Private credit valuations are under the microscope. The FSB's May 2026 report on vulnerabilities in private credit flagged valuation practices, opacity in credit quality, and liquidity mismatches as key concerns. In early 2026, a wave of redemption requests at semi-liquid private credit funds put the question of stale marks front and center. The message from regulators and LPs is consistent: reported NAVs need to be defensible, not just plausible.
At the heart of this is a structural problem. Most private credit borrowers are unrated. There is no S&P or Moody's rating to anchor to. The funds navigate the gap differently, and the range of what "credit monitoring" actually means in practice is wide.
What happens in practice when rating is not available
Reliance on the underwriting case is standard in direct lending, particularly in buy-and-hold strategies. The original credit memo clearly details the rating rationale. Then comes the ongoing monitoring; its goal is to flag significant deviations from the base case rather than re-rating from scratch. Efficient. But it anchors to a view that was formed at origination, often in a different market environment, and it does not produce a standalone rating estimate.
For those running a shadow rating, single-ratio proxies are the most common. Net leverage is the dominant metric, partly because it is commonly embedded into the covenants and hence is easily available from underlying borrowers. ICR comes second. Fast, simple, auditable. But blind to business quality, sector differences, and capital structure nuance. Also, a single ratio never tells the full story. A 4x net leverage ratio means something very different for a capital-light software business and an asset-heavy manufacturer.
At the more sophisticated end, you have multi-factor scorecards. Typically four to six financial ratios with a qualitative business risk overlay, informed by S&P or Moody's observed rating methodologies. Larger managers build these in-house. The challenge is the resource cost of building and maintaining the model, and ensuring the qualitative overlay adjusts as market conditions change.
Third-party tools are growing in popularity. S&P CreditModel is one of the established names. The appeal is standardisation, audit trail, and reduced model risk. Adoption is being driven by LP due diligence pressure and auditor scrutiny on Level 3 valuations.
Of course there is also an option of doing nothing. Keeping all loans at par until proven otherwise. However, not a real option anymore, but more common than anyone would like to admit.
Why getting it right matters more than ever
The pressure is structural, not cyclical. Funds are moving toward open-ended structures, more frequent valuations, and broader retail investor bases. AIFMD II has raised the bar on what adequate risk monitoring documentation looks like. Regulators and investors are paying closer attention, and in a competitive market, the ability to demonstrate rigour is not just a compliance exercise. It builds trust. And trust, over time, drives AUM.
The rating drives the price
What often gets lost in the monitoring conversation is the direct connection to valuation. A credit rating is not just a risk label. It determines the credit spread. The credit spread determines the discount rate. The discount rate determines the fair value of the loan. In a fund with quarterly NAV reporting and active subscriptions and redemptions, a stale credit rating does not stay in the risk monitoring team, it flows straight into the price at which investors enter and exit the fund. Getting it wrong is not just an audit problem. It is a fairness problem.
What this environment demands is a rating that produces a number investors can trust, and that will stand up to regulatory scrutiny. That means documentation as much as methodology. The number matters, but so does the trail behind it: what data was used, how it was weighted, when it was last updated, and what it implies for the credit spread used in valuation.
That trail is what a standardised rating tool is for. The smartZebra synthetic rating scores a borrower on credit ratios across five areas of credit assessment, weighted by sector or individually, adds a qualitative overlay, and returns a shadow rating together with the matching credit spread for maturities from 1 to 30 years across 15 sectors - each figure carrying the source timestamp and calculation path an auditor asks for when the question turns to where the discount rate came from.
References
- Financial Stability Board, Report on Vulnerabilities in Private Credit, 6 May 2026 - names valuation opacity and reliance on private credit ratings as factors that "can amplify strains in stress"; puts the market at an estimated USD 1.5-2.0 trillion at end-2024; recommends closing data gaps and sharing supervisory approaches to risk management and valuation practices
- Directive (EU) 2024/927 of 13 March 2024 (AIFMD II), amending Directive 2011/61/EU - published in the Official Journal on 26 March 2024, in force since 15 April 2024, Member States to apply the transposed rules from 16 April 2026 (the Article 24 reporting measures from 16 April 2027). New Article 15(3)(d) AIFMD: an AIFM managing a loan-originating AIF must implement effective policies, procedures and processes for granting loans, assessing credit risk and administering and monitoring its credit portfolio, keep them up to date and review them regularly and at least once a year. Wording paraphrased from secondary summaries; the exact Official Journal text was not retrieved for this draft
- IFRS 13 Fair Value Measurement, paras. 72-90 - the fair value hierarchy; para. 86 defines Level 3 inputs as unobservable inputs for the asset or liability, which is where an unrated private loan priced off an internal rating sits; para. 93 the Level 3 disclosure requirements, including the valuation techniques and inputs used
- PGIM, Investors at the Private Credit Gate, 2026 - describes the surge in redemption requests at non-traded BDCs in the first quarter of 2026 and the three responses seen: a secondary sale of the portfolio, redemptions above the contractual cap funded from firm resources, and holding redemptions to the contractual 5% with the excess queued. Industry commentary, not a primary source
- S&P Global Market Intelligence, CreditModel - the third-party scoring model the author names; the product's own methodology documentation was not reviewed for this draft
Related pages
- Credit Spreads & Interest Rates - synthetic ratings for unrated borrowers, plus the spread and yield-curve leg of the discount rate, by tenor, sector and rating
- Data Solutions for AIFMs - what fund managers use smartZebra for under AIFMD valuation and risk-management requirements
- Introduction to Loan Valuation and the Private Debt Market - how credit risk gets priced where there is no market quote to check against
- PIK Toggle Loans: Why Monitoring Matters - the FSB's PIK data and why payment-in-kind removes the early warning signal a rating has to replace
- Valuation Frequency vs. Data Lag - Friday Digest - open-ended structures, more frequent NAVs and what that asks of the inputs
- Can We Trust Automation? - Friday Digest - the audit trail argument applied to platforms and AI
- Grounding / Facts - entity definition, data provenance and methodology
