Costs Broke the Quarterly Guide and the Discount to Spot Became a Premium: Downgrading Agnico Eagle to Hold

Results vs. Consensus

Q3 2026 Scorecard

Year-Over-Year Comparisons

Operating Metrics, Year Over Year

The 2025 cost figures above are the restated ones the company prints alongside the 2026 numbers. Effective January 1, 2026 the composition of total cash costs and AISC changed to adjust for the NTI Payment at Meadowbank, and the comparatives are restated onto that basis. On the old composition, Q3 2025 total cash costs were $994 per ounce and AISC was $1,373. A reader comparing this quarter against a figure lifted from last year's release is comparing two different measures, and the restatement flatters the year-over-year cost increase by roughly $42 per ounce at the total cash cost line.

Quarter-Over-Quarter Comparisons

Both quarters sit on the same cost composition, so the sequential comparison needs no adjustment. It is the cleanest read available on direction, and the direction is unambiguous: every operating and cash line is worse, and the cash margin per ounce has now fallen for two consecutive quarters, from a record $3,768 in the first quarter to $3,429 in the second and $3,131 here. Roughly three-quarters of that sequential margin decline is the gold price and one-quarter is unit costs. Last quarter the cost line offset part of the price decline instead of adding to it, which is the change that matters.

Nine Months to September 30

Quality of Beat

• Revenue. A 17.1% increase on 5.0% fewer ounces sold. The realized price did all of it: $4,264 against $3,476 a year ago is $788 per ounce of pure price, worth roughly $650M on 825,084 ounces sold, against a reported revenue increase of $524.1M. Volume was a $151M negative. The variance table the company publishes attributes the entire $524.1M to revenue growth and names price and higher sales volume at Meliadine, LaRonde and Kittila as the drivers, offset by Canadian Malartic and Meadowbank. There is no organic revenue growth in this print.
• Margins. Gross margin expanded 159bp year over year to 60.1% and contracted 370bp sequentially from 63.8%. The compositional fact matters more than either: total cash costs of $1,133 include a $92 per ounce by-product credit against $57 a year ago and $75 last quarter. Hold the credit at the year-ago level and the quarter prints $1,168 rather than $1,133. Hold it at last quarter's level and it prints $1,150. Silver and copper strength is doing $35 per ounce of work on a cost line that missed its own quarterly guide by roughly $11, which means the underlying operating cost deterioration is worse than the headline.
• EPS. IFRS net income of $1,352.9M includes $45M of non-recurring tax benefits, worth $0.09 per share, which is the single largest adjustment and runs in the company's favour. The effective tax rate of 31.4% sits below the 34% to 36% full-year guide. Adjusted net income of $1,324M strips that plus $11M of purchase-price inventory allocation, $8M of derivative losses, $7M of foreign exchange and $12M of other items, and the $2.62 basic adjusted figure is the clean one. Finance costs fell 37.7% and care and maintenance fell 65.3% as Hope Bay moved from care to construction, which together contributed roughly $21M of pre-tax benefit that will not repeat at that rate.
• Free cash flow. This is the line that deserves the most attention and got the least. Free cash flow of $852M is down 28.4% year over year and 36.2% sequentially, in a quarter when net income rose 28.2%. Two things did it. Operating cash flow fell 3.4% despite an 18.0% increase before working capital, because the Nunavut summer sealift built $236M of inventory and other current assets absorbed a further $173M, against a year-ago quarter that released $190M from income taxes payable. And capital spending rose 43.0% to $921M. On a per-share basis free cash flow fell 29.1% to $1.68. Free cash flow before working capital rose only 2.1%, and that is the number describing the business: the gold price gave 22.7% and costs and capital took almost all of it.

Segment Performance

Agnico does not publish revenue by region in the quarterly release, so the tables below use payable production and total cash cost per ounce, both of which the company discloses per mine and per region. The 2025 cost figures are on the restated composition, so the comparison is like for like. Regional costs for the prior-year quarter are production-weighted averages of the disclosed per-mine figures; the same weighting reproduces both disclosed consolidated figures exactly and each disclosed current-quarter regional figure to within a dollar.

Production and Unit Costs by Region

Canada supplied 85.2% of payable production, essentially unchanged from 85.4% in the second quarter, which is the jurisdictional pillar of this thesis expressed in ounces. Every one of the six regions produced gold more expensively than a year ago. That was also true in the first quarter, but not in the second, when Finland and Mexico both improved. The return of that breadth matters more than the magnitude.

Production and Unit Costs by Mine

Six of ten mines grew production year over year, against three of ten in each of the first two quarters of 2026. Two of ten produced gold more cheaply, against four of ten in the second quarter and one of ten in the first. The distributions moved in opposite directions, and that is the single most useful summary of the quarter: the operating base got broader and the cost base got worse everywhere at once. The two cheaper mines, Meliadine and Goldex, are 15.5% of production between them, and Goldex's improvement is a by-product credit rather than a productivity gain.

Abitibi Ontario — the engine, now with a visible grade step-down

Ontario produced 263,700 ounces, up 3.3%, at $969 per ounce, up 24.5%. Detour Lake delivered 183,118 ounces on a higher-grade sequence, positive local grade reconciliation and a third consecutive quarter of record mine performance at 32.5 million tonnes of ore and waste extracted, and poured its eight millionth ounce. The mill lost fourteen days: a seven-day planned shutdown and a seven-day unplanned one to replace a tailings line after a minor breach on August 29. Macassa set a third consecutive quarterly mill throughput record at 2,870 tonnes per day, which partly offset lower-than-expected grades from the AK deposit, and the LZ5 mill at LaRonde processed 93,000 tonnes of AK ore for roughly 10,500 ounces.

"Detour's excellent performance was a function of a few factors. We saw improved productivities, we continued mining in a high-grade portion of the deposit, and we saw some localized positive grade reconciliation as well. As expected, the operation has now transitioned into a lower grade mining sequence that we anticipate to continue throughout the fourth quarter." — Natasha Vaz, EVP and COO, Ontario, Australia and Mexico

Macassa remains the portfolio's worst year-over-year cost comparison at +80.4%, and the base is genuinely flattering: $659 per ounce in the third quarter of 2025 was an exceptional print. The nine-month figure of $1,152 is the representative number, and it is 79.2% above the comparable nine-month 2025 figure, so the deterioration is not a base effect alone. The tailings line breach released a small quantity of tailings and process water into secondary containment, with water quality returning to pre-incident conditions within 48 hours.

Assessment: Ontario did exactly what we said in July it would do, in both directions. It carried the quarter on process improvement, and the grade sequence has now turned down and will stay down through the fourth quarter, in which the Line 2 SAG mill gear replacement adds thirteen days of running at roughly 60% of total mill throughput on top of a seven-day planned shutdown. Management frames the extension as preventative work pulled forward from 2027 and still expects to meet the site's production guidance. Treat Ontario's cost advantage as compressing from here with nothing going wrong.

Abitibi Quebec — the damage, quantified and bottoming

Quebec produced 218,392 ounces, down 18.4%, at $1,238 per ounce, up 28.8%. Canadian Malartic is nearly the whole variance: production fell 29.1% to 111,202 ounces on lower throughput and grade, and unit costs rose 57.2% to $1,508 as the mill ran on low-grade stockpile while Barnat was shut. Odyssey mined a record 391,000 tonnes and East Gouldie contributed 27,250 ounces. LaRonde ran lower grades on a higher proportion of LZ5 ore and completed planned underground and mill shutdowns. Goldex was the portfolio's second-cheapest mine at $902.

"First highlight on operation, Barnat Pit. It is a good example of the resilience I just mentioned. On July 1st, the team had a movement of the rock into the Barnat Pit, and within 24 hours of the rock movement, the team did the assessment and they established a safe path forward. The plan was completed on September 27th, including building rock catch bench, and that was done slightly ahead of the plan. The mining is restarted and we are now mining safely into the Barnat Pit." — Dominique Girard, EVP and COO, Nunavut, Quebec and Finland

The restart on September 27 came in ahead of the commitment made in July, which was remediation during the third quarter and mining resuming in the fourth, and management described the remediation itself as slightly ahead of plan. The quantified damage is unchanged at 60,000 to 80,000 ounces in the second half of 2026 and up to 150,000 ounces in each of 2027 and 2028, with low-grade stockpile partially mitigating. One new consequence surfaced this quarter: the redesigned mine plan carries a higher units-of-production amortization rate because the life-of-mine tonnage fell, and the company now expects full-year depreciation and amortization towards the upper end of its $1.55 billion to $1.75 billion range for that reason.

Assessment: This is the cleanest piece of good news in the release and it is a cost item, not a production item. Canadian Malartic produced roughly 46,000 fewer ounces than in the prior-year quarter, on lower throughput and grade following the wall movement, and the pit is being mined again. The amortization step-up is the quiet confirmation that the 370,000 sterilised ounces have a balance-sheet consequence as well as a cash one, and the reserve treatment at year-end is still the item to watch. Quebec's unit cost should improve sequentially in the fourth quarter as Barnat ore returns to the mill in place of low-grade stockpile.

Nunavut — the model was wrong at Meadowbank

Nunavut produced 211,650 ounces, down 8.0%, at $1,117 per ounce, up 10.9%. Meliadine was in line with plan at 96,149 ounces and was one of only two mines in the portfolio to lower its unit cost year over year, to $1,046 from $1,128, on mill throughput above plan and strong open-pit performance. Meadowbank missed plan at 115,501 ounces, down 15.2%, on grades below plan from negative ore-tonnage reconciliation in the IVR pit and reduced underground production on lower mechanical availability.

"Really what happened, it is really in a specific area in the IVR pit where we have the discrepancy between the model and what we see. Our teams are on that to update the model and to make sure that we have the right plan. That does not affect at all the underground and the Whale Tail pit. It is a localized area into the IVR pit." — Dominique Girard, EVP and COO, Nunavut, Quebec and Finland

The region's full-year production is still described as on track within the guidance range, with Meadowbank grades improving in September and expected to rise further in the fourth quarter on the planned mining sequence. Meliadine's underground development was temporarily diverted to install additional ground support at key intersections and is expected to return to planned rates in the fourth quarter. The summer sealift drove the $236M inventory build that took the free cash flow line down.

Assessment: A localised geological model error is a smaller problem than a systemic one, and the company was specific about its boundaries. It is still the second consecutive quarter in which Meadowbank missed its plan, and it is the asset with the portfolio's highest diesel exposure and the NTI Payment. Nunavut is 26.0% of production and the place where the diesel headwind lands hardest, so its unit cost is the one to track as the $50 to $60 per ounce fourth-quarter diesel impact arrives.

Finland — the recovery problem closed

Kittila produced 59,907 ounces, up 3.4%, at $1,162 per ounce, up 12.2%, on mill recoveries of approximately 85%. That resolves the one unquantified operational item we carried from the last two quarters. Recovery had been stuck at 81% with four named improvement initiatives and no target or date attached to any of them; the release now attributes the improvement to specific work including reagent optimisation based on feed blend.

"The other highlight I would like to do for the quarter is the good work done by the Kittilä team, especially the mill team. ... The team worked systematically over the last years to improve their mill recovery. That was not an easy challenge. In September, they achieved the highest monthly recovery in the last five years." — Dominique Girard, EVP and COO, Nunavut, Quebec and Finland

Kittila has a 17-day fourth-quarter shutdown for mill maintenance and autoclave relining, the longest scheduled full shutdown anywhere in the portfolio. Alongside it, the newly acquired Ikkari project began drilling with three rigs completing 12,332 metres, rising to five rigs in the fourth quarter once winter freezing improves access to wetland platforms. The company holds 207,654,166 contingent value rights outstanding from the Rupert acquisition, with payment conditional on reserve and production milestones at Ikkari.

Assessment: Four percentage points of recovery on 1.57 million tonnes milled over nine months is worth real ounces, and this is the clearest evidence in the release that the productivity argument management makes across the portfolio is more than narrative. Kittila's unit cost still rose 12.2% because royalties scale with the gold price, which is the pattern everywhere this quarter. The internal evaluation of an optimised Ikkari mine design is targeted by the end of 2027, so the Finnish growth story remains a 2028 and beyond item with no public economics attached.

Australia and Mexico — both small, both more expensive

Fosterville produced 40,090 ounces, up 14.7%, at $1,189 per ounce, up 11.5%. Production beat plan on positive grade reconciliation at Robbins Hill and Phoenix and higher mill recovery, and the mine set a monthly development record of 1,154 metres in August. Grinding circuit upgrades and the tie-in of a new bacterial oxidation tank continued, targeting 3,300 tonnes per day from 2028. Pinos Altos produced 20,273 ounces at $2,093, the portfolio's highest unit cost and up 9.8%, on lower grades and a stronger Mexican peso.

Assessment: Fosterville is the only mine in the portfolio where the production trend improved and the cost trend is explainable by something other than input inflation. Together these two assets are 7.4% of production and neither moves the consolidated case. Pinos Altos at $2,093 per ounce still earns a $2,171 cash margin at the current realized price, which is the clearest illustration of why a high-cost asset is not a problem at this gold price and becomes one quickly below $3,000.

Key Topics & Management Commentary

Overall Management Tone: Deliberately redirected. Management opened by conceding that results were "largely in line with expectations" with "no big surprises," then spent the bulk of the call on the growth pipeline rather than the quarter, and closed by pointing to the November 18 Capital Markets Day. On costs the posture was factual and unusually specific, naming the diesel impact to the dollar per ounce for three separate periods and volunteering that quarterly cash costs came in above the top of the internal guide. On 2027 the posture was consistently defensive and deferred to February, including on the cost base, the payout ratio and the reserve price. The question period was short, with eight questioners and no queue remaining.

1. Costs Broke the Quarterly Guide, and Management Said It First

The CEO's second sentence of the call conceded the cost miss against the company's own internal quarterly guidance, before the CFO reached the financials. That is a deliberate choice and it is the right one, because the alternative was to let the AISC line speak for itself. At $1,648 per ounce, third-quarter AISC printed $98 above the top of a full-year range the company still says it will land inside.

"Gold production of 814,000 ounces came in at the midpoint of our adjusted production guidance for the quarter, and cash costs of $1,133 came in 1% above the top end of our cost guidance. This is no small accomplishment in a quarter where oil is $100 a barrel, where the price of diesel is almost double, than where trade, tariffs, and war are challenging established supply chains and logistics." — Ammar Al-Joundi, President and CEO

The full-year framing moved with it. In July, production was "near the lower end" and costs were simply "within the range." Now production is still near the lower end and both cost measures are "near the upper end." The ranges themselves did not change, which lets the company describe guidance as reaffirmed. The distribution inside the ranges did, and that is the substance.

Assessment: Both full-year midpoints are now arithmetically out of reach. Holding the $1,070 total cash cost midpoint would require a fourth quarter near $1,004 per ounce, roughly $50 below the best quarter the company has printed this year, in the quarter with the heaviest diesel load. Holding the $1,475 AISC midpoint would require roughly $1,314. The ranges are still reachable, but the AISC range requires sequential improvement from here and the schedule argues the other way.

2. Diesel: $30 in the Third Quarter, $50 to $60 in the Fourth, Up to $80 for the Year

The company quantified the diesel headwind with more precision than it has quantified anything else this year, and the numbers escalate across every horizon. Diesel is approximately 11% of operating costs, which is below the sector because the Ontario and Quebec operations run on grid hydro and nuclear power and the portfolio is predominantly underground.

"We disclosed the impact for Q3, the higher diesel price relative to what we budgeted had an impact of about $30 an ounce. That could be $50-$60 an ounce in Q4, and for full year 2026 could be as much as $80 an ounce. So it is about a 7% increase, potentially year-over-year." — Jamie Porter, EVP and CFO

On third-quarter production, $30 per ounce is roughly $24M. On a 3.33 million ounce full year, $80 per ounce is roughly $266M. Canadian retaliatory tariffs add approximately $5 per ounce to fourth-quarter costs, already incorporated in guidance, and the CFO said that figure rises without a trade deal. Offsetting partially: a stronger US dollar and higher silver and copper prices, and roughly 62% of the remaining 2026 Canadian dollar exposure hedged with a floor below 1.38 C$/US$.

Assessment: This is the most quantifiable cost headwind the company has disclosed in the three quarters we have covered, and unlike the Nunavut diesel pre-purchase flagged in July it is not a one-time, already-incurred item. It is a run-rate that escalates into the fourth quarter and that management explicitly declined to forecast for 2027. The relative advantage is real and worth something; the absolute number is still $266M.

3. Sustaining Capital Did as Much Damage as Diesel

The AISC bridge is more informative than the AISC number. Of the $317 per ounce year-over-year increase, $181 came from total cash costs and $136 from the three items above them: sustaining capital up $78, non-cash reclamation and sustaining leases up $50, and G&A up $8. Sustaining capital per ounce of $360 compares with $286 in the second quarter and $296 for the nine months.

Sustaining capital of $287M in the quarter was concentrated at Detour Lake ($96M), Nunavut ($75M) and Quebec ($47M), and the release attributes the AISC increase in part to higher sustaining capital at Detour Lake and Fosterville. The reclamation line is non-cash and partly a function of discount rates, and the balance sheet shows the current reclamation provision rising to $233M from $145M at year-end while the non-current provision fell to $1,170M from $1,318M.

Assessment: Nearly 43% of the AISC increase sits above the cash cost line, and about half of that is capital intensity rather than operating cost. That matters for the forward model: sustaining capital of $360 per ounce is not a diesel problem that reverses with the oil price, it is what maintaining a portfolio of ten mines costs while four growth projects are being built through the same organisations. The nine-month figure of $296 is the better forward anchor, and it is still 18.4% above the prior-year nine months.

4. Free Cash Flow Fell 28% in a Quarter When Earnings Rose 28%

Free cash flow of $852M is the second-lowest quarterly figure of 2026, above only the $732M first quarter that absorbed the 2025 tax catch-up, and the weakest year-over-year comparison in the release. Operating cash flow fell 3.4% to $1,755M, which is remarkable against an 18.0% increase before working capital. The Nunavut sealift drove a $236M inventory build, other current assets absorbed $173M, and the year-ago quarter had the benefit of a $190M release from income taxes payable.

"We delivered solid operational results and cost control, generating free cash flow of $852 million." — Jamie Porter, EVP and CFO

On the other side, capital spending rose 43.0% to $921M, of which $848M was capital expenditure and $73M capitalised exploration, and development capital alone was $625M. Free cash flow before working capital rose 2.1% to $1,057M. Free cash flow per share fell 29.1% to $1.68.

Assessment: The working-capital explanation is correct and the sealift genuinely reverses. The capital explanation does not. Nine-month free cash flow is down 5.5% year over year on a 40.9% higher realized gold price, which is the clearest single statement of what rising unit costs and a record capital programme are doing to cash conversion. The CFO said total capital will be "north of $3 billion this year" and "that's likely to increase over the next few years," which means the cash-conversion question is structural and not seasonal.

5. The Growth Pitch Moved to the Front of the Call

Management restructured the call around the pipeline and said so explicitly. The four largest projects are now framed as adding 1.3 to 1.5 million ounces a year by the mid-2030s, with the first production step-up in 2030 and the potential to exceed 4.0 million ounces a year in the early 2030s.

"That said, with results largely in line with expectations and with no big surprises, this morning, we wanted to spend less time talking about this past quarter, and we want to spend more time talking about the substantial and important value drivers of our business going forward." — Ammar Al-Joundi, President and CEO

The underlying numbers are new or newly precise. Canadian Malartic is framed as 400,000 to 500,000 ounces of incremental annual production towards one million ounces a year starting as early as 2033, through a second Odyssey shaft, the Marban satellite pit and the Wasamac underground project. Upper Beaver is 200,000 to 225,000 ounces plus 3,600 tonnes of copper at 5,000 tonnes per day. Hope Bay is 400,000 to 435,000 ounces with completion in 2029 and first gold in 2030. Detour Lake is targeting one million ounces a year by the early 2030s with underground production possible as early as 2028.

Assessment: The pipeline is the thesis for the 2030s and it is advancing on or ahead of schedule at every project. It is also the reason the conversation about 2027 was deferred twice. A company that redirects its own earnings call away from the quarter towards a November 18 Capital Markets Day is telling you where the equity story is going to be made, and it is not in the next two prints.

6. Hope Bay Is Ahead of Plan and the Sealift Landed Complete

Hope Bay is the largest single capital commitment in the company's history and it is tracking. Overall construction reached approximately 13% at September 30, described as slightly ahead of plan, and the 2026 sealift delivered approximately 207,000 cubic metres of cargo and 16 million litres of fuel using nine cargo vessels, a dedicated wind turbine vessel and a fuel tanker. Development capital at Hope Bay was $190M in the quarter and $338M for the nine months.

"The project is moving from investment decision to a disciplined execution. It is the fourth mine we are building in Nunavut, and I am very happy with the team and the strong plan we have in hand." — Dominique Girard, EVP and COO, Nunavut, Quebec and Finland

A new framework agreement with the Kitikmeot Inuit Association was finalised in October, extending the Inuit Impact and Benefit Agreement to 2056 and expanding its coverage across the entire Hope Bay belt. Boston deposit assays arrived as expected: 7,500 metres in eight holes, with 11.0 g/t gold over 12.3 metres at 298 metres and 17.0 g/t over 4.5 metres at 308 metres, from the first drilling at the deposit since the 2021 acquisition. A more aggressive 2027 field and drill programme is planned.

Assessment: Two commitments from the last note are closed here: Boston assays arrived in the October release and were good, and the sealift landed. The IBA extension to 2056 across the whole belt is the quieter and more valuable item, because it converts the social licence for a single mine into a licence for a district. The disclosure is less tidy than the execution: the same release gives detailed engineering as approximately 73% in its highlights and approximately 76% in its project section, with 75% said on the call.

7. Odyssey Shaft One Is Four Months Early and Shaft Two Enters Permitting

The first phase of shaft sinking at Odyssey completed on July 9 at 1,586 metres, which management described as four months ahead of schedule, having described it as three months ahead in July. Mine development advanced 4,874 metres in the quarter, the main ramp reached 1,232 metres, and the headframe changeover to the production hoist is underway with production sheaves installed and sinking infrastructure being dismantled. Shaft-hoisted production from East Gouldie remains scheduled for the second quarter of 2027.

"We are on target for the first production via the shaft in Q2 2027, with over 70% of the construction completed. ... That shovel have dig a 6.5-meter diameter hole to 1.5 kilometer deep, and we completed that first phase of sinking with our team four months ahead of the schedule." — Dominique Girard, EVP and COO, Nunavut, Quebec and Finland

Phase 2 of the paste plant progressed towards fourth-quarter testing and second-quarter 2027 commissioning. The second phase of shaft sinking is now dated: it commences in 2029, completes in 2031 at a final depth of 1,870 metres, with a third loading station commissioned in 2031. A potential second shaft is in internal technical evaluation, with a decree modification application expected in the first quarter of 2027 and a permitting process of approximately 18 months.

Assessment: This is the project on which the next three years of production growth depends, and the dates have held or improved at every disclosure. The second-shaft permitting timeline is the new information: an application in the first quarter of 2027 plus 18 months puts approval in late 2028, which is consistent with the "as early as 2033" framing for one million ounces at the complex and tells you the decision is being made on the East Gouldie mining rate study now in progress.

8. The Balance Sheet Got Larger Than the Business Needs

Net cash rose $149M to $3,416M against total debt of $197M, unchanged. The revolver was undrawn at quarter-end with approximately $2 billion available. After quarter-end the company amended and restated that facility, increasing it to $3 billion with an uncommitted accordion of $1.5 billion and a maturity extended to October 28, 2031, and plans to file a new base shelf prospectus in November with no present intention to issue.

"Yeah, I'd say it's really just right-sizing our overall liquidity. We had an opportunity, the bank market's quite competitive currently, so we had an opportunity to upsize the facility on very attractive terms, without really increasing our incremental cost much. ... Our capital spending is going up. We'll be north of $3 billion this year in total capital, and that's likely to increase over the next few years." — Jamie Porter, EVP and CFO

Committed liquidity is now approximately $6.6 billion, comprising $3,613M of cash and the $3 billion revolver, against a capital programme guided at $2.9 billion to $3.2 billion including capitalised exploration. Cash taxes of $2.9 billion have been paid in the first nine months, including the $1.3 billion catch-up for the 2025 taxation year, with $525M to $575M remaining.

Assessment: A $3 billion committed facility with a further $1.5 billion accordion, plus a shelf, at a company with $3.4 billion of net cash and $197M of debt, is more liquidity than the base business requires. The CFO's framing is defensive positioning for a lower gold price while a rising capital programme runs, and that is a sensible reading. It is also the shape of a balance sheet built to act, and the Capital Markets Day is on November 18.

9. The Buyback Shrank 37% and Paid 8.6% Above the Quarter's Average Price

The company repurchased 1,297,940 shares at an average of $192.60 for $250.0M, against 2,235,947 shares at $178.86 for $399.9M in the second quarter. Total returns of $478M comprise the $250M buyback and a $0.45 per share dividend worth approximately $228M, representing 56% of quarterly free cash flow. Nine-month returns are 51% of free cash flow against an initial 40% target, and the company now expects to exceed 50% for the full year.

The quarterly dividend has been $0.45 per share all year and did not move. The weighted average basic share count rose 0.88% sequentially to 506.1 million as the Rupert consideration fully entered the average, which is almost exactly the step-up we flagged in July.

Assessment: We asked in July whether repurchases were being executed with any reference to the share price, after the company averaged $178.86 into a quarter that closed at $155.13. The answer this quarter is no: it bought 42% fewer shares, at a 7.7% higher price, 8.6% above the quarter's own average close, and 13% above the price at which the stock trades today. Buying less when the stock is more expensive is at least the right direction of travel on size, but a programme that systematically pays above the period average is transferring value rather than creating it.

10. Exploration Is Running at Record Volume With Results That Matter

Drilling totalled more than 330 kilometres in the quarter across more than 120 rigs, for 1.1 million metres year to date against a 1.4 million metre budget. Exploration and corporate development expense rose 29.3% to $77.1M on regional targets in Finland and Canada, and capitalised exploration was $73M. The company expects to replace and grow global mineral reserves and resources at year-end for a further consecutive year.

"So we are anticipating good progress in reserve and resources addition both at Malartic, Detour. Hope Bay, we're going to be updating our reserve at year-end. There's also a fair bit of positive replacement at the existing operations, such as Meliadine, Kittilä, Macassa. So we're in pretty good shape overall in terms of reserve and resources replenishment." — Guy Gosselin, EVP, Exploration

The results carrying the most forward value: 18.9 g/t gold over 26.4 metres at 331 metres in Domain 54 at Detour Lake, within the high-intensity programme feeding the planned bulk sample; 2.8 g/t over 34 metres at 1,950 metres at East Gouldie, described on the call as roughly 100 metres below the current shaft depth and 100 metres west of the first shaft's planned ultimate depth; 9.5 g/t over 18.9 metres in a newly named Artemis porphyry sub-zone at Odyssey; and the first Helmi results in Finland, with hole IKK26-124 returning 3.2 g/t over 10.7 metres at 476 metres on a capped basis.

Assessment: Exploration is the one part of this business where the input cost inflation buys more rather than less, because the constraint is drill metres and geology rather than diesel. The East Gouldie intercept below the planned shaft depth and the Detour Domain 54 results are the two that could change a mine plan, and both land in the February reserve statement. That statement is now carrying three questions at once: reserve replacement, the Barnat sterilisation treatment, and the reserve gold price.

11. Labour Availability, Not Wages, Is the Named Ten-Year Risk

Asked directly about labour cost inflation, management separated price from availability and said the cost side is quiet. The availability side is not, and the company is building four mines simultaneously in regions where it is already the largest employer.

"We're not seeing, at this point, material increases in labor costs. The bigger challenge is in availability of high-quality labor. The canary in the coal mine, as you know, you've been in this industry for a while, is usually on the exploration side with drillers, because they tend to be more transient." — Ammar Al-Joundi, President and CEO

The disclosed response is a centralised recruitment hub covering all of Canada rather than Nunavut alone, a targeted immigration programme aimed at trades and heavy-duty mechanics being used at Macassa and trialled at Detour, skills development programmes with guaranteed job offers, and Macassa development miners seconded to Detour Lake, which let Detour internalise its exploration ramp development contract and set a quarterly development record. Hope Bay's camp is 800 to 900 beds with a workforce roughly double that; Upper Beaver is around 700 people. Housing is named as a constraint, with land acquisition and construction underway with the municipality in the Kirkland Lake district.

Assessment: This is the most candid labour disclosure we have heard from the company and it is a long-dated risk rather than a 2027 one. The Macassa-to-Detour secondment is the specific evidence that the regional operating model has value beyond procurement, and it produced a measurable result this quarter. Housing is the item with the least progress attached: management described the plans as "still in the oven."

12. 2027 Is Still a Blank Page, and the Date Is February

Every forward question was deferred. The 2027 cost base, the payout ratio, the three-year production plan, the reserve gold price and the labour requirement by project were each answered with a version of the same sentence: the budget is not finished and guidance comes in February. One number did get volunteered.

"Beyond that, I don't know where we're going to end up on labor and other consumables and everything else, but yeah, 5% or 6% is a reasonable assumption." — Jamie Porter, EVP and CFO

"We don't have any more insight into long-term diesel prices in 2027, I think, than anyone else. We have hedged some of it. We've bought some of it. We use less than most of our peers. But if I was in your position, that is going to be the tough one to model." — Ammar Al-Joundi, President and CEO

On the reserve price, the answer was a constraint rather than a number: inflation in fuel and other inputs is being taken into account, the view is being finalised, and the objective is to keep the cut-off grade stable. On the payout, 50% was described as a reasonable threshold and not an unreasonable assumption for 2027, subject to the gold price and the capital programme.

Assessment: A 5% to 6% inflation assumption applied to a fourth-quarter 2026 exit rate near $1,164 per ounce puts 2027 total cash costs in the $1,220 range before any productivity offset, which is $150 above this year's guidance midpoint. That is the number the February release has to deal with, and it is why we now anchor our forward framework to $1,215 per ounce rather than this year's $1,090. The cut-off grade commitment is the useful disclosure: it tells you the reserve price will rise with costs, which protects ounces at the expense of the reserve grade narrative.

13. The Capital Markets Day Is the Catalyst, and Management Said What It Is For

The November 18 Capital Markets Day was referenced in the prepared remarks and three times in the question period. Management declined to pre-announce content but was unusually direct about purpose.

"I think the objective of that day is two things. One is we are spending a lot of our owners' money on developing these projects, and they are going well, and we want to frankly make sure our owners know that their money is being well spent and that the projects are going well. Two, we will, and I will not go into detail, we will talk about some additional ideas that we have, that we want to highlight and surface, because there is some pretty interesting stuff." — Ammar Al-Joundi, President and CEO

What will not be there, on management's own account: the Detour full-potential study, which is a 2027 item and explicitly will not feed the February three-year guidance, and any commitment on 2027 costs or payout. What might be: a second Odyssey shaft framing, the Detour incremental underground case for 2028, and Hope Bay and Ikkari project returns the CEO offered to share "to the extent our general counsel allows us to."

Assessment: An investor day held on November 18, three weeks after a cost miss, on a stock that just fell 9.7% in two sessions, is a well-timed opportunity and also a risk. The pipeline detail is the strongest part of this story and a day spent on it should help. But the questions the market asked on this call were about 2027 unit costs and the payout, and management has said neither will be answered until February.

Guidance & Outlook

No numeric range moved. Three skews moved, all in the same direction: total cash costs and AISC to the upper end, and depreciation towards the upper end because the redesigned Barnat mine plan carries a higher units-of-production amortization rate on a smaller life-of-mine tonnage. One item improved: the payout target rose from approximately 40% of free cash flow to more than 50%, which the company has already delivered at 51% through nine months. Cost guidance embeds the $5 per ounce tariff impact and assumes 1.36 C$/US$, 1.18 US$/EUR, 1.40 A$/US$ and 17.50 MXN/US$.

Implied Q4 ramp. Nine-month production of 2,494,937 ounces against a full year "near the lower end" of 3.30 to 3.50 million implies a fourth quarter of roughly 805,000 to 855,000 ounces, in line with or slightly below the third quarter. The cost arithmetic is tighter than the production arithmetic. Using derived nine-month cost dollars of approximately $2,724M and $3,815M:

The total cash cost range has room for a further deterioration of roughly $70 per ounce. Management's figures imply an incremental diesel and tariff load of roughly $25 to $35 per ounce against the third quarter: diesel at $50 to $60 against budget versus about $30 already carried, plus $5 of tariffs. That leaves room on this line. The AISC range does not: even the top of it requires the fourth quarter to come in $35 per ounce better than the third. Against that requirement, the fourth quarter carries diesel at $50 to $60 per ounce against budget rather than $30, $5 per ounce of tariffs, and scheduled mill maintenance at seven of ten mills totalling 46 days, plus a 13-day extension at Detour Lake during which total mill throughput runs at approximately 60% of capacity.

Street at: the Street's modelled full-year AISC has been tracking near the midpoint; the only published operating estimate we could cross-reference carried $1,467 per ounce for the third quarter against $1,648 actual. Expect full-year AISC estimates to move towards and in some cases above the top of the guided range.

Guidance style: Agnico's convention is to hold numeric ranges and move the language inside them, and it has now done that twice in two quarters on production and once on costs. The company has not missed a guided range in the period we have covered. The honest read is that this is a credible guider that is running out of room on one line, and that the fourth quarter is the test rather than the formality the production number is.

Analyst Q&A Highlights

Eight questioners and no queue remaining when the operator closed. The distribution is informative: no questioner engaged the $1,648 AISC print itself, though one noted in passing that cost guidance had moved to the upper end of the $1,400 to $1,550 range, and the period was dominated by 2027 and the decade beyond it.

Labour Cost Inflation Versus Labour Availability

The first question of the call went straight to the gap in the release: diesel and consumables were quantified in detail and labour was not mentioned. Management separated the two, said wage inflation is not currently material, and identified availability of skilled people as the decade's constraint, with exploration drillers as the leading indicator. The answer was candid and it conceded that the company is investing significant effort in a problem it has not yet solved.

Q: "In the press release, there was a lot of commentary on the diesel cost and consumable cost inflation that you're seeing. I didn't see as much about labor cost inflation. Maybe you could touch on that as well, and how you're thinking about it for 2027."
— Fahad Tariq, Jefferies

A: "We're not seeing, at this point, material increases in labor costs. The bigger challenge is in availability of high-quality labor. The canary in the coal mine, as you know, you've been in this industry for a while, is usually on the exploration side with drillers, because they tend to be more transient. So internally, things are going pretty well. We continue to have, I think, a very good, very happy workforce, low turnover."
— Ammar Al-Joundi, President and CEO

Assessment: A useful answer to the right question. Wage inflation absent from a cost base rising 19% year over year is genuinely good news, because it means the increase is in commodities that can reverse rather than in a structurally higher wage bill. The availability problem is real but it is a 2030s execution risk on the growth pipeline, not a 2027 margin risk.

Whether the Odyssey Shaft Can Make Up the Sterilised Barnat Ounces

A recurring line of questioning on the call sought upside offsets to the Barnat sterilisation, and this was the most direct version: with shaft production starting in the second quarter of 2027, can Odyssey cover the up-to-150,000 ounce annual hole. The answer was a flat no with a small conditional, delivered in one sentence and not elaborated.

Q: "Just switching gears to Malartic for a moment, with the first shaft production expected in the second quarter of next year, do you expect any potential, I guess, production makeup for the lost Barnat ounces? Is that possible for next year?"
— Fahad Tariq, Jefferies

A: "No, we don't see something, let's say, concrete right now. If we're doing better, maybe, but I will not say it at this stage where we are with that."
— Dominique Girard, EVP and COO, Nunavut, Quebec and Finland

Assessment: This closes a door the bull case had left open since July. The 150,000 ounces removed from each of 2027 and 2028 are not being backfilled by the shaft ramp, which means the 2027 production guidance arriving in February starts from a base roughly 4.5% below what a pre-Barnat model carried. It also means the February guide and the 2027 unit cost are linked: fewer ounces over a largely fixed cost base.

What the 2027 Cost Base Should Be Built From

The dominant topic in the question period, raised in different forms by several participants, was how to model 2027 unit costs. Management refused to guide but gave the components: the third quarter already reflects most of the diesel and consumable inflation, diesel has roughly doubled against last year's budget assumption, and 5% to 6% is a reasonable overall inflation assumption against which productivity is the offset. The CEO then undercut the exercise by saying the largest variable is unforecastable.

Q: "Just on the cost, Jamie, should I be thinking that the rate you exit 2026, that higher cost rate and factoring in above average inflation for 2027 of maybe 5%, is what would be a good base for 2027 costing?"
— Tanya Jakusconek, Scotiabank

A: "I would say, if you look at the level of cost increases that we've seen, so Q3 of 2026 relative to Q1, that reflects a lot of the higher diesel costs and other inflation that we're seeing. ... Beyond that, I don't know where we're going to end up on labor and other consumables and everything else, but yeah, 5% or 6% is a reasonable assumption. Again, our job is to try to offset that. So we'll see how much of that cost inflation we can reduce through productivity and operational efficiency."
— Jamie Porter, EVP and CFO

Assessment: The most consequential exchange on the call. An exit-rate-plus-inflation construction puts 2027 total cash costs near $1,220 per ounce before productivity, roughly 14% above this year's guidance midpoint, and management did not dispute the method. The honest translation is that the Street should carry a 2027 cost base well above 2026 and that February will confirm rather than surprise. The company's three-year record of offsetting about half of input inflation is the only thing standing between $1,220 and a worse number.

Whether the 50% Free Cash Flow Payout Becomes a Standing Threshold

A recurring line of questioning pressed on whether this year's payout upgrade is a one-off reflecting a high gold price or a new floor. Management moved from "a reasonable threshold" on first asking to "I do not think it is an unreasonable assumption at this point" on second asking, and in both cases attached the gold price and the capital programme as conditions, with the decision deferred to the February guidance release.

Q: "Then I guess on capital allocation, with now targeting above 50% of free cash flow towards shareholder returns this year, given the strong financial position, should we view this as maybe a threshold that's likely to extend into 2027?"
— Bennett Moore, JPMorgan

A: "I think we're well on track. On a year-to-date basis, we've returned 51% of our free cash flow, and we now target being over 50% for the full year. I think 50% is a reasonable threshold, looking ahead to 2027. But again, that's something that as we work through our budgeting process and have discussions with our board, that we'll ultimately decide on and publish in conjunction with our February guidance release."
— Jamie Porter, EVP and CFO

Assessment: A 50% payout on a shrinking free cash flow base is a smaller number than a 40% payout on this year's. Our model has full-year 2026 free cash flow at approximately $3,837M, so 50% is roughly $1,920M of returns; a 2027 with 150,000 fewer ounces, higher unit costs and higher capital generates materially less. The payout percentage is the headline and the free cash flow base is the number that matters, and the immediate follow-up from the CEO that "it will also be a function of the gold price" is the honest framing of both.

Whether the 2028 Production Outlook Is Already Stale

One line of questioning tried to extract upside from the Detour Lake underground story, on the logic that several quarters of positive commentary about a larger underground mine starting as early as 2028 must make the existing outlook conservative. Management confirmed two separate teams are working on it, one on near-term incremental underground ounces and one on the full-potential case, and then ruled out any of it reaching the February three-year guidance.

Q: "Could we see some early results from the study we're expecting in mid-2027 on Detour factored into the three-year guidance that we will get in February?"
— Lawson Winder, Bank of America Securities

A: "No. It's still very early stages on that, Lawson."
— Natasha Vaz, EVP and COO, Ontario, Australia and Mexico

Assessment: The shortest answer of the call and one of the most useful. The Detour full-potential case, which could include a larger underground mine, a pushback of the open pit and higher milling capacity, is a 2027 disclosure and will not appear in February. Anyone underwriting near-term upside from it is underwriting a date, and the date just moved further away than the question assumed. The 2028 incremental underground tonnes from the shallower zone remain live and are a function of a bulk sample completing in the first half of 2027.

Why the Credit Facility and Shelf Were Upsized at $3.4 Billion of Net Cash

The most pointed question of the call asked what a company with a net cash position and almost no debt wants with a 50% larger revolver and a fresh shelf prospectus. Management gave a liquidity-management answer and volunteered the forward capital number inside it. The CEO then added an unprompted denial that anything should be read into it.

Q: "Then just one on the balance sheet. You finished the quarter with $3.4 billion net cash. So what is the thinking around increasing the size of the credit facility and the shelf prospectus? Is this just preparation for peak CapEx with so much going on?"
— Matthew Murphy, BMO Capital Markets

A: "Really just giving us the liquidity that I think makes sense for a company of our size and scale. Our capital spending is going up. We'll be north of $3 billion this year in total capital, and that's likely to increase over the next few years. But we're fortunate in that we're very profitable, and we're going to be able to fund that from operating cash flow and have the balance sheet very defensively positioned. So if the gold price were to drop meaningfully, we've still got the financial wherewithal to be able to move forward."
— Jamie Porter, EVP and CFO

Assessment: Two pieces of genuinely new information came out of a question about plumbing. Total capital rises from above $3 billion over the next few years, which is the first forward capital framing the company has given beyond 2026. And the stated purpose of the liquidity is to withstand a meaningful gold-price decline while continuing to build, which is the correct priority and also an acknowledgement of where the risk sits. The follow-on comment that nothing should be read into it is the kind of denial that keeps the question alive ahead of the November 18 investor day.

Whether the Returns on the Capital Being Spent Justify It

The sharpest challenge of the call came as arithmetic: roughly $3 billion committed at Hope Bay and potentially more than $6 billion in Finland by the time a mine is built, against the entire balance sheet of a much smaller developer, so either the production targets are too low or the returns are poor. Management declined the comparison and answered with a return range instead.

Q: "At Hope Bay, with the CapEx plus the monies already spent, Agnico will have spent over $3 billion. For Lapland, the acquisition was $4 billion, so that by the time you build a mine, it might be over $6 billion. Should I interpret that you're going to produce three times more gold than you're indicating at Hope Bay and Lapland to justify the CapEx, or should I interpret that the company with the $2.4 billion balance sheet will have five times your ROE, or both?"
— John Tumazos, John Tumazos Very Independent Research

A: "If we can build projects making somewhere between 25% and 60% return on capital, we will do them till the cows come home. If you can find somebody at 80%, invest in that. But we are pretty happy at the 25%-60%."
— Ammar Al-Joundi, President and CEO

Assessment: The question deserved a better answer and the answer contains a loose end. The prepared remarks put project returns at approximately 25% to 50% at current spot prices; the response widened the top of the range to 60% without a basis. Neither figure has published supporting economics, and the CEO's offer to share the numbers "to the extent our general counsel allows us to" is the tell that they exist and are not public. Project-level returns at Hope Bay and Ikkari are the single most valuable thing the Capital Markets Day could disclose.

The Reserve Price and the Cut-Off Grade

The last question of the call asked where the reserve gold price assumption will land for the year-end statement. The answer avoided the number and gave the governing constraint instead, which is arguably more useful because it determines what the reserve statement will look like rather than what one input will be.

Q: "I just wanted to ask about reserve resource pricing as you go into next year. Do you have an idea of where that might land at this stage?"
— Anita Soni, CIBC World Markets

A: "As discussed, we are taking into consideration some of the inflation into our input on fuel and the rest of it. We are in the process of finalizing our view, but our objective remains the same, to keep the cut-off grade stable. Whatever we are going to do is not going to drive cut-off grade lower and it's going to impact our view of what we want to mine in the next few years."
— Yves, Agnico Eagle

Assessment: Committing to a stable cut-off grade while input costs rise means the reserve price must rise, and it means the company is choosing not to convert a higher gold price into lower-grade tonnes. That is the conservative choice and the right one, and it is also the reason February's reserve statement should not be read as a reserve-growth event even if the ounce count rises. The closing clause matters most: the reserve work will change what the company plans to mine over the next few years, which is a mine-plan statement disguised as an accounting one.

What They're NOT Saying

1. No quarterly cost guidance is published, yet the quarter was graded against one. The CEO said cash costs came in "1% above the top end of our cost guidance" and production "at the midpoint of our adjusted production guidance for the quarter." Neither quarterly guide is public. The company is grading itself against a yardstick investors cannot see, and in a quarter where that yardstick produced a miss, it is the only reason we know the miss was small.
2. The AISC breach of the full-year range is never named as such. The release reports $1,648 per ounce and separately reaffirms a $1,400 to $1,550 range. Nowhere do the two sit in the same sentence, and no reconciliation of the quarter to the range is offered. One questioner referred to the shift to the upper end of that range and asked about input cost pressures; nobody asked how a quarter $98 above the top of it is consistent with landing inside.
3. No 2027 cost guidance, no 2027 production guidance, no three-year plan, and the date moved. All of it defers to February. The November 18 Capital Markets Day will not carry it. That is a normal calendar and an abnormal amount of forward uncertainty to carry into it, given diesel at $50 to $60 per ounce against budget and 150,000 ounces removed from 2027.
4. The sustaining capital step-up is not explained at the consolidated level. $360 per ounce against $282 a year ago and $286 last quarter is 24.6% of the AISC increase. The release attributes it to Detour Lake and Fosterville and stops. There is no statement about whether $360 is a timing artefact or the new run rate, and no sustaining capital guidance exists to check it against.
5. Hope Bay detailed engineering is given three different ways. Approximately 73% in the release highlights, approximately 76% in the release's project section, and 75% on the call. The sealift is 207,000 cubic metres in the release and 220,000 on the call. These are small, but a company whose core claim is disciplined execution should not publish two different completion percentages for its largest project in one document.
6. The $5 per ounce tariff is described without naming the tariff. The release calls it Canadian retaliatory tariffs; the question on the call called it US retaliatory tariffs; the answer addressed neither. The CFO said "we would actually expect that impact to increase" absent a trade deal, without sizing the increase.
7. The project return range moved and nobody reconciled it. Prepared remarks said approximately 25% to 50% at current spot. The CEO later said 25% to 60%. No supporting economics are published for either, and the offer to share them was conditioned on legal review.
8. The Nunavut diesel bill got worse and the call did not discuss it. Last quarter the company had bought roughly 130 million litres, about 70% of its requirement through the 2027 sealift, at prices roughly 30% above budget. This release reports the purchases complete at approximately 185 million litres for the balance of 2026 and through the 2027 sealift, at prices approximately 45% higher than budgeted net of hedges, and a fourth-quarter production cost impact of approximately $21M. That blended figure is above the roughly 30% disclosed for the first tranche, so the rest was bought at a higher price. Neither the step-up nor what it implies for 2027 Nunavut costs came up on the call.
9. The Detour tailings line breach gets four sentences and no cost. A seven-day unplanned mill shutdown and a release of tailings and process water on August 29, with most solids retained in secondary containment, limiting impacts to Easter Creek, with no remediation cost, no regulatory status and no statement on whether the secondary containment strengthening is complete.
10. Safety is absent. After two quarters in which fatalities were addressed at length and ahead of the financials, this call contained no safety section and no safety update. Silence after three fatalities in twelve months is better than bad news, and it is still a change in disclosure pattern that was not explained.
11. The contingent value rights remain unvalued. 207,654,166 rights outstanding from the Rupert acquisition, conditional on Ikkari reserve and production milestones, with no framework for what they are worth or when they could trigger. This is the second consecutive quarter the count is disclosed and the value is not.

Market Reaction

• Pre-print setup: AEM closed at $183.11 on October 28, up 8.0% year to date from $169.53, down 0.6% over the trailing 30 days from $184.15, and up 16.8% over the trailing twelve months from $156.79. The 52-week closing range entering the print was $136.34 to $252.19. The S&P 500 was up 13.1% year to date over the same stretch.
• The stretch since our last note: AEM rose 26.8% between the two pre-print closes, from $144.43 on July 29 to $183.11 on October 28, while gold spot was effectively flat at +0.4% and the VanEck Gold Miners ETF rose 14.1%. AEM outperformed its own sector proxy by roughly 1,270 basis points over that window, reversing the 580 basis points of underperformance we measured in July.
• Reaction session: AEM opened at $177.00 on October 29, a 3.3% gap lower, traded between $173.70 and $179.29, and closed at $175.92, down 3.9% or $7.19. Volume of 2.1 million shares was 1.2 times the 30-day average. The S&P 500 fell 0.3%.
• Peer and commodity context on the reaction session: gold spot fell 0.5% to $4,031.40, the gold-miner ETF fell 3.1% to $81.33, and Newmont fell 1.9% to $111.80. AEM underperformed the ETF by roughly 81 basis points and Newmont by roughly 207.
• Follow-through session: on October 30 gold closed at $3,898.90, its lowest close of the trailing twelve months and the first sub-$4,000 close since July 16. AEM fell 6.0% to $165.38 on 4.9 million shares, against the ETF down 5.2% and Newmont down 4.6%. Two-day cumulative: AEM down 9.7%, the ETF down 8.2%, gold down 3.8%.
• Where it trades now: gold recovered to approximately $3,955 on November 2 and AEM closed at $170.41, up 3.0%, leaving the stock 6.9% below its pre-print close and 32.4% below the 52-week closing high.

The reaction is best read in two parts, and only the first part is about Agnico. On October 29, with gold down half a percent, the stock fell 3.9% and gave up roughly 80 basis points against its sector proxy and 200 against its largest peer. That is the market's price for the print, and it is a small number for a quarter in which AISC printed $98 above the top of the guided range. The second part, the 6.0% decline on October 30, is a commodity event: gold broke below $4,000 to a twelve-month low and the whole complex sold, with AEM giving up another 78 basis points of relative performance in the process.

What makes this quarter's reaction interesting is the setup rather than the response. In July the stock entered the print down 21.3% from our prior note and 42.7% below its 52-week high, so the question was whether the print contained a fresh negative. This time the stock entered up 26.8% from the same reference point, two dollars from a 30-day high, having outperformed its sector by nearly thirteen percentage points while the metal went nowhere. That is positioning that requires the print to be clean, and this print was not clean on the one line the market cares about in a gold miner at a cycle-high price: unit costs.

The thing we said to watch in July has happened, and it happened in the opposite direction from the one that would have justified holding the rating. We wrote that the reaction to watch was "whether the second consecutive quarter of production above budget starts to compress the discount at which this equity trades to spot gold," and that so far it had not moved at all. Between July 29 and October 28 it closed completely and then inverted. The market spent three months re-rating Agnico Eagle while the commodity did nothing, which is exactly the opposite of the previous quarter and removes the specific asymmetry the Outperform was built on.

Street Perspective

Debate: Is the Cost Breach a Diesel Event or a Structural Reset?

Bull view: The bull case being made on the Street is that the entire miss is identifiable, external and reversible. Diesel contributed about $30 per ounce against budget, tariffs $5, royalties scale mechanically with a gold price that is itself the reason earnings beat, and the comparative base is restated onto a composition that inflates the increase. The company still guides inside both ranges and has not missed a guided range.

Bear view: The bear camp contends that diesel explains roughly a sixth of the year-over-year total cash cost increase and nothing at all of the $136 per ounce that sits above the cash cost line in AISC. Sustaining capital of $360 per ounce, non-cash reclamation up $50, every one of six regions more expensive than a year ago, and eight of ten mines more expensive, is not a fuel story. It is what a ten-mine portfolio costs while four projects are built through the same organisations.

Our take: The bears have the better of this. Diesel is the loudest component and the smallest surprise; the quiet one is capital intensity, which does not reverse with the oil price and which management has now guided higher for several years. The useful framing is that approximately 57% of the AISC increase is operating cost, of which roughly half is commodity-driven and reversible, and 43% is capital and non-cash items that are not.

Debate: Does the Pipeline Justify a Premium Multiple Through the Spending Years?

Bull view: Some sell-side desks argue this is the best-funded, lowest-risk growth pipeline in large-cap gold, and that a company adding 1.3 to 1.5 million ounces a year by the mid-2030s at 25% to 50% returns, from brownfield positions in Canada, with net cash and an A-range credit profile, should trade at a premium while it builds. Every project is on or ahead of schedule and the balance sheet was just made larger than it needs to be.

Bear view: The skeptics note that the growth is 2030 and beyond, that the near-term profile is going backwards with 150,000 ounces removed from each of 2027 and 2028 and explicitly not backfilled, and that the spending years are exactly when free cash flow per share falls. Free cash flow per share is already down 29% year over year and total capital rises from above $3 billion. Paying a premium multiple for ounces that arrive in 2030 while the per-share cash flow declines until then is a long-duration bet dressed as a quality one.

Our take: Both describe the same cash-flow trough from different ends. We keep the premium multiple at 8.7 times because the jurisdictional and balance-sheet case for it has not weakened. What has changed is that the trough is now visible in the reported numbers rather than projected, and the market is paying for the far side of it in advance.

Debate: Does the February Guidance Release Contain a Negative Surprise?

Bull view: The bull case is that February holds no surprises because the components are already public: the exit cost rate, a 5% to 6% inflation assumption, a quantified Barnat impact, and a company with a three-year record of offsetting roughly half of input inflation through productivity. A stable cut-off grade commitment means the reserve statement should be clean.

Bear view: The bear camp contends that February has to deliver four things at once that all point the same way: 2027 production lower on Barnat, 2027 unit costs materially higher on diesel and labour, capital higher again, and a reserve price revision. On the Street's own arithmetic, exit-rate-plus-inflation puts 2027 total cash costs near $1,220 per ounce against a 2026 guidance midpoint of $1,070, and management did not dispute the method when it was put to them.

Our take: The bears are right about the direction and probably wrong about it being a surprise. A 14% step-up in the 2027 cost base is the arithmetic consequence of disclosures already made, so the information is in the market even if the guidance is not. The risk is not the number but the compounding: a lower production guide and a higher cost guide and a higher capital guide, landed together, read worse than any one of them alone.

Debate: Is a 7% Pullback From the Pre-Print High an Entry Point?

Bull view: A growing consensus view is that nothing structural broke: the company beat on earnings and revenue, reaffirmed both guided ranges, raised the payout target, grew net cash through its heaviest spending quarter, restarted Barnat a quarter early, closed the Kittila recovery problem, and has an investor day on November 18 to reframe the growth story. Two post-print target increases came with unchanged neutral ratings, which is the Street marking its gold deck to market rather than changing its view.

Bear view: The bear camp points out that the stock is still 18% above where it traded three months ago on a business whose unit costs have deteriorated in every quarter since, that gold just printed a twelve-month low, and that the equity has lost more than $40 billion from its 52-week high and is still not cheap on forward numbers. On a forward-twelve-month basis at the current spot price the business supports roughly $151 per share against a $170.41 close.

Our take: The bears have the valuation and the bulls have the business. Our framework says the current price requires a forward realized gold price near $4,300 per ounce, 8.7% above spot, where three months ago it required one 8.5% below spot. The pullback is not an entry point; it is a partial unwind of a run that already discounted a better commodity than the one that exists.

Model Update & Valuation Framework

Changes to Our July Estimates

FY2026E Assumptions

Two assumptions carry most of the risk and they are the same two as last quarter with the signs changed. The first is the fourth-quarter gold price, held near the October average rather than at any forecast; a $100 per ounce move is worth roughly $84M of EBITDA over one quarter. The second is the fourth-quarter AISC, where we model $1,610 per ounce against the $1,613 that the very top of the guided range allows. That is a deliberately tight assumption and it is the one most likely to be wrong in the unhelpful direction: it requires a $38 per ounce sequential improvement in a quarter carrying more diesel, a tariff, and 46 days of scheduled mill maintenance across seven mills plus a 13-day Detour extension at 60% throughput. If sustaining capital holds at the third quarter's $360 per ounce rather than easing, the full-year AISC lands above $1,550 and the company misses a guided range for the first time in our coverage.

Valuation and Price Target

At the November 2 close of $170.41 on 505.5 million shares outstanding net of shares held in trust, the market capitalisation is roughly $86.2B. Deducting $3,416M of net cash gives an enterprise value of roughly $82.7B, or 7.9 times our FY2026E adjusted EBITDA of $10,451M. That compares with 6.8 times at the July print and 7.6 times at initiation in May. On our estimates the free cash flow yield is 4.5% and the adjusted price-to-earnings ratio is 14.9 times.

We cut the 12-month price target to $160 from $175, roughly 6% below the current price. The multiple is unchanged at 8.7 times enterprise value to adjusted EBITDA plus net cash, which we continue to regard as a defensible premium for jurisdictional concentration, net cash and a funded growth profile. The target again sits between two anchors. Applying 8.7 times to FY2026E adjusted EBITDA, which contains three quarters of realized prices that will not repeat, produces $187. Applying the same multiple to a forward twelve months at the current spot price and the higher forward cost base produces $151. We weight the forward anchor 75% because ten months of 2026 are banked; last quarter, with five months remaining, we weighted it closer to half.

Gold Price Sensitivity

The table runs the forward-twelve-month version of the framework. Production is held at 3,300 thousand ounces, which allows for the Barnat effect in 2027 that management confirmed will not be backfilled. Total cash costs are based at $1,215 per ounce at a $4,000 realized price, built from the fourth-quarter 2026 exit rate plus the 5% to 6% inflation the CFO called a reasonable assumption, net of a modest productivity offset. Royalty and NTI costs absorb roughly 2.5% of the change in revenue in each direction, so the cost line moves with the price. Fixed costs of $645M comprise G&A, exploration expense and other costs at roughly 4% above the 2026 guidance midpoints. The multiple is 8.7 times.

Thesis Scorecard Post-Earnings

The table scores the same six pillars established in May and carried through July. One status tag moves this quarter.

Overall: Thesis weakened, and specifically on the pillar we had just upgraded. In July we moved Bear 1 from emerging to contained on the strength of a single quarter with both cost measures below their guidance midpoints, and we wrote that it was contained rather than resolved because roughly half the cushion was by-product credits and currency. One quarter later the cushion is gone, the deterioration again spans all six regions, as it did in the first quarter, and the AISC line has breached the top of its own full-year range. The three bull pillars are each in better shape than in July, and the growth pipeline in particular is advancing faster than scheduled. That is not enough. In a gold producer at a cycle-high metal price, the unit cost line is the only variable management controls, and it moved the wrong way in every region at once.

Action: Trim into strength; hold a core position for the 2030s pipeline. The rating moves to Hold because the two legs of the Outperform both weakened in the same quarter: the demonstrated ability to hold the cost guide, and an 8.5% discount to spot gold that has become an 8.7% premium. Neither the asset base nor the balance sheet nor the pipeline is the reason. This is a price and a cost call. The level that would restore the rating is a forward implied gold price back at or below spot, which from here means roughly $150 per share with gold unchanged, or a February guidance release that holds 2027 total cash costs below $1,180 per ounce. Treat a sustained break of $3,700 per ounce in spot gold, now 6.4% below the November 2 level, as the point at which the framework stops supporting even a Hold.

Commitments to Watch Next Quarter

1. Fourth-quarter AISC at or below roughly $1,613 per ounce if the top of the $1,400 to $1,550 full-year range is to hold, against $1,648 in the third quarter. The midpoint is gone. This is the single most important number in the February release, and it requires sequential improvement in the quarter with the most mill downtime of the year.
2. Fourth-quarter total cash costs at or below roughly $1,204 per ounce to hold the top of the $1,020 to $1,120 range, against $1,133 in the third quarter. There is roughly $70 per ounce of room against roughly $25 to $35 of incremental diesel and tariff load relative to the third quarter, so the range should hold unless other cost lines also deteriorate.
3. Whether sustaining capital per ounce retreats from $360 towards the $296 nine-month average, or whether $360 is the new run rate. This is 24.6% of the year-over-year AISC increase and the part that does not reverse with the oil price.
4. The November 18 Capital Markets Day: whether project-level returns for Hope Bay, Ikkari, Upper Beaver and the Detour underground are published with supporting economics, and whether the 25% to 50% and 25% to 60% return ranges are reconciled.
5. 2027 production guidance in February against a base reduced by up to 150,000 Barnat ounces that management confirmed will not be backfilled by the Odyssey ramp.
6. 2027 total cash cost guidance in February. An exit-rate-plus-inflation construction implies roughly $1,220 per ounce. Anything below $1,180 would be a genuine productivity result; anything above $1,250 breaks our forward framework.
7. The year-end reserve statement: the reserve gold price, whether the cut-off grade is genuinely held stable, how the 370,000 sterilised Barnat ounces are treated, and whether the Detour Domain 54 and East Gouldie deep intercepts convert.
8. Whether the payout target above 50% of free cash flow is formalised for 2027, and on what free cash flow base, given total capital rising from above $3 billion.
9. Buyback execution relative to price. The company has now paid above the period average close in one of two measurable quarters and above the quarter-end price in both. Watch whether the fourth-quarter average is below the quarter's own average close.
10. Meadowbank grade recovery in the fourth quarter after the IVR pit model discrepancy, and whether the updated geological model changes the 2027 plan for the complex.
11. Detour Lake fourth-quarter throughput through the 13-day extended SAG mill gear replacement at approximately 60% of capacity, and confirmation that the site still meets its production guidance.
12. Kittila recovery holding at or above 85% through the 17-day autoclave relining shutdown, and whether a formal recovery target is finally published.

Bottom Line

Agnico Eagle delivered almost everything we asked of it in July, on the company-specific list. Barnat remediation finished and mining restarted on September 27, ahead of the date promised. Kittila's recovery problem closed, with September the best monthly recovery in five years. The Boston assays arrived in the October release and were good. The Odyssey shaft finished its first phase four months early and the headframe changeover began without disturbing the second-quarter 2027 commissioning date. Hope Bay's sealift landed complete and construction is slightly ahead of plan. The Rupert share count stepped up 0.88% into the weighted average, almost exactly as forecast. Net cash grew again. The payout target rose.

And the cost line broke. Total cash costs of $1,133 per ounce came in above the top of the company's own internal quarterly guide, which the CEO said in his second sentence. AISC of $1,648 came in $98 above the top of a full-year range the company still says it will land inside, and holding even the top of that range now requires the fourth quarter to improve $35 per ounce in the quarter carrying the heaviest diesel load, a tariff, and 46 days of scheduled mill maintenance across seven mills. Both full-year midpoints are arithmetically out of reach. Every one of six regions and eight of ten mines produced gold more expensively than a year ago, a breadth last seen in the first quarter and reversed in the second. Free cash flow fell 28.4% while net income rose 28.2%, and free cash flow per share fell 29.1%.

Diesel is the loudest cause and it is not the most important one. It added about $30 per ounce against budget this quarter, rises to $50 to $60 next, and may reach $80 for the year, which is real money and roughly a sixth of the year-over-year increase in total cash costs. The $136 per ounce sitting above the cash cost line in the AISC bridge is the part that matters: sustaining capital at $360 per ounce against $282, non-cash reclamation up $50, G&A up $8. That is not a fuel price. That is what maintaining ten mines costs while building four projects through the same organisations, and the CFO said total capital rises from above $3 billion for several years.

The stock, meanwhile, did precisely the thing we said in July it had failed to do. We wrote that the question was whether two consecutive quarters of production above budget would compress the discount at which this equity trades to spot gold, and that so far it had not moved at all. It moved. AEM rose 26.8% between the two pre-print closes while gold was flat and the sector rose 14.1%. Then it kept going, and the 8.5% discount to spot that we measured at $183.56 in May and again at $144.43 in July is now an 8.7% premium at $170.41. The market closed the gap and overshot it, in the same three months in which the company's unit costs deteriorated in every region.

We are downgrading to Hold and cutting the target to $160. The Outperform rested on two legs and both weakened in the same quarter. This is not a verdict on the asset base, which remains the best-located in the industry; nor on the balance sheet, which is stronger again; nor on the pipeline, which is advancing faster than scheduled and is the reason to own this company in 2030. It is a verdict on a price that now requires a gold price 8.7% better than the one that exists, on a cost base that has lost the cushion it had in July, with no 2027 guidance until February and four pieces of that guidance all pointing the same way. We would rather own this at $150 with the discount restored, or at $170 with a February guide that proves the cost base can be held. Neither exists today.