BENCHMARK SPREAD 7 METRICS v1.0 — 9 MAY 2026 FIELD-LEVEL PROVENANCE

Kabanga Nickel Project — Benchmark Spread

What this is. Public-source divergence on the seven highest-stakes screening metrics for the Kabanga Nickel Project: after-tax NPV, mine life, mineral reserves vs resource, all-in sustaining cost, pre-production CAPEX, after-tax IRR, and recovery rates. Each metric carries source attribution with dates, the structural source of divergence, and the most-defensible reading with reasoning visible.

What this is NOT. Not predictive. Not commissioned by Lifezone Metals or any party. Not a substitute for an Independent Engineer report or institutional credit memo work. The methodology is published at /methodology/benchmark-spreads/; the asset's full screening framework is at the parent Kabanga Pre-FID dossier.

Summary view — seven metrics

MetricOperator FS (July 2025)Operator IA (June 2025)Spread / driverMost-defensible readingConfidence
After-tax NPV @ 8%$1.58bn$2.37bn$0.79bn (50% of FS) — scope differenceFS $1.58bn for bankability; IA upside is real but not bankableHIGH
After-tax IRR23.3%22.9%0.4 ppFS 23.3% — converges across bothHIGH
Mine life18 years22 years4 years — reserve-vs-resource scopeFS 18-yr for screening; 22-yr extension dependent on infill drillingHIGH
Mineralisation tonnage52.2 Mt P+P reserves67.9 Mt M+I+Inferred15.7 Mt (23%) — categorisationFS reserves bankable; resource conversion 77% indicates mature engineeringHIGH
Pre-production CAPEX$942Mnot directly comparableFS scope is initial-development phase onlyFS $942M for screening; sustaining capex separate per platform DCF v1.0.38 frameworkMEDIUM
AISC ($/lb Ni net of Cu+Co credits)$3.36/lb (first quartile per CRU)not separately disclosedIndustry curve context: Indonesian laterite-HPAL benchmark $5-7/lbFS $3.36/lb anchors first-quartile positioning; structurally distinct from Indonesian cohortMEDIUM
Recovery rates87.3% Ni / 95.6% Cu / 89.6% Co (flotation)Same — IA used same metallurgical basisHydromet Technology upside potential — Q1 2026 FS targetedFS rates for screening; Hydromet upside as roadmap optionalityHIGH

Confidence tier convention. HIGH = direct primary-source operator disclosure with regulatory filing (Form 6-K, S-K 1300, NI 43-101) or matched cross-source convergence. MEDIUM = single-source or scope-limited operator disclosure where additional context would strengthen the reading. LOW = inference or where multiple defensible readings exist without single-anchor consensus. Same convention applied across all four case-study Benchmark Spreads.

Metric 1 — After-tax NPV at 8% real discount rate

The spread

SourceValueDateScope
Lifezone Form 6-K Feasibility Study TRS$1.58bn after-tax18 July 202518-year M&I-only reserves; initial development phase; $8.49/lb Ni reference price
Lifezone Initial Assessment TRS$2.37bn after-tax2 June 202522-year mine plan; M+I+Inferred resources; $8.49/lb Ni reference price; same recovery basis
Cycle-1.5 reference DCF (platform Python)$5.5bn target reconciling at $1.06/lb Ni cash cost4 May 2026Reference reconciliation harness; cycle-1.5 results
Cycle-1.5 platform DCF v1.0.38 at literal $0.48/lb cash cost$8.59bn (over-credits at literal first-5-years cash cost)9 May 2026Platform DCF with default v1.0.38 fixes; documented over-credit at literal-spec inputs

Spread analysis

The FS-vs-IA delta of $0.79bn (50% of FS NPV) is the headline divergence. The structural driver is scope, not disagreement: the FS is bankability-focused on the initial 18-year M+I-only reserves; the IA includes a broader 22-year scenario incorporating Inferred resources that are categorised as too speculative geologically to support reserves under SEC S-K 1300. Neither is wrong; they answer different questions.

The platform-DCF reconciliation layer adds a separate dimension: at literal-spec $0.48/lb cash cost (which is the first-5-years cash cost from the FS, used as if it applied LoM), both reference and platform DCF over-credit substantially. The reconciling cost where the reference DCF produces the FS-aligned $5.5bn proxy is $1.06/lb. The platform v1.0.38 with three structural fixes converges toward but does not fully match the reference, with platform reconciling cost of $1.28/lb. Cycle-2 work item is named.

Most-defensible reading

For DFI bankability screening: $1.58bn (FS). This is the auditable, M&I-only, sponsor-disclosed value supporting Final Investment Decision. The IA $2.37bn upside is real and represents the resource-extension trajectory under successful infill drilling, but is not bankable for project-finance debt sizing without conversion to reserves under S-K 1300. The platform-DCF reconciliation layer is informative for stress-testing the FS's underlying assumptions and surfaces that literal-spec first-5-years cash cost should not be used as LoM constant — but the FS's reconciling-cost framing produces an internally consistent $1.58bn central estimate.

What the spread tells the reader

The 50% FS-vs-IA delta is structural, not anomalous. It is the difference between bankable economics (FS) and full-resource-potential economics (IA). For DFI screening, the FS is the floor; the IA is the upside contingent on subsequent drilling and conversion. A senior reader should expect institutional credit memo work to anchor on FS numbers with sensitivity overlays for Inferred-conversion scenarios.

Metric 2 — After-tax IRR

The spread

SourceValueDateScope
Lifezone FS TRS23.3% after-tax18 July 2025$8.49/lb Ni reference; 18-year scope
Lifezone IA TRS22.9% after-tax2 June 2025$8.49/lb Ni reference; 22-year scope
Platform DCF v1.0.38 reconciling cost72.5% IRR at platform reconciling cost $1.28/lb9 May 2026Cycle-1.5 reconciliation; published as documentation of platform tool sensitivity
Platform DCF v1.0.38 literal $0.48/lb102.5% IRR (no longer implausible after v1.0.38 fixes; was 160.6% under v1.0.37)9 May 2026Literal-spec inputs documented as over-crediting at first-5-years cash cost

Spread analysis

FS and IA converge at ~23% IRR despite the NPV scope difference — IRR is a rate-of-return measure that is structurally less sensitive to mine-life extension than NPV. The 0.4 percentage point delta between FS (23.3%) and IA (22.9%) is methodologically minor and indicates the same project-economics underpinning across both.

The platform DCF reconciliation produces a higher IRR than the FS at the FS-aligned reconciling cost — this is documented in cycle-1.5 results as the residual platform-vs-reference convergence gap that cycle-2 will address.

Most-defensible reading

23.3% (FS). The IA-FS convergence at this dimension is informative — both operator analyses produce the same rate-of-return signal despite different mine-life scopes, which strengthens confidence in the underlying economic model. The platform DCF over-credit at literal-spec is documented; cycle-2 will narrow the residual gap.

Metric 3 — Mine life

The spread

SourceValueDateScope
Lifezone FS TRS18 years18 July 2025P+P reserves only; initial development phase
Lifezone IA TRS22 years2 June 2025M+I+Inferred resources; broader scenario
Operator stated long-term potentialBeyond 22 yearsJuly 2025 webcast and IR statementsResource estimate at broader project level supports stated long-term production extension potential

Spread analysis

The 4-year FS-vs-IA mine-life difference is the same scope question as the NPV spread. FS captures M&I-supported reserves; IA captures Inferred-included resources. Industry benchmark for Inferred-to-Reserves conversion in sulphide deposits ranges 40-70% depending on geological continuity and infill drilling success. The 4-year extension corresponds to ~15.7 Mt of additional M&I+Inferred resource not yet converted to reserves.

Most-defensible reading

For DFI screening: 18 years (FS reserves-supported). Project-finance debt sizing should anchor on M&I-only reserves. The 4-year extension is realistic but contingent on subsequent infill drilling — a credit committee should treat as upside-not-base. Operator's stated potential beyond 22 years (resource-extension scenarios) is further-removed and should be treated as longer-horizon optionality, not screening assumption.

Metric 4 — Mineralisation tonnage (reserves vs resources)

The spread

SourceTonnageGradeDateCategorisation
FS Proven + Probable Reserves52.2 Mt1.98% Ni / 0.27% Cu / 0.15% Co18 July 2025Reserves under S-K 1300 (Modifying Factors applied)
IA Measured + Indicated + Inferred Resources67.9 Mt1.93% Ni / 0.26% Cu / 0.14% Co2 June 2025Resources (Inferred is too speculative geologically to support Modifying Factors per S-K 1300)
Total contained metal (FS basis at LoM)902 kt Ni / 134 kt Cu / 69 kt Co18 July 2025Operator-published over 18-year LoM at 100% project basis

Spread analysis

The 15.7 Mt delta (23%) is the Inferred-to-Reserves gap. Reserves are 77% of the M+I+Inferred resource — at the high end of the 60-85% conversion range typical for sulphide deposits at FS stage. This reflects mature engineering on the asset (cumulative historical investment exceeding $435M and approximately 620 km of drilling per Lifezone disclosures), not aggressive scope.

Average grade is essentially identical across reserves (1.98%) and resources (1.93%) — the 0.05 pp difference is within typical reserve-resource grade variance and indicates consistent geological characterisation across the broader resource envelope.

Most-defensible reading

52.2 Mt P+P reserves at 1.98% Ni for screening. The reserve-resource conversion ratio of 77% is mature; the 15.7 Mt resource-but-not-reserves represents real upside contingent on infill drilling and successive Modifying Factor application. A DFI screening posture should treat reserves as the bankable basis and resource-extension as the upside scenario in stress-test sensitivities.

Metric 5 — Pre-production CAPEX

The spread

SourceValueDateScope
Lifezone FS TRS pre-production CAPEX$942M18 July 2025Initial development phase: 3.4 Mtpa underground mine + concentrator
Comparable-asset benchmark: $/tpa concentrator capacity$277/tpa ore for KabangaDerived from FSAfrican underground sulphide development reasonable range $200-400/tpa
Sustaining capex (platform DCF v1.0.38 default)$50M/yr default9 May 2026Default placeholder per platform DCF; user-overridable per asset

Spread analysis

The FS pre-production CAPEX of $942M is the operator-disclosed figure. Comparable-asset benchmarks for African underground sulphide developments at 3-5 Mtpa concentrator scale typically fall in the $200-400/tpa range — Kabanga at $277/tpa sits comfortably mid-range. The figure is internally consistent with the asset's scope.

Sustaining capex is structurally separate. The platform DCF v1.0.38 default of $50M/yr is a starting placeholder; institutional users should override per asset using operator-published sustaining-capex guidance where available. Lifezone has not disclosed a separate LoM sustaining capex figure as of the May 2026 cutoff; this is a known unknown for institutional credit memo work.

Most-defensible reading

$942M pre-production CAPEX (FS) for screening. The figure is consistent with comparable-asset benchmarks at $277/tpa ore. Sustaining capex remains operator-undisclosed at the LoM-aggregated level; institutional users should request specific guidance from Lifezone IR or apply industry benchmarks of 1.5-3.0% of pre-production CAPEX per year for sulphide concentrators.

Metric 6 — All-in sustaining cost (AISC)

The spread

SourceValueDateScope
Lifezone FS TRS AISC$3.36/lb Ni net of Cu+Co credits18 July 2025LoM average; net of byproduct credits at FS price assumptions
CRU 2025 nickel cost curve (Lifezone-attributed)First quartile of global nickel AISC industry-wide2025CRU Group nickel asset services cost model — comparison curve
Indonesian laterite-HPAL benchmark$5-7/lb (highly variable by asset and HPAL ramp-up status)Multiple sources 2024-2025Concentration: 64% of global nickel production from Indonesia per IEA Critical Minerals Outlook 2025
First-5-years cash cost (FS)$0.48/lb Ni18 July 2025First-5-years operating cost from start-up; not LoM constant

Spread analysis

Two distinct cost concepts must not be conflated. The AISC of $3.36/lb is the LoM average all-in sustaining cost net of Cu+Co byproduct credits — the industry-standard benchmark for cost positioning. The first-5-years cash cost of $0.48/lb is the early-LoM operating cost, materially lower because (a) higher early-mine grades, (b) byproduct credits highest at Cu+Co price-strength periods, (c) sustaining capex deferred to later years. Treating $0.48/lb as a LoM constant produces the over-credit documented in cycle-1.5 reconciliation; the platform DCF v1.0.38 form-clarification deferred to v1.0.39 addresses this user-input distinction.

First-quartile positioning per CRU's 2025 cost curve is structurally distinct from Indonesian laterite-HPAL economics. The Indonesian cohort represents 64% of global nickel production but operates on a different cost-curve regime (HPAL refining; tropical climate; political risk); Kabanga's African sulphide-flotation route produces different cost dynamics that comparable-cost analysts must treat separately.

Most-defensible reading

$3.36/lb LoM AISC for cost-curve positioning. $0.48/lb first-5-years cash cost as ramp-up indicator only. Institutional users running DCF stress-tests should apply LoM-average AISC trajectory (which moves from $0.48/lb early-LoM toward end-of-mine cash costs as grades decline and sustaining capex applies) rather than a LoM-constant cost. The first-quartile CRU positioning is the asset's structural advantage; Indonesian laterite comparable analysis requires separate cost-curve treatment.

Metric 7 — Recovery rates

The spread

SourceNi recoveryCu recoveryCo recoveryDate / scope
FS TRS metallurgical basis87.3%95.6%89.6%18 July 2025 — conventional froth flotation
IA TRS metallurgical basis87.3%95.7%89.6%2 June 2025 — same froth flotation basis
Hydromet Technology potential upsideQ1 2026 FS targetedGlencore partnership on US PGM recycling demonstrates Hydromet commercialisation pathway; Kabanga-specific FS not yet published

Spread analysis

FS and IA produce essentially identical recoveries (Cu rounding difference of 0.1 pp is methodologically negligible). This convergence indicates the metallurgical basis is well-characterised across scenarios and not a structural source of forecast risk.

The Hydromet Technology layer is the optionality. Lifezone's proprietary refining process targets feasibility study completion in Q1 2026 for Kabanga-specific application. If Hydromet delivers materially higher overall recoveries or eliminates intermediate concentrate logistics steps, the asset's economics could improve. The Glencore partnership on US PGM recycling demonstrates Hydromet commercialisation outside Kabanga; transferability of the technology to Kabanga's specific ore is the FS question.

Most-defensible reading

FS recoveries (87.3% Ni / 95.6% Cu / 89.6% Co) for screening; Hydromet upside as roadmap optionality. Institutional users should anchor base-case sensitivity on conventional flotation recoveries; treat Hydromet as conditional upside contingent on Q1 2026 FS publication. Hydromet downside (technology risk, ramp-up complexity, capital intensity) should also be considered if institution incorporates the upside.

How to use this Benchmark Spread

For an institutional capital allocator screening Kabanga for project-finance, equity, or off-take engagement, the spread provides three operational inputs:

  1. Anchor central estimates on FS values for bankability work. NPV $1.58bn; IRR 23.3%; mine life 18 years; reserves 52.2 Mt P+P; CAPEX $942M; AISC $3.36/lb; recoveries 87.3/95.6/89.6%. These are the auditable, M&I-supported, S-K 1300-compliant disclosures that DFI credit memo work should anchor on.
  2. Treat IA values as upside scenarios in stress-tests. NPV $2.37bn; mine life 22 years; resource 67.9 Mt M+I+Inferred. These are real but not bankable. Sensitivity overlays should incorporate them as optionality, not central estimates.
  3. Use the platform DCF v1.0.38 with cycle-1.5-disclosed limitations. The platform tool is a screening aid, not an institutional engineering substitute. The reconciling-cost framing produces internally-consistent NPV at $5.5bn target, but residual platform-vs-reference convergence gap of $0.22/lb-recovered is published; institutional users should validate against own reference DCF before committing analyst time.

What this Benchmark Spread does NOT do

Update triggers

This Benchmark Spread will be updated on three triggers per the published Benchmark Spread Methodology: (a) new primary disclosure from Lifezone (quarterly updates, FID-stage updated FS, Hydromet Q1 2026 FS); (b) substantive new third-party analysis with material divergence; (c) the platform's quarterly deep audit cohort (next Kabanga audit_due 2026-07-18 per the audit-overdue badge mechanism). Each update is recorded in the public Audit Log.