Reading a Mining Company's PEA — What Equity Analysts Miss

A Preliminary Economic Assessment is the first document where a junior miner translates geology into capital. It is also where most equity analysts stop reading too early. The IRR headline is the least informative number in the report — the discount rate, the contingency line, and the recovery assumption decide whether the project is real.

What a PEA is, exactly

A Preliminary Economic Assessment (PEA) is the first of three NI 43-101 economic study levels: PEA → Pre-Feasibility Study (PFS) → Feasibility Study (FS). It is permitted to include Inferred Resources in the production case, which is why a PEA can be published long before drilling is complete. That same allowance is also the main reason PEA economics are typically the most optimistic version of a project's life-of-mine model.

The standard PEA contains four blocks that compound on each other: a resource model (Measured + Indicated + Inferred), a mine plan (open-pit or underground, throughput, strip ratio), a processing flowsheet (recoveries by metal), and a financial model (CapEx, OpEx, sustaining capital, NPV, IRR, payback). The financial summary is what gets quoted in the press release. The geology and processing assumptions are what determine whether the financial summary is defensible.

The seven numbers analysts under-weight

Most coverage anchors on three numbers: after-tax NPV, after-tax IRR, and initial CapEx. Those three numbers are the surface of the report. The seven numbers below are where the project's real economics live.

1. The discount rate

Industry convention for a gold or copper PEA is a 5% real discount rate. That number is very low relative to the actual cost of capital for a pre-production junior — which commonly sits between 10% and 15% real, especially in higher-risk jurisdictions. A PEA NPV computed at 5% can be 40-60% higher than the same project at 8%. Always check the rate, and re-discount the NPV at 8% before any peer comparison.

2. The metal price deck

PEAs are required to disclose the metal-price assumptions used. Common practice is a three-year trailing average, but some issuers use forward strip or consensus broker price decks. A PEA assuming $2,400/oz gold against a $2,100/oz three-year average is materially optimistic. The standard sensitivity table at +/-10% to +/-20% price moves tells you how much of the NPV is leveraged to that assumption — if a 10% price drop wipes out 30% of NPV, the project is price-fragile.

3. The recovery rate

Metallurgical recovery is the percentage of the metal in the ore that ends up in the saleable concentrate or doré. For gold this is commonly 88-94% in open-pit heap-leach or CIL circuits. A PEA assuming 95%+ recovery without supporting metallurgical test-work flagged as "preliminary" is a yellow card. For complex sulphide or refractory ores, recoveries below 80% are realistic and require pressure-oxidation or bio-leach circuits — which add tens of millions to CapEx.

4. The contingency line

Initial CapEx in a PEA includes a contingency, typically 15-25% of direct costs. PEAs running at 10% contingency are under-provisioned. PEAs at 30%+ are usually a signal that the engineering team has high uncertainty. The real-world historical track record for mining-project CapEx overruns is around 25-30% on average from PEA to actual construction[^1] — so a 15% PEA contingency is mathematically below the historical base rate.

5. Sustaining capital

Initial CapEx gets the headline. Sustaining capital — the capital spent each year over the life of mine to replace equipment, expand pits, develop new underground levels — is what determines free cash flow. A PEA showing flat sustaining CapEx is unrealistic for any operation longer than 8 years. Underground mines typically run sustaining CapEx at 15-25% of revenue. Open-pit operations run lower, 8-15%. Check the LOM schedule, not just the year-1 number.

6. Strip ratio (open-pit) or dilution (underground)

Strip ratio is waste-to-ore tonnage. A PEA assuming 2.5:1 strip in year 1 and 4:1 in year 10 needs scrutiny on the year-10 number — late-life strip is where pit-shell optimizations tend to break down. For underground operations, the equivalent is planned dilution: the percentage of waste rock mined alongside the ore. Selective methods like long-hole stoping can run 10-15% dilution; bulk methods like sub-level caving can run 20%+. Higher dilution dilutes the head grade entering the mill.

7. Payback period in real years

Payback is usually quoted in after-tax discounted years from start of production. Read the timeline carefully: a PEA published in 2026 with "3-year payback" may assume permitting takes 2 years, construction takes 2.5 years, ramp-up takes 1 year — meaning the payback clock starts in 2031, not 2026. The CapEx is exposed for 5.5 years before the payback even begins. That is the actual capital-risk window.

Why PFS and FS economics tend to be worse

A PEA can include Inferred Resources. A Pre-Feasibility Study cannot — only Measured and Indicated Resources can be converted into Mineral Reserves at the PFS or FS stage. This single rule is responsible for a large fraction of the historical PEA-to-FS economic deterioration: projects that looked great on Inferred-heavy resource bases often shrink dramatically when only M+I tonnes are allowed in the reserve.

Published studies of PEA-to-PFS economic outcomes show, on average, that NPV declines by roughly 10-30% and CapEx rises by roughly 15-25% as projects progress from PEA to PFS. That is a base rate — well-engineered projects can beat it, but if a PEA is the only economic study available, an analyst should mentally apply a haircut before peer comparisons.

Three red flags

• Inferred-heavy resource base. If more than 40% of the production tonnage in the LOM plan is Inferred, the PFS is likely to lose a significant chunk of those tonnes. The mine life shortens, fixed-cost absorption rises, NPV falls.
• No metallurgical bulk sample yet. Recoveries quoted from variability test-work on small drill-core composites are not the same as bulk-sample recoveries. A PEA without bulk-sample data is a recovery-risk PEA.
• Single-pit base case. If the entire NPV depends on one optimized pit shell and no underground option, the project has no operational optionality. Permitting delays or pit-wall geotech issues become existential.

How Mineralis approaches PEAs

The MAS-Score methodology (currently 3 of 8 sub-scores live) reads each PEA, PFS, or FS as part of the document index. When v0.2 ships, the Stage sub-score will explicitly weight a project's economic-study level: PEA, PFS, FS, in-production. A junior with a single PEA on Inferred Resources scores below the same project at PFS on M+I Reserves. That gap is one of the structural differences between exploration risk and development risk — and one of the most common places where retail investors and even some institutional desks misprice junior miners.

The PEA is a serious document. Read it past the headline.

References

1. McKinsey & Company, "The mining industry's capital project performance" (2024) and earlier industry reviews report average mining-project CapEx overruns of 25-30% versus initial sanction. Order of magnitude consistent across studies by IPA Global, EY, and various NI 43-101 retrospectives.
2. CIM Definition Standards on Mineral Resources and Mineral Reserves (2014), referenced throughout NI 43-101 Companion Policy 43-101CP. Defines the M+I vs Inferred classification rules used in PEA vs PFS economic models.