Fund repos: trace N-PORT collateral beyond cash labels
A fund’s repo holding is not adequately described by a cash-like label. The research task is to identify who lends cash, what collateral supports repayment, when the agreement matures and which values are being compared. Collateral coverage is a useful calculation, but it is not a guarantee of immediate recovery.
This guide reads the SEC reference Form N-PORT marked effective December 11, 2023, specifically the scope statement, Part C.2 and C.10, on October 26, 2026. It uses that document for the cited field definitions, not as a claim that every filing deadline or public-release rule in the reference copy remains current. All positions, counterparties and amounts below are fictional.
Establish the fund’s direction before calculating
Form N-PORT’s scope and Item C.10.a identify the relevant fund population and distinguish the transaction direction. Under this form’s terminology, the fund is the cash lender receiving collateral for a repurchase agreement; it is the cash borrower posting collateral for a reverse repurchase agreement.
The form excludes funds regulated as money market funds under Rule 2a-7. Do not substitute this exercise for an N-MFP analysis merely because a product name contains “cash,” “liquidity” or “money market.” Identify the actual reporting vehicle and fund series.
For each holding, retain fund identity, reporting date, accession and amendment. The monthly portfolio date is not the date the public obtained it, and a downloaded snapshot is not a live collateral monitor. Verify the applicable filing and availability rules separately if the research question depends on information timing.
Separate three different amounts
Part C.2 and C.10.f provide investment amount and value fields and information about the collateral, including principal amount, collateral value and category. The principal amount of a collateral security is not automatically the cash lent under the repo.
Use a directionally explicit worksheet: cash-lending principal, reported repo holding value, repayment claim at the selected date, collateral face amount and collateral market value. They can differ. An investor should not choose whichever number makes coverage look highest.
For example, a collateral bond can have face amount 108 but current value 104. Comparing the repo loan of 100 with face amount 108 produces 108%, while comparing it with current collateral value produces 104%. The former is not the market-value coverage calculation in this guide.
The assumed repo principal of 100 is a separately supported input in the example. If a public filing supplies only a holding value and insufficient terms to reconstruct principal or accrued amounts, state that limitation. Do not rename a reported value “principal” without support.
Work the original coverage and two shocks
Assume a fund lends 100, current collateral value is 104 and the simplified comparison claim is principal only. Ignore accrued interest, margin transfers, substitution, currency changes and costs for this first calculation. Collateral face amount is not used.
| Scenario | Collateral value | Claim compared | Coverage | Principal shortfall |
|---|---|---|---|---|
| Initial | 104.00 | 100 | 104.00% | 0 |
| Collateral falls 3% | 100.88 | 100 | 100.88% | 0 |
| Collateral falls 6% | 97.76 | 100 | 97.76% | 2.24 |
Initial coverage is 104 / 100 = 104%. Under the first shock, 104 × 0.97 = 100.88. Under the second, 104 × 0.94 = 97.76 and the principal-only shortfall is max(100 − 97.76, 0) = 2.24.
The 6% price decline is not a 6% loan loss, and the 2.24 is not an observed realized loss. The result is a controlled collateral-value gap assuming no response, before considering the contract and counterparty recovery. A margin call, payment or substitute collateral could change the facts.
The initial 4% excess measured against the loan is also different from a haircut measured against collateral value. With a haircut defined here as (104 − 100) / 104, the result is approximately 3.8462%. State the denominator instead of using “4% haircut” as an interchangeable label.
Add interest and liquidation assumptions in separate layers
Assume separately a 5% annual simple repo rate, a 360-day convention and 30 days of interest, with no other payments. Interest on principal 100 is 100 × 0.05 × 30 / 360, approximately 0.416667. The comparison claim would be 100.416667 under those stated terms.
Do not apply that maturity claim to a different date without adjusting accrual. The public rate and maturity fields do not by themselves disclose every contractual detail, day-count convention or already-paid amount. Those are explicit extra assumptions in this calculation.
For a separate liquidation scenario, apply an additional 2% realization discount to the shocked 97.76 collateral value. Proceeds become 97.76 × 0.98 = 95.8048. Relative to principal only, the scenario gap becomes 4.1952. This discount is invented, not an observed transaction cost or an official required haircut.
Keep the market shock, realization discount and repayment basis in different columns. Applying a second 6% shock accidentally, or subtracting the original 4 excess again, would double count changes already reflected in collateral value. Recovery also depends on enforceability, timing and other available claims, which a valuation worksheet cannot resolve alone.
Trace counterparty and collateral structure
Items C.10.b–e ask about central clearing, counterparty identification where applicable, tri-party status, repo rate and maturity. These fields help direct further research; a tri-party indicator is not a statement that the custodian guarantees repayment.
Distinguish collateral issuer from repo counterparty. A Treasury collateral category does not transform the cash borrower into the U.S. Treasury. Conversely, a named private counterparty does not establish that all supporting collateral is its own debt.
The form permits certain collateral information to be aggregated by issuer. A category or aggregate collateral value does not necessarily identify every individual security, maturity or liquidity characteristic. Do not invent a granular holdings schedule from an aggregate line.
Keep each agreement and reporting date distinguishable. Adding collateral across unrelated counterparties and dividing by total loans can conceal one undercovered agreement. A netting assumption needs the actual supported legal scope; common branding alone is insufficient.
Avoid turning collateral into extra free assets
In the simplified initial case, the fund has a repo claim of 100 supported by collateral valued at 104. It does not follow that it owns 204 of freely usable investment assets. The claim and its support describe linked legal and economic roles.
Whether collateral can be sold, substituted, reused or realized after default depends on the actual arrangement and applicable rights. A field showing collateral value does not prove present unrestricted disposal. Avoid counting the same pledged security both as the repo’s support and as a separate unencumbered portfolio holding without an appropriate accounting and legal basis.
For a reverse repo, the fund’s cash-borrowing and posted-collateral direction changes the analysis. Do not add its liability and collateral as though it were another cash-lending investment with the same coverage formula. Resolve the role before aggregating exposures.
Execute the collateral-evidence register
| Check | Input to retain | Missing-data treatment |
|---|---|---|
| Vehicle | Fund series, filing and scope | No substitution for money market fund data |
| Direction | Cash lender or borrower | No formula before role is known |
| Amount | Principal, holding value and claim date | Unavailable basis stays unknown |
| Collateral | Face amount, value, issuer and category | Face amount not used as market value |
| Structure | Counterparty, clearing and tri-party fields | No guarantee inferred |
| Contract | Margin, substitution, netting and realization rights | No assumed immediate recovery |
| Scenario | Shock, costs and accrual assumptions | Conditional gap not called actual loss |
If sources disagree, compare reporting date, currency, principal-versus-value basis, collateral aggregation and amendments. A later collateral schedule can differ legitimately from an earlier filing. Do not average unlike dates to make the discrepancy disappear.
Update the table when an amendment, repayment, margin transfer, collateral substitution or supported contract term changes an input. Preserve the prior snapshot and recalculate the affected scenario. The deliverable should show the repo claim, its collateral support and the unresolved recovery conditions without manufacturing a cash guarantee or an observed loss.
Distinguish curing a gap from restoring the initial margin
After the 6% shock, adding 2.24 of eligible collateral value would bring 97.76 back to 100, assuming the principal-only claim is unchanged. Restoring the original collateral value of 104 would instead require 6.24. A statement that a margin call was met needs its actual target and terms; the two amounts do not describe the same requirement.
The break-even decline before principal-only collateral falls below 100 is 1 − 100 / 104, approximately 3.8462%. It is below 4% because the shock is measured on initial collateral, while the original excess was measured on the loan. These calculations still assume no price movement during collection and no costs.
FAQ
Does collateral of 104 against principal 100 mean 204 of free assets?
No. The repo claim and collateral support have linked roles and cannot simply be added.
Is the 2.24 shock shortfall a realized fund loss?
No. It is a conditional principal-only collateral gap before responses and recovery conditions.