquid vehicle channel, where the valuation question and the liquidity question are the same question, as that report sets out.
The genuine structural improvement should not be lost in the risk discussion. Long-dated committed capital is a more stable holder of illiquid assets than demandable deposits. If leveraged lending must sit somewhere, a closed-end fund with a ten-year life is a better place for it than a bank balance sheet. That argument is sound and it survives everything else in this report.
The qualification is that it applies to closed-end committed capital and applies less as the asset class's funding mix shifts toward vehicles offering redemption.
Evidence the asset class is holding up: non-accrual rates rising modestly while realised recoveries stay near modelled assumptions; PIK income share stable rather than climbing; amendment activity declining.
Evidence it is not: PIK share rising quarter over quarter, particularly amended rather than contractual PIK; non-accruals concentrated in the 2021 vintage and spreading; recoveries below syndicated comparables.
Evidence the systemic channel is activating: redemption pressure in semi-liquid vehicles; gates imposed; bank lending to non-bank financial institutions contracting, which the Federal Reserve's H.8 release tracks weekly.
Evidence the structural case remains intact: bank share of leveraged lending staying low, confirming that the regulatory origin argument still holds and the asset class's supply is durable regardless of how this cycle resolves.
This is the closing report of a five-part sequence (2016, 2020, 2022, 2024, 2026), and the sequence made a single claim repeatedly. It should be stated plainly so it can be judged.
The claim: private credit's structural case is sound and its performance claims are unverified, and the two have been consistently conflated.
2016 set out the regulatory origin — bank retrenchment from leveraged lending is a rule change, not a cycle — and identified three claims that could only be tested by a default cycle: better underwriting, better workouts, better information.
2020 argued that the record generated that year measured the policy response rather than the underwriting, because the stress test was cancelled within weeks. The asset class then grew substantially citing that record as evidence of resilience.
2022 argued that floating rate protection is two-sided — the lender's yield and the borrower's distress being the same cash flow — and that the resulting stress was being deferred through five mechanisms rather than resolved. It identified the vintage split that has since become the most important analytical distinction in the asset class.
2024 argued that scale had converted an allocation question into a systemic one, through three transmission channels that were plausible and unquantified, and that supervisors had identified interconnection as a data gap rather than a resolved question.
2026 argues the deferral capacity is approaching its limits and specifies what evidence would settle each of the original three claims.
What has held: the structural case, entirely. Banks have not returned to leveraged lending and the supply of loans requiring a non-bank lender is durable. The asset class exists for the reason 2016 identified and that reason has not changed.
What remains open: every performance claim. Nine years after the first report in this sequence, the three propositions that distinguish direct lending from syndicated markets have still not been tested by the conditions that would test them.
The falsifiable version: if defaults normalise and realised recoveries come in at or above modelled assumptions with PIK share stable, the claims are vindicated and this sequence was excessively cautious. If recoveries disappoint and PIK share climbs, the sequence was right that a benign-period record was being over-interpreted.
An asset class can be structurally sound and commercially unproven at the same time. Private credit has been both for a decade, and the persistent error has been treating evidence for the first as evidence for the second.
An outlook, not a retrospective. Its purpose is to specify what the long-deferred test of private credit consists of and what evidence would settle each of its central claims.
Where figures appear they carry a numbered source. Mechanisms — deferral capacity exhaustion, PIK as a leading indicator, vintage-dependent outcomes, the separation of performance and systemic questions — are analysis with reasoning shown.
Every forward-looking statement is framed as a scenario or a monitoring question. None should be read as a forecast, and this report makes no claim about how the cycle will resolve.
AI Investment Report 2025 — Where the Value Accrues precedes this report in the asset class sequence.
Private Credit Outlook 2027 — Reading the Answer follows this report in the asset class sequence.
Global Investment Outlook 2026 — The Return Question covers the same year at global multi-asset level.
US Venture Capital Outlook 2026 — What Has to Happen covers the same year in North American private markets.
Accredited investors receive our market reports, private event invitations and curated deal flow.
.webp){kind=logo}
.png)
.png)