The miner digs the ounce. The streamer paid for it first.

The price did the work in the first half, then stopped

Gold's first quarter of 2026 was the best on record for anyone selling it.

A U.S. investor who wants gold exposure through a Canadian stock has two obvious kinds of company to choose from, usually discussed as if they were the same bet at different volumes. They are not. Agnico Eagle digs ounces out of the ground; Wheaton Precious Metals bought the right to ounces other companies will dig, years before the digging. Both report in U.S. dollars and both trade on the NYSE, so the comparison is clean on currency and listing [8][14]. It is not clean on anything else.

What Agnico Eagle owns: ore, mills and a cost line that moves

Agnico Eagle operates ten mines across Quebec, Ontario, Nunavut, Australia, Finland and Mexico [4]. At December 31, 2025 its proven and probable reserves were 55.4 million ounces of gold at a grade of 1.30 grams per tonne [5]. That is the inventory. Turning it into metal costs money every year, and the cost is the thing to watch.

Total cash costs rose from US$979 an ounce for 2025 to US$1,093 in the first quarter, and AISC from US$1,339 to US$1,483 [5][6]. Gold mines pay for labor, diesel and contractors in local currencies, and a record price has brought rising bills. The 2026 guidance of US$1,020 to US$1,120 in total cash costs and US$1,400 to US$1,550 in AISC says management expects that to continue, with production of 3.3 million to 3.5 million ounces a year through 2028 [5].

Then there is capital. Agnico guided 2026 capital expenditures to US$2,175 million to US$2,395 million excluding exploration [5], and on May 19, 2026 approved construction of Hope Bay in Nunavut: about US$2.4 billion of initial capital for 400,000 to 435,000 ounces a year over an initial 11-year life, with first gold possible as early as 2030, on economics struck at US$4,500 an ounce [7]. That is a producer's version of buying an ounce in advance: money in now, ounces in four years, return set by a price nobody controls.

What Wheaton owns: contracts on other people's mines

Wheaton's own description of its business is short. It purchases a percentage of the metals produced by a mine for an upfront payment plus an additional payment on delivery, with the delivery payment generally below the prevailing spot price [9]. At March 31, 2026 it had such agreements on 22 operating mines and 26 development and other projects [10]. It owns no shovels.

The first quarter shows what those contracts throw off at a high price. Attributable production was 211,951 gold equivalent ounces (GEOs, the company's conversion of silver and other metals into gold at prevailing prices), of which 97,106 ounces were gold and 6.636 million ounces silver; sales were 181,743 GEOs at a realized gold price of US$4,849. The average cash cost was US$681 per GEO, leaving a cash operating margin of US$4,279 per ounce sold. Revenue was US$901 million, net earnings US$582.044 million, and general and administrative expense US$12.971 million [10].

A head office that costs US$12.971 million a quarter is a very different object from ten mines. The balance sheet shows where the money went instead. Wheaton held US$2,164.505 million of cash at March 31, 2026, the day before it paid BHP, and funded Antamina with a US$1.5 billion term loan and a draw on its US$2.0 billion revolving facility [10]. The capital a producer spends year by year on sustaining its mines, the streamer spends in a lump at signing, and in this case borrowed to do it.

How each earns from the same ounce

Take one ounce of gold sold in the first quarter at roughly the same price by each company. Agnico keeps the realized price less what it cost to mine, process and sustain the operation, which AISC puts at US$1,483 [6]. Wheaton keeps the realized price less the delivery payment written into the contract, which across its portfolio averaged US$681 per GEO [10]. On the face of it the streamer keeps more of every ounce.

That is the wrong place to stop. Agnico's US$1,483 buys the ounce and the right to the next one: it includes the sustaining capital that keeps the mine producing and the exploration that replaces reserves [5]. Wheaton's US$681 buys only this ounce; the right to the next one was paid for in 2007, 2013 or April 2026, and that money is not in the cost line [11][12][13]. What a stream transfers is cost risk, which moves to the operator because the delivery payment is fixed, while production risk stays with Wheaton: if Vale mines less at Salobo, Wheaton receives less, with no cost it can cut [9][11]. An honest comparison charges the streamer for its upfront capital over the life of each contract and the producer for its growth projects, and only then compares margins.

What the gold price does, and does not do, for each

What the price does not do for the streamer is replace the ounces. Wheaton's contracts run for the life of each mine, and the ounces that arrive depend on reserves, grades and operator decisions it does not control [9][11][12]. Its forecast of about 1,200,000 GEOs a year by 2030, from 2026 guidance of 860,000 to 940,000, rests on projects other companies must build [10]. The gold price can make every delivered ounce worth more; it cannot make Vale or Newmont deliver one more.

What the price does not do for the producer is hold its costs still. Agnico's own guidance has 2026 costs above 2025 costs, and the first-quarter figures came in above the 2025 average [5][6]. A producer at a record price is a producer whose suppliers and workforce know it is a record price.

How a U.S. reader should line them up

The practical points first. Both trade on the NYSE under the same symbols as on the TSX, AEM and WPM, so there is no over-the-counter step [8][14]. Both pay quarterly dividends in U.S. dollars, US$0.45 a share at Agnico and US$0.195 at Wheaton as of their latest declarations, and Canadian withholding tax is deducted for a non-resident holder in a taxable account [5][8][10][14].

The comparison itself needs the same date, the same currency and the same treatment of capital. Enterprise value for Agnico should reflect its net cash of US$2,915 million at March 31, 2026 [6]; for Wheaton it should reflect the debt taken on the next day to pay for Antamina, not the cash that was about to leave [10]. Cash flow for Agnico is already after sustaining capital, because AISC includes it; cash flow for Wheaton should be taken after an allowance for the upfront payments that produced it, which no reported figure does. Hope Bay's US$2.4 billion and Antamina's US$4.3 billion are both bets on ounces not yet delivered, made at prices that were recent records [7][13].

What would change our view

For Agnico, a second-quarter report with AISC above the top of the US$1,400 to US$1,550 range, or a change to the Hope Bay capital estimate, would tell us the cost line is running away from the price [5][7]. For Wheaton, the test is deliveries: the first quarters of Antamina silver under the BHP contract against the 33.75% entitlement, and the price those ounces fetch, which was US$84.52 in the first quarter [10][13]. For both, the gold price itself is the least informative input. It was US$5,405 on January 29 and US$4,546 at the end of May [2], and the ore at Agnico's mines and the contracts in Wheaton's portfolio were the same on both dates.