‹ DFI screening tools

Debt-sizing & DSCR — the lender's lens

The screen and the report answer the equity question (what is the asset worth). A development-finance institution is usually a lender: its credit committee underwrites how much debt the cash flows can carry, and whether that debt still covers in a downturn. This tool sizes indicative debt capacity and stresses the debt-service coverage ratio (DSCR) across price scenarios.

A sizing indication, not a credit model. The cash-flow input is the returns feed's Derived annual gross operating margin (100% project basis) from a simplified model (production × (price − C1 cost)) — it is before tax, royalties, and sustaining capital, so a haircut is applied to approximate cash available for debt service; true free-cash-flow coverage is lower. Financing terms (rate, tenor, target DSCR, haircut) are illustrative assumptions you set, not a real facility. This tool indicates whether an asset's cash generation can plausibly carry debt and survive a price downturn; it does not structure, price, or approve a loan.

Two limits a credit officer must hold. (1) 100%-project basis, not borrower-attributable — the debt capacity shown is against the whole project's cash flow; a single sponsor's supportable debt scales by its ownership share (e.g. a ~40% sponsor supports ~40% of the figure). (2) Scope — these are the producing assets that carry a Derived margin; the development pipeline a DFI would actually finance has no margin scenario yet, so DSCR is Pending there (see the credit-memo generator). The magnitude of debt capacity moves with the haircut and terms; the downside-coverage verdict does not (it depends only on margin elasticity).

Cash available for debt service (spot)
gross margin × (1 − haircut)
Indicative debt capacity
sized to target DSCR at spot
Implied gearing

DSCR under price scenarios Derived

The down-case coverage result is independent of the rate, tenor and haircut you set — those cancel out. It follows from the asset's margin elasticity (how much of spot margin survives the downturn), so it is a property of the asset, not of the assumptions.

ScenarioCash for debt serviceDSCR

Coverage simulation Derived · Monte Carlo

Given debt sized to the target DSCR at spot, the price is drawn many times from a lognormal distribution with the base case as median, calibrated so the down case sits at the 10th percentile — i.e. the down scenario is treated as a plausible P10 outcome with real probability mass below it, not an impossible floor. Each draw maps through the asset's exact linear margin model to a DSCR. The result is the probability that coverage falls below 1.00× — a single-factor (price) screening probability, not a market-risk or default model.

P(DSCR < 1.00×)
screening probability — not a default rate
P10 DSCR
weak case
P50 DSCR
median
P90 DSCR
strong case

Red bars = draws below 1.00× coverage. The simulation is exact in the margin model (margin is linear in price); the assumptions are the price distribution (lognormal, median = base) and that the down case is a P10 outcome. This reconciles with the deterministic table above: the down-case DSCR there is roughly the P10 of this distribution. Single-factor — only price varies; capex overrun, schedule slip, ramp-up and grade are held fixed, and those construction-phase risks dominate real project-finance defaults. The breach probability is sensitive to the percentile choice: if the down case is nearer P25 than P10, it roughly doubles. A screening probability conditional on the model — not a probability of default, and true coverage is lower after tax/royalties/sustaining.

Cross-asset downside resilience

For each producing asset with a Derived margin: how much of spot operating margin survives the down-price case, and where a debt sized to 1.40× DSCR at spot would land in the downside (≥1.10× comfortable, 1.00–1.10× thin, <1.00× breached). Assets without a margin scenario are Pending.

AssetGross margin (spot ↓ down)Downside retentionDown-case DSCR @1.40×Capex base
Kansanshi
Zambia · Cu · operating
US$2,877m (↓US$2,259m)79% retained1.10×US$1,250m
Kamoa-Kakula
DRC · Cu · operating
US$2,723m (↓US$1,872m)69% retained0.96×US$3,040m
Sentinel
Zambia · Cu · operating
US$2,373m (↓US$1,864m)79% retained1.10×n/a — operating
Sukari
Egypt · Au · operating
US$1,060m (↓US$810m)76% retained1.07×n/a — operating
South Deep
South Africa · Au · operating
US$656m (↓US$511m)78% retained1.09×n/a — operating
Obuasi
Ghana · Au · operating
US$564m (↓US$431m)76% retained1.07×n/a — operating
Tenke Fungurume
DRC · Cu-Co · operating
margin scenario PendingPendingn/a — operating
Khoemacau Zone 5
Botswana · Cu-Ag · operating (ramp)
margin scenario PendingPendingn/a — operating
Ngualla REE
Tanzania · REE · development (pre-FID)
margin scenario PendingPendingUS$320m
Manono
DRC · Li · development (contested)
margin scenario PendingPendingUS$546m
Motheo
Botswana · Cu · operating
margin scenario PendingPendingUS$259m
Balama
Mozambique · graphite · operating
margin scenario PendingPendingUS$138m
Colluli
Eritrea · Potash (SOP) · development
margin scenario PendingPendingUS$302m

Margin returns.json (Derived, gross, 100% basis) · capex bankability.json · equity-economics companion: project economics estimator · method: how debt-sizing is built · v2.137.23
This is the lender's-lens analytical layer. It does not host a data room, model a specific facility's security/intercreditor terms, or anchor a live credit process — those sit outside a static analytical layer.