Who Marks the Book? The Case for Independent Loan Administration in Private Credit

Private credit has spent a decade being valued largely by the people who own it. In a rising market, that was tolerable. In a market where redemptions are being tested, transfers are being priced and defaults are being worked out, it is the central question of the asset class.

The question is not whether private credit marks are wrong. It is who is in a position to demonstrate that they are right. And in a loan transfer there are always two sides: the fund selling or retaining the exposure, and the lenders or buyers taking it on. Most providers of record sit on one side of that line. The argument of this article is that the record itself should not.

1. The Ares continuation fund and the return of the valuation question

The clearest recent illustration came this year. Ares Management was forced to scale back a one billion euro private credit vehicle after investors pushed back against the valuation of the loans being placed into the continuation fund. The firm had sought to complete the vehicle but could not obtain buy-in on pricing from investors.

It would be a mistake to read this as a firm-specific stumble. Ares is among the largest and most sophisticated direct lenders in the world. The same period saw pressure across the sector as a whole: several of the largest non-traded business development companies received redemption requests well above their standard 5% quarterly caps and were obliged to prorate, and the sector recorded its first net outflow quarter in early 2026, with sponsors meeting close to 7 billion dollars of redemptions against under 5 billion dollars of gross sales. Managers have generally maintained, with justification, that the underlying portfolios continue to perform.

What failed was not the credit. What failed was agreement on what the credit was worth. A continuation fund requires the selling vehicle and the incoming capital to converge on a price for a portfolio that has no observable market. When they cannot, the transaction does not happen. That is a verification problem before it is a valuation problem, and certainly before it is a credit problem.

2. The same loan, three prices: private credit’s valuation problem

The dispersion in private credit marks is now well documented, and the examples are striking.

Two different lenders valued the same loan to Magenta Buyer, a cybersecurity company, at 79 cents and 46 cents on the dollar. Loans to the software company Medallia were valued simultaneously at 91%, 82% and 77% by three of its lenders, a 14-point spread in which one lender treated the loan as distressed while another marked it at 91%. A loan to HDT, an aerospace supplier, was valued between 85 and 49 cents on the dollar. These cases surfaced through the Bloomberg analysis that put the issue on the agenda under the heading of flawed valuations threatening the private credit boom.

The important analytical point is that this dispersion is not, for the most part, evidence of misconduct. It is evidence of information asymmetry and inconsistent data. Research into appraisal practice found that lead lenders consistently received higher appraisal values than participant lenders on the same loans during the pandemic, because lead lenders have direct access to borrowers through renegotiations and board observation rights, giving them nuanced information that participant lenders struggle to convey to appraisers in a verifiable format.

The same research points to what actually disciplines marks. Fifty-eight per cent of BDCs (US vehicles) are explicitly required by their own lenders to use third-party valuations, and where they do, only 4% of reported loan values exceed the appraiser’s recommended range and just 1.4% exceed it by more than one percentage point.

Mainstream practitioners make the same case. One large manager’s published view is that differences in third-party valuation providers, modelling assumptions and methodologies can reasonably produce a five to fifteen point range in marks across otherwise similar portfolios, particularly for stressed loans, and that this is primarily a transparency and timing issue rather than a fundamental credit story, with improved disclosure and more frequent marking likely to narrow the gap.

We agree with that diagnosis. It is precisely why the infrastructure question matters. If the problem is verifiable information reaching the right parties in a usable form, then the answer is architectural.

3. When the marker sets the mark: valuation, fees and conflict

The second axis of the problem is harder to solve with better data alone, because it concerns incentives.

Regulators have moved from general observation to specific concern. The Department of Justice has publicly warned about creative marks and divergent valuation practices in private portfolios. The SEC’s examination priorities identify overreliance on internal marks without sufficient independent validation or challenge, inconsistent application of valuation policies where methodologies differ across strategies or vehicles, failure to adjust valuations in distressed or covenant breach scenarios, and potential undisclosed conflicts of interest between valuation outcomes, performance reporting and fee calculations.

That last item is the structural point. Valuations in these vehicles are determined by fund managers exercising significant discretion, subject to external audit only annually, and because reported net asset value affects the fees a manager earns, valuation disputes implicate fee calculations directly and raise conflict of interest questions. The consequences are no longer hypothetical: the SEC has settled claims against a private fund manager for selling loans to private equity sponsors at par value without consideration of the loans’ fair market value.

The advice now being given to allocators reduces this to a single test. Investors are being told to ask each manager, in writing, who marks the portfolio and what triggers a write-down, on the basis that a manager who marks his own book is not the same as one whose valuations are reviewed by an independent third party.

That is the right question. It is also only the first half of it.

4. From static NAV to a live credit book: monitoring loans in real time

Whatever view one takes on valuation governance, the operational demands of the asset class have changed. The emerging consensus is that private credit should be monitored less like a static income allocation and more like a live credit book, looking beyond headline yield to valuation quality, cash interest coverage, payment-in-kind dependency, sector concentration and fund liquidity.

The evidence for that is accumulating on several fronts. Headline default rates have remained relatively low, but stress is more pronounced once selective defaults and liability management exercises are included, alongside falling interest coverage ratios and a marked rise in the use of payment-in-kind facilities. Structural triggers are being hit: a manager fee waiver followed the breach of an over-collateralisation test at a private credit CLO, a signal of portfolio stress sufficient to divert cash flows to senior tranches until metrics cure. Concentration is real, with software representing roughly 20% of the United States direct lending market, closer to 30 to 35% including adjacent industries, and around 10 to 15% for European direct lenders.

None of this can be monitored from quarterly fund-level financial statements. It requires loan-level data: reconciled positions, documented servicing history, covenant and compliance certificate tracking, accurate interest and PIK accrual, and verified collateral. Market commentary on private credit secondaries makes the point squarely, that portfolio accuracy depends on reconciled source data and documented servicing activity rather than fund-level statements alone, and that the administrative load of loan tapes, compliance certificates, collateral monitoring and borrower reporting is the binding constraint.

It is also where transactions are most fragile. Transfers create servicing continuity risk, because payment processing, interest accrual, covenant monitoring and notices must continue uninterrupted at exactly the moment scrutiny is highest, and any break shows up as a reporting gap in the diligence. Transfer documentation is granular and voluminous, varying by asset type, security, credit support and jurisdiction, and the structure chosen carries distinct consent and perfection consequences: whether the transfer is an outright assignment, an economic participation or a pledge determines what consent rights, disposition restrictions and change of control provisions are engaged under both the fund-level facility and the underlying loan documents.

A portfolio that cannot demonstrate this record does not simply price lower. It may not price at all.

5. Neither side’s agent, and therefore both. The case for independent loan administration.

Which brings us back to the two sides of the line.

The independence debate in private credit has so far been framed narrowly, as third-party review versus internal marks. That framing is necessary but incomplete, because it treats independence as a property measured only against the manager. In a transfer, a continuation fund, a secondary or an asset-backed facility secured on a portfolio of loans, there are two principals with opposed economic interests in the same record.

Under the conventional model, each has its own agent. A security agent acts for the lenders and holds collateral in their interest. A fund administrator acts for the fund and reports in its interest. Each is independent of the other side. Neither is neutral between them. That is not a criticism of either role; it is the definition of agency. But it means the record of what the loan book contains and what it is worth is produced by a party with a principal, and the party on the other side of the table knows it.

The position we think the market now needs is different. An administration and verification agent that is the agent of neither side, and therefore capable of serving as the record for both. A single reconciled source of truth on positions, covenants, collateral, servicing history and accruals, maintained by a party with no economic interest in whether the number is higher or lower, no fee that moves with net asset value, and no principal whose interest the record must serve.

That neutrality is not a marketing claim. It is a structural property, and it is the only basis on which a mark or a loan tape carries trust across a table rather than on one side of it. It is what allows a selling fund and an incoming lender to work from the same file. It is what allows an ABL lender to comfortably lend against a loan book. It is what allows a covenant calculation to be accepted rather than renegotiated. And it is what was missing when a continuation fund with sound underlying credit could not get its pricing agreed.

Private credit is not broken. The scrutiny it is receiving is legitimate, and the asset class will be better for it. But a market of this size can no longer rely on records produced by interested parties. The infrastructure has to catch up to the capital, and independence has to mean independent of both sides.

This is the standard Altrium is built to. We act as facility agent, security agent, escrow agent and collateral administrator across private credit, fund finance and asset-backed structures, and our collateral administration and loan book verification work is deliberately constructed so that we serve no side’s interest in the number. If you are pricing a transfer, standing up a continuation vehicle, or rebuilding a loan book that has to withstand diligence, we would welcome the conversation.