Record Margins, Shrinking Ounces: Agnico's Cost Line Is Now the Whole Debate — Initiating at Outperform

Results vs. Consensus

Q1 2026 Scorecard

Year-Over-Year Comparisons

Operating Metrics, Year Over Year

The Q1 2025 cost figures above are the restated numbers Agnico published in this quarter's release under its revised cost composition, not the numbers that were reported at the time. On the old composition Q1 2025 total cash costs were $903 per ounce and AISC was $1,183. The restatement moves the base down by $8 per ounce on both measures and therefore makes the year-over-year increase look slightly larger, not smaller.

Quarter-Over-Quarter Comparisons

Quality of Beat

Revenue. The +66.1% year-over-year growth decomposes almost entirely into price. Gold sales volume fell 1.6% to 829,651 ounces; the realized price rose 68.1%. Multiply the two and you get essentially the whole revenue line, with by-product silver, zinc and copper contributing roughly $67M against $4,033M of gold. There is no acquisition contribution in the quarter and no accounting change. The revenue beat versus consensus is therefore best read as the Street under-marking the realized price rather than as an operational surprise, which is confirmed by the production line landing 4% short of the same analysts' volume estimate.

Margins. Gross margin of 66.4% expanded 1,443 basis points year over year and 477 basis points sequentially, and both moves are price-led. Production costs rose 24.5% against revenue up 66.1%, so the margin expansion is arithmetic rather than operational leverage. The two cost items worth isolating: amortization was essentially flat at $420.3M despite a heavier capital base, which flatters gross margin by roughly 660 basis points relative to a scenario where amortization had grown in line with revenue; and G&A of $77.9M ran at 31.8% of the full-year guidance midpoint in a quarter that is 25% of the year, an overrun management attributes to share-price-linked compensation. The share price fell through April, so that line should reverse.

EPS. Adjusted EPS of $3.41 basic exceeded IFRS EPS of $3.39 by two cents, and the reconciling items are small and clean: $9M of reclamation adjustments, $7M of derivative gains, $7M of asset disposal losses, $5M of foreign-currency translation on deferred tax liabilities, and $3M of other, all net of tax. That is an unusually tight adjustment bridge for a miner and stands in contrast to Q4 2025, where a $229M impairment reversal pushed IFRS income $172M above the adjusted figure. The effective tax rate of 33.8% is just below the 34% to 36% full-year guidance range, so the tax line is a modest tailwind to the EPS beat that should not repeat. Below the line, the 0.4% reduction in the basic share count added roughly a cent.

Production and costs. This is where the quarter was genuinely weak, and it is the half of the print that the headline beat obscures. Production of 825,109 ounces missed the compiled estimate by 34,317 ounces and fell 5.6% year over year, with declines at Macassa, Meadowbank and LaRonde outweighing a strong quarter at Detour Lake. Total cash costs of $1,093 per ounce came in $36 above the Street. Both cost measures finished above the full-year guidance midpoints. Management's framing that the quarter represents "about 24 percent of the midpoint of our annual guidance" is accurate on volume, but the equivalent statement on costs would be that Q1 ran 2.1% above the cost midpoint, and that statement was not made.

Segment Performance

Agnico reports revenue by mine and rolls it into six geographic regions. The concentration is the point of the business: Quebec, Ontario and Nunavut together produced $3,545.0M of the quarter's $4,099.6M, or 86.5% of revenue, all of it in Canada.

Revenue from Mine Operations by Region

Production and Unit Costs by Mine

Nine of ten mines saw unit costs rise, seven of ten produced fewer ounces, and the one mine with lower unit costs got there on by-product credits rather than operating improvement. The consolidated and Meadowbank cost figures are on the revised composition; the other eight mines are unaffected by the change and are directly comparable.

Abitibi Quebec — 32.3% of revenue, the volume anchor

Quebec produced 277,184 ounces at $998 per ounce, essentially flat on volume against 281,280 a year ago, with Canadian Malartic doing the heavy lifting. Malartic milled 4,707 thousand tonnes at 1.20 g/t and beat plan on grade because mining continued through mineralized zones near historical underground stopes in the Barnat pit. LaRonde is the offset: production fell 10.8% as East Mine delays were partially covered by lower-grade tonnes from LaRonde Zone 5 and Zone 11-3, and unit costs rose 37.9%, the third-worst deterioration in the portfolio. Goldex was the portfolio's only cost improver at $915 per ounce, helped by strong productivity at the Akasaba open pit.

The milestone in the region is underground. East Gouldie delivered first production via the ramp in March, three months ahead of plan, contributing roughly 27,400 ounces. The main ramp is at 1,151 metres and the shaft at 1,514 metres, with first shaft-hoisted production still targeted for Q2 2027.

Assessment: Quebec is doing what it is supposed to do, which is hold volume flat while the Odyssey transition is built underneath it. The LaRonde cost step-up is real but sits against a mine in sequencing transition with an automation program that is already mucking 26% of ore with automated loaders. A ten-day SAG mill shutdown in Q2 will make the second quarter look worse before it looks better.

Abitibi Ontario — 30.5% of revenue, the strongest and the weakest mine in one region

Detour Lake was the quarter's standout: 177,019 ounces, up 15.8%, at $974 per ounce, on a record 29.5 million tonnes of ore and waste from the open pit, record first-quarter mill runtime of approximately 92.5% and recovery of 90.9%. This is a mine hitting operating records while holding unit costs to a 3.0% increase in a year when the consolidated figure rose 22.1%.

Macassa is the opposite. Production fell 35.4% to 55,593 ounces and total cash costs rose 94.7% to $1,256, because the existing paste plant could not support the increased mining rate and extreme cold made it worse. The mill still set a quarterly throughput record of 149,000 tonnes, which is the framing management chose on the call.

"Now despite this progress, total mill tonnage was below plan this quarter, and this was mainly a function of challenges we faced with our old paste plant while commissioning the new one, which we expect to be fully operational in Q2."
— Natasha Vaz, EVP and COO, Ontario, Australia and Mexico

The fix is dated and specific: a new paste plant with roughly 3,600 tpd design capacity, on track for Q2, required to support the ramp to 2,040 tpd by year end. A permit amendment received in the quarter also allows AK deposit ore to be trucked to the LZ5 facility at LaRonde, with milling starting in Q2 and roughly 40,000 ounces of 2026 production forecast from AK.

Assessment: Ontario is where the second-half production weighting has to come from, and it is also where the execution risk sits. Detour needs no help. Macassa needs a paste plant to commission on schedule, an ore-haulage arrangement to work, and a mill ramp to 2,040 tpd by December, all in the same nine months. A doubling of unit costs at a mine this size costs the consolidated figure roughly $41 per ounce on its own, which is more than the entire distance between Q1's $1,093 and the guidance midpoint of $1,070.

Nunavut — 23.7% of revenue, grade-driven and weather-exposed

Nunavut produced 207,693 ounces at $1,117 per ounce, down 13.0% on volume. Meadowbank beat plan on grade through a mining-sequence change and higher-grade stockpile processing, but still produced 18.7% fewer ounces than a year ago at costs 28.0% higher. Meliadine lost open-pit tonnes to weather while setting a quarterly record of 4,114 metres of lateral underground development. Meadowbank has SAG and ball mill liner replacements scheduled in Q2 and Q4, and Meliadine a regular mill-maintenance shutdown in Q3.

The region is also where the next construction decision lands. Hope Bay engineering reached approximately 55% completion, the camp has a third wing going in, the mill building is standing and empty, and a decision is expected in May 2026 on a project management describes as over 400,000 ounces per year.

Assessment: Nunavut carries the highest unit costs of the three Canadian regions and the least weather predictability, and the caribou migration remains an annual production variable that management cannot schedule around. Against that, it is the region with the largest single growth option in the portfolio. The May construction decision is the nearest-dated catalyst in the story.

Australia — 4.4% of revenue, the best operational quarter nobody discussed

Fosterville milled 216 thousand tonnes at 6.31 g/t for 41,443 ounces, above plan on both throughput and grade, helped by positive grade reconciliation and additional development metres from the Phoenix zone. Unit costs still rose 38.1% to $1,123 on a 5.0% volume decline. A primary ventilation upgrade needed to sustain the mining rate in the Lower Phoenix zones is now commissioning in Q2.

Assessment: Fosterville is a high-grade, short-reserve-life asset that has been in managed decline from its peak, and a quarter that beats plan is a good outcome rather than an inflection. The ventilation upgrade is the gating item for the next several years of mining rate, and its slip into Q2 is worth tracking.

Finland — 6.1% of revenue, the operating problem the acquisition headline buried

Kittila produced 48,527 ounces at $1,313 per ounce, the highest unit cost of any mine outside Mexico and 29.7% above a year ago. Production came in below plan on lower mill throughput and recovery, with an external power outage in January and an unplanned three-day shutdown in March for a mechanical failure of the counter-current decantation tank mixer. Mill recovery remained at 81%, below plan, with improvement initiatives running on reagent dosage optimization, new autoclave instrumentation and heated-leach pilot testing. A 17-day mill and autoclave relining shutdown is scheduled for Q4.

Assessment: An 81% recovery rate at a refractory autoclave operation is a metallurgical problem, not a scheduling one, and the fix list reads like a mine still searching for the cause. Kittila also sits inside the Finnish tax change that has already caused at least one other operator to defer capital in the country. The Central Lapland consolidation is a genuinely interesting land package, but the mine that will process the first ounces from it is the one running below plan today.

Mexico — 3.0% of revenue, the tail asset

Pinos Altos produced 17,650 ounces at $2,311 per ounce on a by-product basis and $3,348 on a co-product basis, against a consolidated figure of $1,093. Production costs per ounce were $3,746. The asset is economic only because gold is at $4,861. By-product credits of $18.3M cut the reported figure by $1,037 per ounce, or 31% off the co-product cost.

Assessment: At 3.0% of revenue and more than double the company's unit cost, Pinos Altos is a portfolio question rather than an operating one. Management's only comment on the call was that the team "continues to work very hard on initiatives to safely extract the most value from their asset." That is the language of harvest, not investment, and the San Nicolás joint venture is the only Mexican item drawing capital.

Key Topics & Management Commentary

Overall Management Tone: Management was assured on the balance sheet and the growth pipeline and noticeably brief on mine-level cost deterioration, which went unaddressed in the prepared remarks despite being the quarter's weakest line. Safety was the exception to an otherwise confident call: the CEO opened by taking personal responsibility for two fatalities in five months, and the Q&A follow-up produced an acknowledgement that investigations remain open rather than a description of what has changed. Analyst pushback was almost entirely forward-looking, concentrated on the Finnish consolidation and the Hope Bay capital number, and no one asked why Macassa's unit costs nearly doubled.

1. The Margin Record Is a Price Record

The release is headlined on record operating margins and record adjusted net income, and both claims are correct. What the call did not do is separate the two sources of that record. Realized gold of $4,861 per ounce is 68.1% above a year ago. Total cash costs of $1,093 are 22.1% above a year ago. The gap between those two growth rates is the entire margin story, and it is a commodity outcome rather than a management one.

"We're pleased to announce a solid start to the year, with production slightly above budget and with costs in line with our guidance."
— Ammar Al-Joundi, President and CEO

Both halves of that sentence are defensible on management's own internal plan and neither is true against the external comparison. Production was above internal budget and 4% below the compiled analyst estimate. Costs were inside the guidance range and above its midpoint. Investors marking to the Street rather than to the budget got a different quarter than the one described.

Assessment: The record is real and the framing is generous. For an equity that trades on the gold price, a print where margin expansion is 100% price-driven and volume is negative should compress the multiple, not expand it, and that is roughly what the tape did.

2. Cost Guidance Reiterated, With the Second Half Doing the Work

Full-year total cash costs of $1,020 to $1,120 and AISC of $1,400 to $1,550 were both held. Q1 printed $1,093 and $1,483, inside both ranges but above both midpoints. The CFO's framing was that costs "continue to trend within our full year guidance ranges," which is accurate and also the weakest available true statement.

"Total cash costs were $1,093 per ounce, and all-in sustaining costs were $1,483 per ounce, reflecting higher royalty costs associated with a significantly higher realized gold price, lower production volumes as expected, and a stronger Canadian dollar compared to the first quarter of 2025."
— Jamie Porter, EVP, Finance and CFO

The arithmetic that matters: hitting the $1,070 midpoint on 3.4 million ounces with Q1 booked at $1,093 on 825,109 ounces requires roughly $1,063 per ounce across the remaining 2.57 million ounces. Hitting the $1,120 ceiling allows roughly $1,129. So the range accommodates a further deterioration from here, and the midpoint requires a $30 per ounce improvement. On AISC the equivalent numbers are approximately $1,472 to hit the midpoint and approximately $1,572 to hit the ceiling, which means AISC needs to run roughly flat with Q1 to land mid-range.

Assessment: Guidance is defensible, not comfortable. The midpoint is now a stretch and the range is intact. The mechanism management is relying on is volume: fixed costs spread over the second-half ounces. That works if the second half arrives.

3. The 48/52 Split Is a New Disclosure, and It Was Volunteered

Agnico does not usually publish a half-year production shape. This quarter it did, guiding approximately 48% of full-year production to the first half and 52% to the second. At the 3.4 million ounce midpoint that implies roughly 1,632 thousand ounces in H1, of which Q1 delivered 825.1 thousand, leaving roughly 807 thousand ounces for Q2 and roughly 884 thousand per quarter in H2, about 7% above the Q1 run rate.

"We don't typically give 48 percent, 52 percent. We decided we wanted to do it because we just wanted people to know that, actually, everything's going quite well."
— Ammar Al-Joundi, President and CEO

Assessment: Volunteering a shape you do not normally give, immediately after a quarter that missed the Street on volume, is a confidence signal and a commitment in the same gesture. It converts a full-year target into two checkable half-year targets, and it means the Q2 print now has a number attached to it before the quarter has closed. We read it as net positive and as a raised bar.

4. Free Cash Flow of $732M Understates the Quarter by More Than Half

The reported free cash flow figure absorbed a $1.3B payment settling the remaining 2025 tax liability. Total cash taxes paid in the quarter were $1.8B, roughly half of the $3.4B to $3.6B guided for the full year. Free cash flow before working-capital changes was $1,618M, and cash from operations before working capital was $2,231M against $1,209M a year ago. Income taxes payable on the balance sheet fell from $1,226.3M to $250.2M, which is where the payment shows up.

Remaining 2026 cash taxes are guided to quarterly instalments of $525M to $575M. That is a step-down of roughly $1.25B per quarter from the Q1 outflow, and it is the single largest identifiable driver of sequential free cash flow improvement in the model.

Assessment: Anyone comparing Q1's $732M against Q4 2025's $1,310M and concluding that cash generation deteriorated has read the wrong line. On a pre-working-capital basis Q1 free cash flow rose 113% year over year. The distortion reverses mechanically from Q2.

5. Balance Sheet: A‑ Rating, $2.9B Net Cash, and an Explicitly Imprecise Target

Cash rose $246M to $3,111.9M against total debt of $196.5M, taking net cash to $2,915M from $2,670M. Fitch upgraded the long-term issuer default rating from BBB+ to A‑ on April 29, the day before the print. The revolver remains undrawn with roughly $2B available and an uncommitted $1B accordion. Total equity rose to $26,275.1M from $24,742.5M.

"In terms of minimum cash balance, I mean, I think we're comfortable where we're at now with $3.1 billion of cash, $2.9 billion of net cash, between $3 billion and $5 billion."
— Jamie Porter, EVP, Finance and CFO

"The exact position of cash on the balance sheet is as much an art as it is a science."
— Ammar Al-Joundi, President and CEO

Assessment: A $3B to $5B target range on a company generating roughly $5.6B of annual free cash flow is loose enough to justify almost any capital-allocation decision for two years. That is a feature while the pipeline is being funded and a governance question once it is built. The A‑ rating on $197M of debt is close to irrelevant to the cost of capital and highly relevant to the ability to move fast on an acquisition.

6. The Buyback Halved, and the Ceiling Doubled

Agnico repurchased 721,211 shares at an average price of $207.68 for $150M in the quarter, half the Q4 2025 pace, alongside $203.2M of dividends paid on the cash-flow statement and a declared $0.45 quarterly dividend payable June 15. Total shareholder returns were $375M on the company's basis, which pairs the buyback with the declared rather than the paid dividend, which management characterised as closer to 50% of the quarter's depressed free cash flow against a 40% full-year target. The internal NCIB purchase limit is being raised from $1B to $2B at the May renewal.

"It was $150 million of shares repurchased in the quarter, which was half of what we did in Q4. That will ramp up, certainly, in Q2 and through the rest of the year as our free cash flow is expected to be higher."
— Jamie Porter, EVP, Finance and CFO

Two observations the call did not make. First, the average repurchase price of $207.68 is 13.1% above the $183.56 close on the day after the print, so the quarter's buying was done into a higher tape. Second, the $2B ceiling is not the binding constraint on repurchase volume. On our full-year free cash flow estimate of roughly $5.6B, a 40% payout is about $2.24B, of which the dividend consumes roughly $0.9B, leaving roughly $1.34B of buyback capacity against a $2B ceiling. Raising the ceiling matters only if the payout ratio itself moves higher.

Assessment: The stated forward commitment is unambiguous and dated, which makes it gradeable next quarter. The lever that would make the doubled ceiling meaningful is the portfolio-investment sale management flagged as a way to offset Rupert dilution, against a $1,814.1M investment balance that grew $305.9M in the quarter.

7. Finland: A 2,492 km² Consolidation Paid for in Stock

On April 20 Agnico announced three simultaneous transactions in the Central Lapland Greenstone Belt: proposed acquisitions of Rupert Resources and Aurion Resources, and the purchase of B2Gold's 70% interest in Fingold Ventures, which closed in April. Rupert and Aurion are expected to close late in Q2. The stated ambition is a roughly 500,000 ounce per year Finnish platform within the decade.

"By these transactions, we are aiming to deliver in Finland a platform for multiple mines over multiple decades, similar to the three regions in Canada, that are Quebec, Ontario, and Nunavut, where we will be leveraging our regional expertise."
— Guy Gosselin, EVP, Exploration

The consideration structure is the part worth pausing on. The release commits to evaluating ways to "reduce dilution associated with the Rupert transaction," and management confirmed in Q&A that the all-share structure was the counterparty's requirement rather than Agnico's preference. Timing for a revised technical view is end-2027.

Assessment: Strategically coherent and consistent with how Agnico has built every other regional platform it owns. The cost is a share issuance of undisclosed size at a moment when the stock is 27.2% off its 52-week closing high, and a first-look technical update that is roughly twenty months away. This is a decade-horizon transaction being funded with equity at a cyclical low in the equity, which is the least attractive version of a good idea.

8. Hope Bay: A Construction Decision Three Weeks Out, With a Range Rather Than a Number

Detailed engineering reached approximately 55% completion, the camp is being extended with a third wing, sealift planning is underway, and a redevelopment decision is expected in May. The project is framed at 400,000 to 425,000 ounces of incremental annual production from 2030. On capital cost, management would only bracket it.

"We see a bit of inflation, but it's going to be below $2.5 billion for sure." ... "It's going to be slightly over $2 billion."
— Dominique Girard, EVP and COO, Nunavut, Quebec and Europe

"And the changes are not so much inflation. They're more … instead of ramping up the mill in two stages just because of the economics, you just do it all at once."
— Ammar Al-Joundi, President and CEO

The scope changes named on the call are a mill designed for 6,000 tpd from the start rather than in two stages, fast-tracking Patch 7 to pull ounces forward, and an additional camp wing to sustain exploration drilling. Those are deliberate choices with returns attached, and they are also the reason the number moved.

Assessment: Being told three weeks before sanction that capital will be "slightly over $2 billion" and "below $2.5 billion for sure" is a $500M range on a project the company has been engineering for years. Management's attribution of the increase to scope rather than inflation is the right answer if it holds. The May release is the first hard test of the pipeline's capital discipline, and it is the nearest catalyst in the story.

9. Safety: Two Fatalities in Five Months

The CEO addressed this before handing to the CFO, which is the correct order and was not obligatory.

"Tragically, we've had two fatalities over the past five months. This is not acceptable. I recognize and I accept that the responsibility for the safety of our people rests, ultimately, with myself and with my team."
— Ammar Al-Joundi, President and CEO

The incidents occurred at Fosterville in December 2025 and Canadian Malartic in April 2026. The company mandated a global stand-down across every site. Asked directly in Q&A what had changed in procedures, the answer was that investigations remain open and that the emphasis is on major-hazard identification and critical controls.

Assessment: The tone was right and the disclosure was thin, appropriately so while regulatory investigations are live. For a company whose entire cost advantage rests on stable long-tenured workforces in a small number of jurisdictions, two fatalities in five months is a governance and operating risk rather than only a human one, and it is the risk least visible in the financial statements.

10. The Cost Composition Changed on January 1, and the Comparisons Broke With It

For periods from January 1, 2026, Agnico revised how total cash costs and AISC treat the NTI Payment at Meadowbank. The company restated Q1 2025 on the new basis ($895 and $1,175, against $903 and $1,183 as originally reported) and restated full-year 2025 ($953 and $1,313 against $979 and $1,339). It did not restate Q4 2025, which remains published at $1,089 and $1,517.

The practical consequence is that the year-over-year comparison in this release is clean and the sequential one is not. Reading Q1 2026 AISC of $1,483 against Q4 2025's $1,517 suggests a $34 per ounce improvement that does not exist on a like-for-like basis.

Assessment: The restatement is disclosed clearly and repeatedly in the release, and the direction of the revision makes the year-over-year increase look larger rather than smaller, which is not the direction a company massaging its optics would choose. It is nonetheless a trap for any model carrying a sequential unit-cost series across the December 2025 boundary.

11. Input Hedging: Detailed on Diesel and Currency, Silent on Gold

The release quantifies diesel exposure with unusual precision. Diesel is roughly 10% of operating costs, split about 7% direct consumption and 3% transport. Approximately 54% of 2026 diesel exposure is hedged at an average benchmark of $0.71 per litre against a $0.78 guidance assumption. A 10% move in diesel prices is worth approximately $4 per ounce on direct consumption after hedges and roughly $2 per ounce on transport. Approximately 42% of remaining Canadian dollar exposure is hedged with a floor around 1.38 and participation to about 1.42, against a 1.36 guidance assumption.

"Including, for example, in both Ontario and Quebec, where we produce the majority of our gold, all of our electricity is either hydro or nuclear and, really, not exposed to changes in fuel and diesel prices."
— Ammar Al-Joundi, President and CEO

Assessment: This is a genuine and underrated structural advantage, and the disclosure is better than the sector norm. It is also aimed at the smallest of the three cost drivers management itself identified. The February guidance attributed roughly 60% of the year's cost inflation to royalties and the Canadian dollar and 40% to about 4% general inflation plus mine sequence, and diesel sits inside that 40%.

12. Capital Spending Steps Up Roughly 30% This Year

Q1 capital expenditures were $488.9M excluding capitalized exploration and $573.6M including it, against a full-year guidance midpoint of $2,595M. That is 22.1% of the year in the first quarter, so the spend is back-weighted alongside production. Development capital of $292.3M ran 1.49 times sustaining capital of $196.6M, with the largest development items at Canadian Malartic ($85.1M), Detour Lake ($73.4M plus $4.3M underground), Hope Bay ($31.8M) and Macassa ($24.5M).

"The five key value-driver projects that we're advancing are 30 percent to 60 percent IRR projects in the current gold price environment." ... "I mean, our capital spending has increased from $2.3 billion last year to probably $3 billion this year, all in."
— Jamie Porter, EVP, Finance and CFO

Assessment: A 30% to 60% IRR claim is a gold-price statement as much as a project statement, and management said so. The more durable point is that development capital exceeds sustaining capital by half, which is what a growth pipeline actually looks like on a cash flow statement, and it is being funded without touching the revolver.

Guidance & Outlook

No line of 2026 guidance moved. The only change to the outlook is the newly disclosed half-year production shape.

Implied quarter-over-quarter ramp. At the 3,400 thousand ounce midpoint, the 48% first-half weighting puts H1 at roughly 1,632 thousand ounces. Q1 delivered 825.1, leaving roughly 807 thousand ounces for Q2, marginally below Q1. The second half then needs roughly 1,768 thousand ounces, or about 884 thousand per quarter, some 7% above the Q1 run rate. The identified drivers of that step-up are the Macassa paste plant commissioning in Q2, roughly 40,000 ounces of 2026 production from the AK deposit starting to mill in Q2, the Kittila recovery-improvement programme, and normal sequencing. Working against them are scheduled shutdowns in every quarter of the year, including a ten-day LaRonde SAG mill outage and a five-day Meadowbank liner replacement in Q2, a five-day Macassa mill outage and a Meliadine outage in Q3, and a 17-day Kittila autoclave relining in Q4.

The cost bridge. Reaching the $1,070 total cash cost midpoint requires roughly $1,063 per ounce over the remaining 2.57 million ounces, about $30 below Q1. Reaching the $1,120 ceiling allows roughly $1,129, or $36 above Q1. On AISC, the midpoint requires roughly $1,472 and the ceiling allows roughly $1,572. In short, the AISC guide has meaningful room and the cash-cost midpoint does not. Because guidance assumes $4,500 per ounce gold for royalty and NTI purposes and Q1 realized $4,861, a gold price that stays above the assumption keeps royalty costs above plan, which is the mechanism by which a good revenue quarter produces a bad cost quarter.

Street position. Compiled estimates going into the print sat at 859,426 ounces of production and $1,057 per ounce of total cash costs, so the Street was modelling more ounces at lower unit costs than Agnico delivered. Nothing in the reiterated guidance forces those full-year estimates down, but the Q1 shortfall has to be absorbed somewhere, and the 48/52 disclosure removes the option of assuming it away across four even quarters.

Guidance style. Agnico guides annually rather than quarterly and has a long record of landing inside its ranges, including 2025, when production of 3,447,367 ounces finished above the midpoint while total cash costs of $979 and AISC of $1,339 both came in slightly above the top of that year's guidance on higher royalties. That precedent is directly relevant: the last time gold ran well above the company's price assumption, costs breached the guided range while production did not. The 2026 assumption is $4,500 and Q1 realized $4,861.

Analyst Q&A Highlights

What the Finnish Consolidation Delivers Over the Next 12 to 24 Months

The opening question of the session went straight to the value-creation timeline on the Central Lapland transactions, which had been announced eleven days earlier and had not been discussed on a call. The answer was that the first deliverable is a revised technical view at the end of 2027, and that drilling starts as soon as the acquisitions close because the property boundaries were themselves constraining where holes could be collared. No resource update, study milestone or permitting start date was attached to the intervening period.

Q: "But can you help us understand what are the sort of value creation steps over the next 12 to 24 months to get to a better understanding of what that ultimately looks like? So, resource update, study, when do you expect permitting to start, and just any other considerations along that sort of timeline thinking."
— Lawson Winder, Bank of America Securities

A: "As we discussed in the press release, our plan, by removing the property boundaries, is to have sort of an updated view with the current information by the end of 2027, where we're going to have sort of a better picture of what the optionality that removing the property boundary offers on both the optimum pit design and location of infrastructure."
— Guy Gosselin, EVP, Exploration

Assessment: The question asked for a 12 to 24 month roadmap and got a 20-month wait for a first look. That is honest and it is also the correct read on how long this optionality takes to become quantifiable. Investors buying AEM for Finland today are buying a 2030s asset with no interim marks.

Whether Detour Lake Underground Is Turning Into a Larger System

A follow-up pressed on whether the drill results west of the Detour pit have changed the internal view of what the underground could be. The exploration answer was deliberately deflationary: the headline intercepts sit within an area expected to average 2.5 to 3.5 grams per tonne, in line with the existing model. The operating answer was more interesting, because it introduced scope changes that are not in any published plan.

Q: "With the drilling you've done to date, has your understanding of what the Detour Lake underground can be? Has it evolved and changed in any way in the past, let's say, 12 months? I mean, is it starting to look like it could be a much bigger system?"
— Lawson Winder, Bank of America Securities

A: "But with the combination of the exploration success and the high gold price environment, there is definitely optionality here at Detour. So it's fairly early stages, I would say, but the team is looking at different trade-offs for, potentially, a higher milling capacity, a larger underground, potentially another pushback at the pit."
— Natasha Vaz, EVP and COO, Ontario, Australia and Mexico

Assessment: Three scope expansions were named in one sentence: more mill, more underground, another pit pushback. Each is capital. The exploration team talked the grades down and the operating team talked the scope up, which is the correct division of labour and also a signal that the Detour capital number in circulation is a floor rather than a plan.

Why the Buyback Slowed and Where the Cash Floor Sits

The most useful exchange of the call on capital allocation. The buyback ran at half the prior quarter's pace, and the question established that the cause was the tax payment rather than a change in policy, then extracted an explicit minimum-cash range and a forward commitment on repurchase pace.

Q: "I noticed the buyback pace slowed down quarter over quarter in the first quarter. Is that just a function of the pretty significant cash tax payment? And can you just remind us of the minimum cash you'd like to keep on the balance sheet going forward?"
— Fahad Tariq, Jefferies

A: "Our free cash flow was lower in Q1 as a result of the cash tax payment. So I think the percentage worked out to closer to 50 percent. It was $150 million of shares repurchased in the quarter, which was half of what we did in Q4. That will ramp up, certainly, in Q2 and through the rest of the year as our free cash flow is expected to be higher."
— Jamie Porter, EVP, Finance and CFO

Assessment: A dated, checkable commitment, which is what a scorecard needs. The payout ratio actually rose to roughly 50% of a depressed free cash flow base, so the policy is intact and the denominator was the problem. We will grade the Q2 repurchase figure against roughly $400M per quarter, which is what a 40% full-year payout implies for the balance of the year.

How Input Inflation Is Being Priced Into the Hope Bay Economics

Asked how current input prices are being carried into a project study due within weeks, management answered on process rather than price: engineering completion and existing infrastructure are the cost controls, and the study will publish long-term assumptions with sensitivities rather than spot-based economics.

Q: "I had a question in terms of the Hope Bay project update later this month. In looking at the current market for prices, there's obviously been a high degree of inflation. How are you thinking about incorporating some of those input prices for that project update and thinking about maybe the near-term impact versus what otherwise would a long-term, much more reasonable price be?"
— Josh Wolfson, RBC Capital Markets

A: "In my experience, the most important thing in building projects is to get the engineering done and to have the execution plan well laid out. We have seen some inflationary pressure but, actually, it hasn't been that bad."
— Ammar Al-Joundi, President and CEO

Assessment: "It hasn't been that bad" is a claim the May release will settle in one number. The substantive point in the answer is that Hope Bay is a brownfield restart with camp, backup power, half a mill building and water treatment already standing, which is a genuinely different risk profile from a greenfield build and is the strongest argument for the pipeline generally.

Why the Rupert Consideration Was All Stock

The sharpest question of the call, because it went to a decision that dilutes existing holders at a de-rated share price. The answer was unusually direct: Agnico wanted to pay cash and the seller refused.

Q: "First question is, was there a reason for using Agnico shares rather than cash specifically? And then, I guess, the second part. You alluded to exceeding the 40 percent cash flow payout, potentially, in order to reduce the share count following the acquisition. Can you give us a sense of kind of quantum?"
— Daniel Major, UBS

A: "With regard—it's a very good question on why we used shares instead of cash. And the answer is we wanted to use cash, and they wanted 100 percent shares. I think their view—and rightfully so—is, Agnico shares are good shares to have, and they wanted 100 percent shares. We used full cash on the other deals."
— Ammar Al-Joundi, President and CEO

Assessment: Credible and unflattering to the outcome. A counterparty insisting on stock is a counterparty that thinks the stock is going up, which is a compliment to the asset and a cost to the buyer. The follow-up on quantum did not produce a share count, and the offsetting mechanism is a conditional portfolio-investment sale rather than a committed buyback. This is the one place in the quarter where a real economic cost was disclosed without being sized.

Ranking Reinvestment Against Shareholder Returns as Cash Builds

A recurring line of questioning across the session concerned what happens to capital allocation as the cash balance grows. The answer declined to re-rank the priorities and instead argued the question is not binding at current gold prices.

Q: "As you think about the cash balance increasing, where do your priorities lie in terms of capital? Just rank them again in terms of capital return to shareholders. And, I mean, the balance sheet's pretty strong at this point, and you're accumulating a lot of cash. So where does reinvestment into the business now fall into the—has it moved up over the capital return to shareholders?"
— Anita Soni, CIBC World Markets

A: "Beyond that, I'd say right now, in this gold price environment, we're fortunate in that we can do it all. We can afford to reinvest aggressively in the business. We can afford to deliver very strong returns to shareholders."
— Jamie Porter, EVP, Finance and CFO

Assessment: "We can do it all" is true at $4,861 gold and untested below it. The answer that would have carried more information is which of the five projects gets deferred first at $3,500 gold, and that question was not asked. The stated hierarchy is unchanged: reinvestment first, 40% payout as a floor, balance sheet strengthening throughout.

Whether the Hope Bay Capital Number Has Moved

The question referenced sector-wide capital escalation and asked whether the previously discussed $2B figure still holds three weeks before sanction. The answer moved the number and attributed the move to scope.

Q: "I just wanted to ask, because we've seen capital increases lately at some of the projects out there, should we still be thinking that $2 billion for Hope Bay would be a reasonable capital for the 10-year mine plan that you've been talking about?"
— Tanya Jakusconek, Scotiabank

A: "We see a bit of inflation, but it's going to be below $2.5 billion for sure. And the thing we need to finalize—we have a very good, now, let's say, level of engineering. But there's a decision that we're taking, for example, to fast-track Patch 7 and to do more ounces earlier."
— Dominique Girard, EVP and COO, Nunavut, Quebec and Europe

Assessment: The number went from "$2 billion" to "slightly over $2 billion" and "below $2.5 billion for sure" in the space of one answer, with the CEO stepping in to attribute the change to a single-stage mill rather than to inflation. The scope explanation is plausible and specific. It is also the case that a company confident in its number three weeks before a board decision would have given the number.

What Has Changed in Safety Procedures After Two Fatalities

The final exchange of the call, and the one management least wanted to be brief on. The question was asked twice, the second time to establish whether the answer really was that nothing procedural had changed.

Q: "So my question to you is, from your tragic incidents that you've had at your mine sites, what have you learned? And what changes to procedures and processes have you put in place from these learnings?"
— Tanya Jakusconek, Scotiabank

A: "In both of these instances, the in-depth investigations are still ongoing. The authorities are involved, and the regulatory authorities and so on are involved in these investigations as well. So we can't share the results of these investigations yet but, certainly, there are learnings around this."
— Carol Plummer, EVP, Sustainability, People and Culture

Assessment: The honest answer to a question with no good answer while investigations are live. What was disclosed is a global stand-down, plus a major-hazard identification programme and critical controls that existed before. The absence of a specific procedural change is the finding, and it is not resolvable until the investigations close.

What They're NOT Saying

1. Gold-price sensitivity on the cost line. The release quantifies diesel sensitivity twice, gives the hedged percentage and benchmark price and lays out the currency hedge structure, and management added the litres-per-ounce consumption figure on the call. It contains no sensitivity for what a change in the gold price does to royalty and NTI costs, even though guidance is built on a $4,500 per ounce assumption, Q1 realized $4,861, and the company's own February bridge attributed roughly 60% of 2026 cost inflation to royalties and the Canadian dollar. The largest cost driver is the only one without a published sensitivity.
2. The Rupert share count. The release refers twice to reducing "dilution associated with the Rupert transaction" without stating how many shares are being issued or what percentage of the company that represents. The balance sheet gives shares outstanding at March 31 and the transaction had already been announced on April 20. A reader cannot compute pro-forma per-share metrics from anything Agnico published this quarter.
3. Macassa's unit costs. Macassa appears five times in the prepared remarks: record mill throughput, an LTE network installation, an ore-haulage permit approval, and twice in lists of mine-sequencing effects. Production fell 35.4% and total cash costs rose 94.7% to $1,256 per ounce, a deterioration worth roughly $41 per ounce at the consolidated level. Neither management nor any analyst addressed the unit cost directly.
4. Kittila's 81% recovery. The written release discloses that mill recovery "remained lower than planned at 81%" and lists three separate improvement initiatives still in progress. Kittila is the highest-cost mine outside Mexico at $1,313 per ounce. It was mentioned seven times on the call, entirely in the context of the Finnish acquisition, the Finnish tax change, and a 20-year-old exploration analogy. The recovery problem was not raised.
5. Pinos Altos's future in the portfolio. One generic sentence of call commentary for an asset running at $2,311 per ounce against a $1,093 company average, more than double the group cost, on 2.1% of production. No comment on mine life, no comment on whether it is a sale candidate, no comment on what happens to it below $3,000 gold.
6. A Q2 number. The 48/52 half-year split is a genuinely useful disclosure and it is also a way of implying a second-quarter figure without stating one. Pressed on the quarterly cadence in Q&A, the answer was that it is "more just mine sequencing" with no volume figure attached. Investors must derive roughly 807 thousand ounces themselves, and management retains the ability to say it never guided to it.

Market Reaction

• Pre-print setup: AEM closed at $188.21 on April 30, the day of the release. The stock entered the print up 11.0% year to date and up 60.1% over twelve months, but down 7.3% over the prior thirty days and 25.4% below its 52-week closing high of $252.19. The 52-week closing range was $104.78 to $252.19.
• Sector setup: the sell-off into the print was commodity-led rather than company-specific. Spot gold fell 5.1% from $4,879.60 on April 17 to $4,629.60 on April 30. The VanEck Gold Miners ETF fell 12.0% over the same window and AEM fell 14.5%, so the stock was carrying roughly 250 basis points of underperformance against its own sector before a single number was published.
• Reaction session (May 1): the stock gapped down 1.0% to open at $186.31, traded a $182.80 to $187.69 range, and closed at $183.56, down 2.5% or $4.65 on the day. Volume of 2.8 million shares was 1.0 times the thirty-day average, which is a notably unenthusiastic tape for an earnings reaction.
• Relative move: AEM underperformed on every comparison available. Spot gold rose 0.3% to $4,644.50 on the session. The gold miners ETF fell 1.3%. The largest North American peer fell 2.2%. The S&P 500 rose 0.3%. AEM's 2.5% decline lagged its sector by roughly 120 basis points and the index by roughly 280.

A record quarter that beats on revenue and EPS and reiterates guidance does not usually close down 2.5% on flat volume while the underlying commodity rises. The explanation is that the two numbers the market had not already marked to market were both misses. The gold price was public information for the entire quarter, so the revenue and EPS beats carried almost no news. Production of 825,109 ounces against a compiled estimate of 859,426 and total cash costs of $1,093 against $1,057 were news, and both were negative.

The second factor is positioning. Coming into the print the stock had already given back 14.5% in nine sessions on a 5.1% move in the underlying, which is the shape of a crowded trade unwinding rather than a fundamental repricing. In that setup a print with no upward guidance revision has nothing to arrest the move with. Management reiterated rather than raised, and reiterating is what the release did on every line.

The third factor is the arithmetic of the next quarter, which the release made easier to compute and less attractive to look at. Spot gold on April 30 was 4.8% below the $4,861 Agnico realized in Q1. With the 48/52 disclosure implying roughly 807 thousand ounces in Q2, sequential revenue on an unchanged gold price is heading lower, not higher. Anyone updating a model on the afternoon of May 1 was marking Q2 down even while marking Q1 up.

Street Perspective

Debate: Is the Cost Inflation Structural or Self-Financing?

Bull view: Roughly 60% of the guided 2026 cost increase is royalties and a stronger Canadian dollar, both of which are consequences of the same conditions that lifted revenue 66%. Royalties are a percentage of a bigger number, not an erosion of the business. The cash margin per ounce nearly doubled while unit costs rose 22%, which is the only comparison that matters for cash generation.

Bear view: Roughly 40% of the increase is genuine inflation of about 4% plus mine sequence, and that portion is permanent. Underneath the royalty arithmetic, production costs per ounce rose 31.7%, seven of ten mines produced less, and two mines saw unit costs rise by more than a third. If gold retraces, the royalty component unwinds but the operating cost base does not, and the margin compression is asymmetric.

Our take: The bulls have the better argument on the cash-flow question and the bears have the better argument on the operating-quality question. The decisive fact is that Q1 landed above both cost-guidance midpoints in a quarter management describes as running to plan, which means the plan itself was set at a level Q1 did not clear. We treat this as a live risk rather than a settled debate, and it is the first item on the scorecard we will grade next quarter.

Debate: Does the Growth Pipeline Justify a Premium Multiple?

Bull view: Five funded projects, all brownfield or adjacent to existing infrastructure in jurisdictions where the company has decades of permitting history, a stated path to 20% to 30% production growth over the decade and above four million ounces in the early 2030s, and management's claim of 30% to 60% IRRs at current prices. Growth of that shape is scarce in a sector where most producers are managing decline.

Bear view: The first step-up is 2030. Between now and then the company spends roughly $2.6B a year of capital against a production base that shrank 5.6% year over year. Hope Bay's capital number moved before the study was even published, and Detour's scope is being described as three concurrent expansions with no number attached. Paying a premium today for ounces that arrive after 2030 is paying for four years of execution risk in advance.

Our take: The bear timing objection is fair and is exactly why the multiple has compressed. Where we side with the bulls is on capital risk: brownfield restarts with standing infrastructure and a resident workforce fail differently, and less expensively, than greenfield builds. The May Hope Bay release is the first real test and it arrives before the next print.

Debate: Is the Balance Sheet an Asset or Idle Capital?

Bull view: $2.9B of net cash, an A‑ rating, an undrawn $2B revolver and a $1.8B investment portfolio give Agnico the ability to fund the entire pipeline through a gold drawdown and to move on assets when peers cannot. The doubled NCIB ceiling signals intent to convert that flexibility into per-share value.

Bear view: A $3B to $5B stated cash target on a company producing roughly $5.6B of annual free cash flow is a two-year accumulation with no defined use. Cash earning a money-market return in a business with 30% to 60% project IRRs is value destruction by arithmetic, and the buyback ran at half pace in the quarter the balance sheet grew.

Our take: The bear critique is right in principle and mistimed in practice. Q1's buyback slowdown has a complete and verifiable explanation in the $1.3B tax payment, and management attached a dated commitment to reversing it. If Q2 repurchases do not step up materially toward the $400M per quarter that a 40% payout implies, this debate becomes a governance concern rather than a timing one.

Debate: Does the De-Rating Create an Entry Point?

Bull view: The stock is 27.2% below its 52-week closing high on a quarter that set records for margin, adjusted net income and net cash. At $183.56 the enterprise value is roughly 7.6 times our full-year adjusted EBITDA estimate, against a business with net cash, tier-one jurisdictional concentration and funded growth. The de-rating tracked gold, not fundamentals.

Bear view: The stock is still up 60.1% over twelve months. Gold has rolled over 5% from its April high and gold equities carry roughly three times that beta on the way down as well as up. Buying a producer whose volumes are falling and whose unit costs are rising, on the argument that the commodity will hold, is a commodity bet dressed as a value case.

Our take: We agree that it is a commodity bet and think the entry price compensates for it. At $183.56 the equity discounts roughly $4,250 per ounce gold on our framework, 8.5% below the May 1 spot of $4,644.50 and 5.6% below the $4,500 assumption embedded in the company's own guidance. That is a real cushion, and it is the basis of the rating.

Model Update & Valuation Framework

This is our first published model on Agnico Eagle, so the table below sets the initial assumptions rather than revising prior ones. Volume and capital anchor to company guidance. The realized gold price is our assumption, set at $4,700 per ounce for the full year: Q1 delivered $4,861 and spot on May 1 was $4,644.50, so this holds the balance of the year roughly at spot.

Two assumptions carry most of the risk. We model total cash costs at $1,085 per ounce rather than the $1,070 guidance midpoint, because Q1 printed $1,093 and the improvement required to reach the midpoint is roughly $30 per ounce across the remaining ounces. And we assume no contribution and no share issuance from the Finnish transactions, because the share count has not been disclosed. A Rupert closing late in Q2 will raise the average share count for the year and reduce per-share figures by a proportion we cannot yet size.

Valuation and Price Target

At the May 1 close of $183.56 on approximately 500.0 million shares outstanding net of shares held in trust, the market capitalisation is roughly $91.8B. Deducting $2,915M of net cash gives an enterprise value of roughly $88.9B, or 7.6 times our FY2026E adjusted EBITDA of $11,720M. On our estimates the free cash flow yield is 6.1% and the adjusted P/E is 14.1 times.

We set a 12-month price target of $210, which implies 8.7 times FY2026E adjusted EBITDA plus net cash, and roughly 14.4% upside from $183.56. An 8.7 times multiple is below where the stock traded at its 52-week closing high of $252.19, which on the same estimates was 10.5 times, and is a premium to the sector we consider justified by jurisdictional concentration, net cash and the funded growth profile.

Gold Price Sensitivity

Because every line above is levered to one input, the table below runs the framework across realized-price scenarios, holding volume at 3,400 thousand ounces, capital at $2,595M and the multiple at 8.7 times. Royalty and NTI costs are assumed to absorb roughly 2.5% of the incremental revenue in each direction.

Thesis Scorecard: Initiating Coverage

No prior Aardvark Labs coverage of Agnico Eagle exists, so this quarter establishes the thesis rather than grading one. The pillars below become the standing framework, and each subsequent recap will score this same list rather than inventing a new one.

Overall: Thesis established with three bull pillars intact and two of three risks already active. The bull case rests on assets and balance sheet, both of which the quarter confirmed. The bear case rests on unit costs, which the quarter made worse.

Action: Buy, sized for commodity volatility. The entry price discounts a gold environment roughly 8.5% worse than the one that exists, which is the margin of safety. Position size should reflect that the downside case is a gold drawdown rather than a company failure, and that gold equities carry roughly three times the commodity's beta in both directions.

Commitments to Watch Next Quarter

1. Q2 production of roughly 807 thousand ounces, implied by the 48% first-half weighting against Q1's 825,109 ounces.
2. Total cash costs trending toward $1,063 per ounce for the balance of the year if the $1,070 full-year midpoint is to hold; anything above $1,129 puts the range itself at risk.
3. Share repurchases stepping up materially from $150M, toward roughly $400M per quarter, which is what a 40% full-year payout implies for the remaining three quarters.
4. The Hope Bay construction decision in May, and whether initial capital lands inside the "slightly over $2 billion" framing rather than nearer the "below $2.5 billion" bound.
5. The Macassa paste plant fully operational in Q2, with the mill ramping toward 2,040 tpd by year end and roughly 40,000 ounces of 2026 production from the AK deposit.
6. Kittila mill recovery above 81%, and whether the three named improvement initiatives move it.
7. The Rupert and Aurion closings late in Q2, and the first disclosure of the share count issued.
8. Cash taxes at $525M to $575M per quarter, and the free cash flow step-up that follows.

Bottom Line

Agnico Eagle reported the best financial quarter in its history and the stock fell, and both facts are correct responses to the same set of numbers. The records are real: $4,099.6M of revenue, $1,706M of adjusted net income, $3,011M of adjusted EBITDA, a $3,768 per ounce cash margin, and $2,915M of net cash behind an A‑ rating. They are also almost entirely the product of a $1,970 per ounce increase in the realized gold price, which was public information before the release was published and therefore carried no news.

What carried news was the operating detail underneath. Production fell 5.6% and missed the compiled estimate by 4%. Total cash costs rose 22.1% and missed by $36 per ounce. Both cost measures finished above the full-year guidance midpoints in a quarter management characterised as tracking to plan. Seven of ten mines produced fewer ounces than a year ago and nine of ten produced them more expensively. Macassa's unit costs nearly doubled and drew no question. Kittila's mill recovery is stuck at 81% and drew none either.

We initiate at Outperform anyway, for a reason that has little to do with the quarter and everything to do with the price. At $183.56 the equity discounts a full-year realized gold price near $4,250 per ounce. Spot on the day was $4,644.50. Agnico's own cost guidance assumes $4,500. The stock is priced for a gold environment materially worse than the one that exists, on an asset base that is 86.5% Canadian, carries net cash rather than debt, funds a decade of brownfield growth from operating cash flow, and is about to see roughly $1.25B per quarter of tax drag fall away from free cash flow.

The risks are named and dated rather than vague. If second-quarter production comes in materially below 807 thousand ounces, the second-half weighting stops being a shape and starts being a shortfall. If unit costs do not improve roughly $30 per ounce from here, the cost midpoint goes and the range follows. If the Hope Bay sanction lands nearer $2.5B than $2.0B, the capital discipline argument that underpins the growth premium weakens. And if gold breaks below $4,200, the entire margin of safety in the entry price is gone. None of those is a reason not to own the shares at this level. All of them are reasons to know exactly what you are underwriting, which is a commodity price with an unusually well-run operator attached.