What is a deterministic valuation under IFRS 13?

A deterministic valuation produces a single-point fair value from fixed, documented assumptions — one expected-cashflow path and one discount curve — rather than running probabilistic simulations such as Monte Carlo. Under IFRS 13 it is acceptable provided the inputs reflect market-participant assumptions and the result is an exit price.

Is a deterministic DCF compliant with IFRS 13 for private credit?

Yes. IFRS 13 does not mandate a stochastic model. A deterministic discounted-cashflow under the income approach is compliant when the discount rate embeds market-based credit and liquidity risk, the model is calibrated to the Day-1 transaction price, and the required Level 3 sensitivity disclosures are produced.

How is the discount rate set for a private credit instrument?

The rate is a risk-free benchmark (for example OIS / SOFR) plus a credit spread. Under the calibration approach the spread is solved at origination so the model value equals the transaction price, then carried forward and adjusted for documented changes in market spreads and borrower credit quality.

Deterministic private credit valuation under IFRS 13

A deterministic IFRS 13 mark prices private credit from one expected-cashflow path and one market-participant discount rate, calibrated to the Day-1 transaction price and updated only on documented evidence. It is fully compliant, reproducible, and — unlike a stochastic model — reconcilable to the basis point. This is how Verus builds it.

Conclusion first

Private credit is almost always a Level 3 asset: no active market, valuation driven by unobservable inputs. Under IFRS 13 the fair value is still an exit price built on market-participant assumptions — not the lender’s own view of worth, and not the contractual face value. A deterministic approach meets that standard by computing a single fair value from fixed, documented assumptions rather than a probability distribution. It is compliant, it is defensible, and it has one property a stochastic model cannot offer: the same inputs always produce the same mark.

1. The method: income approach, deterministic DCF

The fair value is the present value of expected future cashflows, discounted at a market-participant rate:

Fair value = Σ_t E[CF_t] / (1 + r_t)^t

where r_t is not a risk-free rate alone but a rate that embeds credit and liquidity risk a market participant would require at the measurement date. There are two deterministic ways to carry the credit risk:

Yield method (discount-rate adjustment). Discount the contractual cashflows at a credit-adjusted yield. This is the most common technique for performing credit.
Expected-cashflow method. Adjust the cashflows themselves for a fixed expected default and recovery (a single deterministic haircut, not a distribution), then discount at a rate adjusted only for market risk.

Verus runs the yield method by default and routes a facility to a recovery (EV-waterfall) or probability-weighted blend when an enterprise-value coverage test indicates the credit is impaired. The router is deterministic: given the deal, the assumptions and the coverage inputs, the same path is taken every time. The full router is set out in the methodology.

2. The Level 3 inputs you must fix

A deterministic model locks a specific value for each unobservable input. The material ones for private credit:

Credit spread / yield. A risk-free benchmark (OIS / SOFR / EURIBOR) plus a spread reflecting the borrower’s current risk.
Prepayment assumption. A single expected path — for example, refinanced at par in month 24 — based on contractual terms and incentives.
Default and loss given default. Where a credit event is anticipated, a fixed LGD applied to the affected cashflows.

3. The Day-1 calibration anchor

IFRS 13 emphasises calibration. On origination the transaction price (entry price) is generally taken to equal fair value (exit price). A deterministic model is built to honour that:

Solve at origination. Invert the DCF against the issuance price to recover the implied spread that makes present value equal the amount invested.
Carry forward. Hold that spread and adjust it only when there is documented evidence that market conditions or the borrower’s credit quality have changed — for example, a move in the relevant market spread benchmark, or a change in leverage.
Trace every change. Record the implied spread, the drift applied, the source of that drift, and the final spread used, stored with the run.

Verus implements exactly this: the spread is solved by bisection at origination and carried forward with a Verus spread-composite drift, keyed by rating and market environment, with every recalibration appended to an immutable history keyed to the deal.

4. Reflecting credit migration

In a deterministic framework, value changes come from altering the fixed inputs, not from re-rolling a simulation:

Improvement. Leverage falls from 5.0x to 3.5x → lower the spread input → fair value rises.
Distress. Performance deteriorates → adjust the expected cashflows down to the specific anticipated restructuring or recovery path.

5. The disclosure IFRS 13 requires

Because the marks rest on Level 3 inputs, IFRS 13 mandates a quantitative table of the significant unobservable inputs and a sensitivity analysis — how fair value would change if those inputs were altered. Example: “a 100 bps increase in the discount rate reduces fair value by £1.2m.” Verus reprices each facility at seven spread percentiles per run and produces the sensitivity grid as a standard output, so the disclosure reconciles to the engine rather than to a hand-built spreadsheet.

6. Why deterministic, and why it matters for audit

The standard does not ask whether a model is deterministic or stochastic; it asks whether the result is an exit price a market participant would agree at the measurement date. A deterministic model meets that test and adds three things an auditor values:

Reproducibility. Same inputs and same engine version reproduce the mark to the basis point. A historical valuation can be re-run from its recorded inputs at any time.
Explainability. Every figure on the mark is an input on the deal, a field on the assumption set, or a documented default. There are no learned weights and no silent overrides.
Calibration evidence. The Day-1 anchor and its drift history are stored, so the chain from transaction price to current mark is auditable end to end.

Where Verus fits

Verus is the deterministic engine Markst runs for private credit funds, banks, LPs and risk committees. It produces IFRS 13 / ASC 820 marks on demand — not quarterly batch — with full calibration lineage and a methodology whitepaper, available under NDA, behind the method. An SR 11-7 / SS1/23 validation pack is being assembled; its outcomes and back-testing sections begin with the first production book.

Methodology — the full equation set, approach router, scenario weights and governance surface.
Model risk — validation framework and challenger model.
Private credit valuation approaches — when a deterministic engine fits versus an advisory opinion or an operations-platform module.
Sample report — a Verus valuation output, screen by screen.

To value your own book under this method, book a demo.