Valuing a private loan book the way the regulator wants it evidenced

Equity in private markets is hard to value because the future is uncertain. A loan stays near par until a discrete event forces recognition. A performing loan pays its coupon and sits near par, which makes par the default answer, moved off only when something undeniable arrives. That reluctance to mark down early is a structural bias, and it maps onto two of the conflicts the FCA named: the incentive to show a smooth return profile, and, where fees or borrowing depend on NAV, the incentive to keep the mark high.

A loan also moves through distinct states, and each wants a different basis. A performing loan is valued by yield: the present value of its cash flows discounted at a rate reflecting current market spreads for comparable credit, so the mark shifts as credit conditions and borrower risk shift, even while the coupon keeps coming. A loan on the watch-list, covenant headroom eroding or performance softening, needs harder scrutiny of the assumptions. An impaired loan has to be valued on a recovery basis, an estimate of what the lender will actually get back, driven by the borrower’s enterprise value, the security, and where the loan sits in the capital structure. Value every loan as if it were performing, with no real watch-list or impairment discipline, and the book reprices in one violent step instead of gradually.

Then there’s the conflict underneath all of it: the originator marking its own loan. The credit team that sourced, underwrote and now runs the loan holds the best information on it and the strongest reason to read it kindly, because a markdown is partly an admission about their own underwriting. The review made functional independence a central expectation for exactly this reason.

Amortised cost can also be used to defer recognition. Where a structure lets loans sit at amortised cost rather than fair value, deterioration is easy to defer, since amortised cost only moves on impairment rather than tracking credit conditions continuously. Most fund vehicles report at fair value under IFRS 13. The expected-credit-loss staging that banks run under IFRS 9, from performing, to significantly increased credit risk, to credit-impaired, is a useful model for a fund’s own watch-list, because it forces deterioration to be recognised in steps rather than in a single jump. A credit manager should know the basis on which each vehicle values its loans, and shouldn’t let an amortised-cost convention, where one applies, become a reason to look past a borrower that’s plainly weakening.

Without a staged trail, marks move from par to a heavy markdown in one jump.

The March 2025 review set out what good valuation looks like across private markets. Each element has a sharper edge in credit.

Start with functional independence. The review found only a few firms could clearly show a valuation function separate in practice from the people running the assets, and it flagged senior investment professionals sitting as voting members of valuation committees. For a credit manager the test is whether a loan’s mark can be set and changed by someone other than the credit team that owns it, and whether there’s a recorded case of the valuation function moving a mark against the originator’s view. If the credit team’s number always prevails, the valuation function is only administrative.

On methodology, the review wanted firms to explain why they use the methods they use and to state the limitations. A credit policy should set out the yield basis for performing loans and the recovery basis for impaired ones, the key inputs, discount rates and spreads, expected loss and recovery assumptions, where those inputs come from, and where each method breaks down, rather than asserting one mechanical answer.

Defined triggers for interim valuation were the review’s central practical finding: most firms had no formal process and no thresholds for valuing assets between scheduled cycles. For a loan book, interim-valuation triggers are the priority. The triggers here are concrete, and the policy should name them: a covenant breach, a missed or deferred payment, an election to pay in kind under stress, a credit downgrade, a material miss against plan, a sharp drop in covenant headroom, a sector-wide shock. Each forces a fresh valuation rather than waiting for quarter-end. Naming them in advance is what answers the par-until-proven-otherwise habit.

The review also expected methodologies applied consistently and a secondary approach used to check the primary one. For loans that means testing the model mark against comparable market yields, against any broker quotes or secondary trade levels you can get, or against a third-party valuation, and reconciling what differs.

Its seven conflicts bear directly on the mark: fees charged on NAV, NAV or subscription facilities whose covenants ease as marks rise, marketing of unrealised performance, and carried interest. The one to watch in credit is the NAV facility, where a higher loan mark can widen the fund’s borrowing base or head off a covenant breach. That’s a direct reason to value optimistically.

On third parties, the review backed valuation advisers as a way to bring in independence and expertise, and for a credit manager an external valuer can be the cleanest answer to the originator-marks-its-own-loan problem, especially on the largest or most contested positions. It was just as firm that responsibility stays with the manager: keep your own people at arm’s length from the adviser so they can’t lean on it, assess the adviser’s strengths and limits rather than treat its output as final, manage the commercial conflict where the adviser’s fee depends on you, and disclose to investors what the third-party service covers. If the deal team briefs the external valuer directly and its assumptions go untested, the conflict returns.

Back-testing drew praise, and in credit it earns it, because loans realise, through repayment, refinancing or recovery, so a manager can set what it finally recovered against what it had been carrying the loan at. Recoveries that keep landing below carrying value are evidence the book was marked high, and a manager that back-tests catches that before a supervisor does.

Behind all of this sits the AIFMD valuation requirement, onshored in the FUND sourcebook at FUND 3.9 [R/UK]: the manager is responsible for the proper valuation of the fund’s assets and for calculating and publishing NAV, done impartially and with due skill, care and diligence under appropriate and independent procedures, whether through a functionally independent function or an external valuer. The depositary is a second line, required to oversee the valuation process under FUND 3.11.25 [R] and, for assets like loans that aren’t held in custody, to verify ownership and keep a record under FUND 3.11.23 [R]. One caveat to respect: FUND sets no numeric valuation frequency, the detailed method sitting in the AIFMD Level 2 regulation, so a manager shouldn’t claim a specific FUND-mandated cadence. Frequency follows the assets and the dealing model. IFRS 13 governs how the marks show up in the accounts.

A loan-valuation framework that meets the standard the review describes has eight working parts.

1. A credit-specific valuation policy that states, by loan type and state, the method, the inputs and their sources, and where each breaks down, and that defines the watch-list and impairment framework. A generic policy that can’t tell a performing loan from an impaired one isn’t fit for a credit book.
2. An independent valuation function with real authority: loans owned, or at least controlled and challenged, by people outside the credit team, with the expertise to interrogate a mark and the standing to change it. The proof it’s real is a record of it having moved a mark against the originator.
3. A watch-list and impairment process that moves a loan from performing to watch to impaired on defined criteria, so stress is recognised in stages rather than denied until default.
4. The named interim-valuation triggers, covenant breach, missed or PIK-elected payment, downgrade, underperformance, headroom erosion, sector shock, each forcing an out-of-cycle revaluation, with the trigger and the resulting mark written down.
5. A secondary validation on every material mark, checked against an independent reference, comparable yields, market levels or a third-party valuation, with differences reconciled and explained.
6. A credit conflicts map that ties the pressures on each loan’s mark, NAV-facility covenants above all, plus fees and carry, to the loan itself, so the firm can show a mark sitting near a borrowing covenant or a fee threshold was set clear of that pressure.
7. Back-testing against realisations: realised recoveries and repayments set against prior carrying values, the pattern analysed and fed back into the approach.
8. Documentation and board oversight, so an outsider could reconstruct every mark, change, trigger and challenge, the board sees management information showing the spread of marks, the watch-list, the impairments and the back-testing, and the valuation committee minutes record the reasoning, not only the result.

Concentration runs across all eight. The biggest loans carry the most valuation risk, both because an error in a large position moves the NAV materially and because supervisory interest in single-name and sector concentration means these are the marks a supervisor or a stress exercise reaches for first. A sensible framework gives them the most scrutiny, more frequent review, independent or third-party validation, deeper documentation, instead of treating a flagship loan and a small one with the same light touch. Independent challenge should scale with position size.

This insight is provided for general informational purposes only and doesn’t constitute legal, investment, or regulatory advice.