Amex Exploration’s Self-Funding Claim: Does the Math Hold Up?

A junior gold miner in Quebec is making a claim that most investors would want to test before they applaud it. Amex Exploration says it can offset almost the entire capital bill for its Perron project, roughly CAD 194 million, before commercial production is ever formally declared.

That is a structurally unusual assertion. Canadian junior miners have long depended on dilutive equity raises and flow-through share structures to fund development, because they typically have no operating cash flow and limited appetite from lenders. A self-funding model that sidesteps most of that, built to work in a strong gold price environment, changes the calculus around dilution risk and capital efficiency.

The Amex Exploration financing strategy deserves scrutiny rather than celebration. After reading this, you will be able to judge whether the self-funding arithmetic holds together under realistic assumptions, which variables carry the most weight, and what specific milestones to watch before deciding the thesis is credible.

The logic of self-funding: why Amex is building the mine before building the mine

Amex is not treating its bulk sample as a geology exercise. It is treating it as the first act of project finance.

The company’s phased plan deliberately sequences revenue from a 40,000-tonne underground bulk sample, then pre-commercial Phase One production, ahead of any conventional financing event. The idea is to generate saleable gold ounces while the mine is still being built, then apply that cash against the capital bill.

The motivation is straightforward once you see the sequencing logic. Three factors drive it:

The broader gold price environment for junior miners in 2026 has not automatically translated into easier project financing, which is one reason Amex’s decision to build its revenue model around current spot levels rather than wait for conventional capital market access reflects a specific read on where the cycle sits.

• Dilution avoidance: management prefers not to issue equity when the share price already reflects a strong gold environment, since that hands value to new shareholders cheaply.
• A favourable gold price backdrop: high spot prices make pre-production ounces far more valuable as an offset than they would be in a weak market.
• Capital market optionality: by funding early stages internally, Amex keeps the timing of any future raise on its own terms rather than the market’s.

Where a gap remains, management has floated a fallback: pre-selling roughly 10,000 ounces in 2027 for approximately US$50 million, with delivery pushed out to around 2029. This is a forward sale, an agreement to deliver gold later at a price agreed now, not a streaming deal that would sell a permanent slice of future production. It is a bridge, and management frames it as one.

Management has gone further, characterising the high-grade Champagne Zone as rich enough to stand alone as a mining operation. That framing matters, because it reclassifies every operational milestone as a financial event worth tracking.

Where the project stands heading into bulk sample execution

The main governmental authorisations for the bulk sample were received on 30 March 2026. Final grade-control drilling results for the Champagne Zone were reported on 21 May 2026, and the decline had advanced more than 75 metres underground as of 21 September 2026. The mining contractor is CMAC, a firm with established Abitibi experience.

Processing is where the model still has an open question. Amex has signed a Letter of Intent (LOI) with Eldorado Gold’s Lamaque facility to mill both the bulk sample and Phase One material, and it has submitted a regulatory project notice designating Lamaque, making that destination a matter of public record.

The LOI has not yet converted to a definitive agreement. Amex has identified at least four to five alternative regional mills that need feed, so the strategy is not single-threaded, but the processing destination remains unconfirmed for now.

Running the numbers: how CAD 230 million in offsets is supposed to work

The offset claim rests on three components that build on each other. Take them in sequence.

First, the bulk sample itself. At 40,000 tonnes and an estimated yield of 20,000 to 28,000 ounces (management models 25,000 ounces), priced at US$4,000/oz and converted at a US$/CAD rate of 1.35, the bulk sample generates roughly US$100 million, or about CAD 135 million.

Second, the capex reclassification. The total bulk sample programme is projected to cost around CAD 60 million, but CAD 40 million of that spend qualifies directly as Phase One capital expenditure because the infrastructure serves both phases. That reclassified spend counts as an offset because it is money that would have been spent on Phase One anyway.

Third, pre-commercial Phase One production, modelled at a more cautious US$3,500/oz and estimated at US$68 million, roughly CAD 91 million at the same exchange rate.

Add the three components at face value and the total lands near CAD 266 million. Management’s headline figure is roughly CAD 230 million, and the difference reflects that the components are not all fully additive: some overlap, and management’s own presentation applies conservative haircuts rather than summing best-case figures. Either way, the projected offset exceeds the CAD 194 million Phase One target.

What matters is knowing where pre-commercial revenue counting stops.

When pre-commercial revenue counting ends Amex defines commercial production as sustaining output of 660 tonnes per day continuously for three months. Every ounce sold before that threshold counts toward the offset. Once the threshold is crossed, the offset window closes.

Here is the interpretive read you should take. The arithmetic is internally consistent, and the components are individually coherent. The question is not whether the model adds up. It is whether the gold price assumptions embedded in it, US$4,000/oz for the bulk sample and US$3,500/oz for Phase One, are defensible. Spot gold traded at US$4,282.98/oz on 25 September 2026, above management’s planning figure, which supports the model at today’s prices. No independent institutional forecast has been retrieved here to validate those assumptions into 2027 and beyond, so treat them as management estimates, not verified guidance.

The infrastructure decisions that make dual-phase allocation possible

The reclassification of CAD 40 million from expense to capex is only possible because Amex sized its infrastructure for Phase One from the outset.

Rather than running on diesel, the company is connecting to the grid at a one-time cost of about CAD 7.7 million, securing hydroelectric power at CAD 0.055 per kilowatt-hour. A water treatment plant is being built to Phase One capacity, not bulk sample capacity.

This is the structural move that converts a geological programme into a financial event. By front-loading infrastructure both phases require, Amex avoids redundant upgrades and gives itself a defensible basis for counting that spend against the Phase One bill.

What makes the Perron deposit suited to this model (and what makes it fragile)

Perron is a better candidate for this approach than most junior projects, and the reasons are geological before they are financial.

The Champagne Zone holds roughly 74,750 tonnes averaging just over 12 grams of gold per tonne, with free gold hosted in quartz and no deleterious elements, meaning no harmful impurities that complicate recovery. Management believes recovered grades could exceed feasibility estimates. Clean, high-grade feed is precisely what makes third-party mills willing to take material and what makes the revenue projections plausible.

The Eldorado Lamaque toll-milling relationship is the logistics enabler. Toll milling means paying an existing mill to process your ore rather than building your own. It sidesteps the two most common sources of cost overruns at this stage: mill sizing and tailings facility construction.

Favourable geology and smart logistics are necessary, but the industry record shows they are not sufficient. Three comparable cases make the pattern clear:

1. Osisko Mining (Windfall, Quebec): bulk samples and toll milling defrayed early costs, yet full mine development still required substantial conventional project financing.
2. Pretium Resources (Brucejack, BC): an early high-grade bulk sample over-represented average grades, and later operations revealed more complex grade behaviour.
3. Pure Gold Mining (Madsen, Ontario): a pre-commercial self-funding strategy ran into grade and operating problems that led to serious financial distress.

The Brucejack lesson Pretium’s bulk sample looked convincing, then the broader ore body behaved differently once at scale. Grade representativity in a sample is not proof of what the wider mining panels will deliver, and that gap is the single most material risk embedded in Amex’s revenue projections.

Industry experience points to three factors that separate the successes from the failures:

• Strict grade control: ensuring what gets mined matches what the model assumed.
• Close mill proximity: short haulage keeps costs and logistics manageable.
• Experienced underground operating teams: the people who keep development on schedule and on budget.

The read for investors is direct. Perron’s deposit characteristics are genuinely stronger than many peers attempting the same playbook, but the historical record insists that good geology is where this model starts, not where it is guaranteed to finish.

TRX Gold’s approach to non-dilutive expansion at Buckreef offers a useful reference point for how self-funding models behave when operational execution matches the initial thesis, and where the model breaks down when it does not.

The assumptions an investor needs to stress-test before accepting the self-funding thesis

The strategic logic and the deposit quality both hold up. What remains is a set of variables you can carry into your own due diligence rather than accept on management’s framing.

Start with gold price, the largest lever. On 25,000 ounces, the difference between US$3,000/oz and US$4,000/oz is a swing in the offset calculation that is material against a target of CAD 194 million, so a sustained pullback in gold does more damage than a headline glance suggests.

The gold price outlook beyond 2026 is the single variable that most determines whether Amex’s offset arithmetic holds, and the range of institutional forecasts for 2027 and 2028 is wide enough that stress-testing the US$3,500-4,000/oz planning range against independent frameworks is a necessary step before accepting the model.

Mill access is a binary dependency. The Eldorado Lamaque LOI has not converted to a definitive agreement, and any disruption to milling compresses or breaks the revenue timeline that the entire thesis depends on.

Grade reconciliation is the failure point the industry record keeps returning to. Bulk samples target the highest-confidence zones, so strong sample results do not guarantee equivalent economics across the full mine inventory.

The interpretive point is that these risks are independent of each other. An unfinished mill agreement, a gold price sitting at the high end of the modelled range, and a team making its first operational transition mean the thesis has at least three separate ways to come under pressure at once, and none of them is fully within management’s control.

What the balance sheet says about how much buffer exists

The financials show a company burning cash faster as it approaches operations. Amex reported a net loss of CAD 8,744,706 for the year ended 31 December 2025, up sharply from CAD 739,554 in 2024.

Liquidity as of 31 March 2025 stood at roughly CAD 10.2 million in cash and equivalents, up from CAD 3.5 million at the end of 2024. That is the baseline against which any working capital gap will be measured, and underground development costs land months before revenue does.

The forward sales fallback narrows the gap but introduces its own trade-off. Locking in a price and a delivery obligation on 10,000 ounces reduces exposure to a bridging shortfall, but it also removes upside optionality if gold keeps climbing and adds a firm delivery commitment the company must meet regardless of operational reality.

Amex’s forward sale mechanism sits within a wider evolution of modern mining financing structures that has seen streaming, royalties, and forward delivery agreements displace conventional equity raises as the preferred tools for developers who want to retain shareholder value through the construction phase.

A credible thesis, not a guaranteed one: what investors should do with this model

What Amex has built is genuinely differentiated within the Canadian junior peer group. The infrastructure sequencing, the toll-milling strategy, and the explicit capex offset architecture are not standard at this development stage, and the Champagne Zone’s grade and metallurgy give the approach a real foundation rather than a hopeful one.

The tension sits between the CAD 194 million capital target and the roughly CAD 230 million projected offset. That offset is a management projection resting on gold price assumptions at the optimistic end of institutional ranges, unverified in the context of this analysis, with the forward sale of 10,000 ounces for around US$50 million available as a defined but optionality-reducing fallback.

The right way to hold this model is as a framework that shifts financial risk away from equity markets and onto operational execution. Your task is to judge whether the team and the resource are up to carrying that risk, and to test each quarterly update against three triggers:

• Definitive mill agreement: has the Eldorado Lamaque LOI converted, or has an alternative been locked in?
• Grade reconciliation: do bulk sample results match the feasibility model as material is mined?
• Gold price tracking: where does spot sit relative to the US$3,500-4,000/oz planning range?

Watch those, and you evaluate Amex against milestones rather than headlines, which is a real advantage in a sector where the narrative often runs ahead of the fundamentals.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors, and the forward-looking figures discussed here are speculative and subject to change based on market developments and company performance.