Sprott said it in plain language on October 7, 2026. Gold miners have entered a new era, one defined by stronger balance sheets, greater capital discipline, and record earnings, and valuations still reflect the industry's past. The author was Kenny Zhu, a research director at Sprott Asset Management. The piece is called "Gold Mining 2.0." It is a sector note. It does not name five stocks to buy. The headline that pairs his sentence with a shopping list is doing work he did not do.
The work worth doing is smaller. Zhu's claim is that the last boom's bad habits have been paid down, and the share prices have not noticed. A second Sprott voice, John Hathaway, said something that fights the first if you let it. The big producers have sat on more than a billion dollars of cash for years. He called those lazy balance sheets. Cash that sits there loses purchasing power. Cash that gets spent on the wrong mine repeats the last cycle. A stock to watch, in that argument, is not a stock Sprott blessed. It is a stock where you can see which of those two uses is actually happening.
Eric Sprott is not the author of either note. He is the man whose name is on the door, and he has been writing checks to juniors. People who ask why Eric Sprott is bullish on gold miners are usually asking about the man and answering with the firm, or the other way around. Those are different bull cases. Mixing them is how a research PDF becomes a tip.
This is not a recommendation to buy or sell any miner, any fund, or the metal. Sprott manages gold funds. A firm that sells the exposure has a reason to describe it kindly. Read the filings anyway.
What Zhu actually counted
The numbers in the October 7 note are index numbers, not company numbers. On the NYSE Arca Gold Miners Index, twelve-month earnings expectations were about $49.60 at the end of 2023. By September 30, 2026, they had reached $219.77. That is nearly fourfold. Forward enterprise value over EBITDA had compressed to 5.7 times, from a peak of 9.9 times in June 2014. Zhu's point is the mismatch. Earnings expectations exploded. The multiple did not. In a normal story, a fourfold rise in expected earnings pulls the multiple up. Here the multiple shrank over a decade, and the new earnings arrived into a market that still will not pay the old peak multiple.
He ties the earnings to a change in behavior. Growth at any cost, he writes, has been replaced by capital discipline and healthier balance sheets. Record earnings have been aimed at paying down debt and paying owners. Many producers now hold net cash. Debt has declined. Returns on capital and shareholder distributions are at multi-decade highs, and he allows that some of those marks may be all-time highs. The reliance, in his telling, is on cash the mines generate, not on a new loan.
That is a real change if the charts are right, and the charts are the firm's. A reader who does not open the underlying index data is trusting Sprott's picture of Sprott's market. Trust is fine as a start. It is a bad place to stop, because the firm runs the Sprott Gold Miners ETF and the Sprott Gold Equity Fund. The note is an argument that the thing they sell is misunderstood. Arguments like that can be true. They are not neutral.
Zhu also dates the market's suspicion. After the 2011 peak, aggressive building and deal-making erased about $129 billion of shareholder value from 2012 to 2017, a figure he sources to McKinsey. Investors who lived through that do not hear "record earnings" as a new religion. They hear a cycle. If the discipline is only the discipline of a high gold price, the multiple should stay low. If the discipline survives a flat gold price, the multiple is the mistake. Zhu thinks the second reading is available and unused. He also warns, in the risk language around the piece, that precious metals carry headline risk, political risk, and the risk that the improvement is temporary. The warning is the adult sentence. The slogan is the other one.
The rally that makes the sentence feel urgent
The note is not written from a quiet tape. It says a sharp second-quarter correction took the miners down about 25.3 percent from a March 2 peak to a July 16 trough, on energy shocks, inflation, and a stronger dollar. Then August brought a surprise rally of about 30 percent, which the note also describes as a climb of about 32 percent in gold mining stocks that month. The spark, in Sprott's telling, was fiscal fear. Deficits. Debasement. A softer dollar. A Treasury that was buying back long bonds. The kind of story that does not need a new mine. It needs a loss of faith in paper.
Hold that next to this week's price. Gold in early October 2026 has been a little above $4,100. One close on October 7 was $4,111. The late-January high was an intraday print near $5,589 and a closing high near $5,405. The metal is down about a quarter from that spike. The August rally in the shares did not repeal that. It sat inside it. A reader who hears "record earnings" and pictures a gold price still at the January high is using a stale picture. The earnings power was built at very high prices. The price now is lower than the second-quarter average near $4,512, and it is still far above the cost of producing an ounce at a major.
That gap is the whole cash story. It is also the part that can close without any change in "discipline." John Hathaway, at the Precious Metals Summit in Beaver Creek, said the industry is more profitable than at any time in his experience since 1998. He also said the quiet part. If gold marks time in the mid-$4,000s for two or three years, margins get squeezed, because costs inflate. He put the real cost pressure near 8 to 10 percent, from capital spending and labor, not from a central bank's 2 percent target. Paul Harris, the interviewer, offered a sector snapshot Hathaway did not himself calculate. Average all-in sustaining cost around $1,900, and a margin of roughly $2,000 to $2,400 an ounce. Hathaway's reply was the squeeze, not the celebration. Record cash is a photograph. Inflation is a process. A process wins if the price stands still.
Why Eric Sprott is the wrong answer to this headline
Search interest asks why Eric Sprott is bullish on gold miners. The honest answer starts with a correction. The October 7 sentence was not his. Eric Sprott built the firm, left the day-to-day work years ago, and has spent the recent bull market as a private buyer of small names. In 2026 his checks, as disclosed in company releases, have gone to places like Cerrado Gold, Silver Storm, and Goldgroup. Those are junior placements. They are not Agnico. They are not a comment on index earnings of $219.77.
Why he buys them is a guess you should not dress up as his essay, because he did not write Zhu's essay. The visible pattern is an old one. He buys pieces of companies that do not yet have the balance sheet Zhu is praising. He is early, concentrated, and willing to be diluted or rewarded in a single name. That is a person's risk. It is not a model of "stronger balance sheets." A strong balance sheet is what a junior is trying to deserve. Eric Sprott is often on the side of the story before the balance sheet exists.
People mix the two because the name is the same. The mix is convenient for anyone selling a basket. If Eric is buying, the mood says, the sector is on. His buying says he likes specific small stocks enough to fund them. Zhu's note says the large miners are already rich and still cheap on a multiple. Hathaway says the large miners may be too rich in cash and too slow to use it. Three claims. One surname. A stock to watch has to survive without the surname.
Lazy cash, or grown-up cash
This is the argument inside Sprott, and it is more useful than the slogan. Zhu treats the cash as proof. Debt down. Net cash up. Dividends and buybacks at highs. The industry learned from the $129 billion lesson. Hathaway treats a pile of cash as a problem with a clock. He said larger producers have held more than a billion dollars for five years. Over those years the purchasing power of cash fell. Sitting on it is a habit from the lean years after 2011, when managers who had nearly died did not want to look hungry. He would rather see buybacks, higher dividends, a careful acquisition, or even bullion on the balance sheet. Cash that does nothing is not discipline. It is fear with a good excuse.
Both can be right about different companies at the same time. A miner that spent 2013 like a drunk, and now refuses to borrow, has grown up. A miner that has been net cash for five years, while its reserves age and its shares trade at a low multiple, may be hoarding. The low multiple Zhu cites, 5.7 times forward EBITDA, is itself a clue. If the market will not pay up, one reason is that investors remember the waste. Another reason is that they do not believe the cash will reach them. A dollar in a corporate treasury is not a dollar in your account. It becomes yours when it is paid out, or when it is spent on a project that earns more than it costs. Until then it is a claim on management's nerve.
Hathaway thinks the nerve is about to show up as deals. He said mergers and bolt-on buys are just around the corner, because cash burns a hole. He also said, in a line a careful reader should not adopt, that valuations are so low you could almost throw darts at big producers, growth stories, and takeout candidates. Dart throwing is what the last cycle called diversification, right up until the write-downs. The fact that a Sprott partner can say it is a sign of how wide the margins look. It is not a method. Every asset and jurisdiction differs. He said that too, in the same conversation. Keep the second sentence. Throw away the darts.
Five ways the cash can go
The five names below are not Sprott's list. They are five machines that sit inside the debate Zhu and Hathaway are having. Several of them are large weights in the Sprott Gold Miners ETF, ticker SGDM, as of October 7. That is a fact about a fund, not a blessing. Agnico was about 9.8 percent of the fund. Newmont about 7.7 percent. Barrick about 7.5 percent. Wheaton Precious Metals about 7.2 percent. Franco-Nevada about 6.0 percent. Kinross about 4.1 percent. The fund's net asset value that day was $71.53, down 3.34 percent, on about $627 million of assets. A down day in a product is not a verdict on the research note published the same week. It is a reminder that the note and the price can disagree by Tuesday.
Agnico Eagle is the cost end of the argument. In the second quarter of 2026, against a gold price that averaged about $4,512, Agnico reported an all-in sustaining cost near $1,459. That cost had ticked down from about $1,483 in the first quarter, while several peers paid more. Implied margin, which is only the average price minus that cost, was about $3,053. At a spot price a little above $4,100, the same cost still leaves a margin near $2,700, if the cost holds. This is what "record earnings" looks like in one company. It is also what capital discipline looks like when the discipline shows up as a lower cost, not as a press release. The thing to watch is whether the cost stays down when the gold price no longer flatters every decision. A low cost that only exists in a boom is the boom. A low cost that survives the mid-$4,000s is the business Hathaway said could work even if the metal does not rise.
Newmont is the scale. It reported an all-in cost near $1,621 in that same quarter, and an implied margin near $2,891. A note on October 8 said the company's 2026 plan assumes gold at $4,500. Spot is below that. The cash is still vast at $4,100, because $1,621 is not close to $4,100. The plan is a different object from the cash. Scale lets a company return money and build at the same time, which is what Zhu praises. Scale also lets a company write a deck at a price the market has left, which is what you watch. On October 7 the shares were reported near $113.54, down about 2.5 percent on a day the metal fell less. The gear is working in both directions. Consider Newmont as the test of whether the largest balance sheet in the group gets redrawn honestly, or whether the $4,500 assumption lingers in the model after it has left the tape.
Barrick is the cost that has been walking up. About $1,581 in the fourth quarter of 2025. About $1,708 in the first quarter of 2026. About $1,866 in the second. The direction matters more than the level. A balance sheet can strengthen on paper while the mine gets more expensive to run. If earnings are at a record because the gold price did the work, and the cost line is the leak, then "stronger" is temporary. Barrick's implied margin at $4,512 was still about $2,646. That is not distress. Distress would be a cost that meets the price. The watch item is the next print. If the walk stops, the balance sheet story and the cost story agree. If it does not, Hathaway's warning about 8 to 10 percent cost inflation has a name.
Kinross is the split decision. A late-September account of its revised guidance said production for 2026 and 2027 had been cut by about 7 to 8 percent, to roughly 1.84 to 1.86 million ounces, and that all-in costs had been lifted by about 9 percent, toward $1,850 to $1,900. The same account put net cash near $1.9 billion and a target of returning about half of free cash flow. Read those lines together or do not read them. Net cash is Zhu's proof. A cut to ounces and a rise in cost is the thing the proof does not cancel. A lazy-cash critique does not fit a company that is already promising to hand half the free cash back. A record-earnings critique does fit, if the record was partly a volume story that just got smaller. Kinross is the stock that stops a sector sentence from becoming a sector order.
Franco-Nevada is the fifth, and it is not a miner. It is a royalty company. SGDM holds it beside the diggers, which is already a lesson. Zhu's sentence is about operators who used to borrow and build. A royalty does not run the pit. It takes a slice of revenue from pits other people run, and it tries to keep a clean balance sheet by design. Hathaway's wish that producers would hold some bullion, so the cash is metal rather than paper, is a cousin of what a royalty already is. You do not watch Franco-Nevada for an all-in sustaining cost. You watch whether the contracts keep paying if the operators behind them cut volume, the way a guide gets cut. Wheaton, the streamer sitting just above Franco in that same ETF, is the same family with a different contract. Putting either of them on a list of "gold miners" because a fund did is how labels replace mechanics. The mechanic is the point. A royalty can have the strong balance sheet without having the operating leverage that makes a miner's earnings "record" in a spike and painful in a drop.
What the multiple is refusing to pay for
A forward multiple of 5.7 times EBITDA, if the index math holds, is cheap next to 9.9 times in 2014. Cheap next to a past peak is not the same as cheap against what happens next. The earnings expectations that went from about $50 to about $220 were written while gold was doing something it had not done in the lives of many portfolio managers. Part of that $220 is a gold price. If the price stays near $4,100, some of the $220 remains, because costs are still far below. If the price goes back toward $3,500, less of it remains. If costs rise 8 to 10 percent a year while the price marks time, Hathaway's squeeze arrives on a calendar. The multiple may be low because the market is subtracting that calendar in advance. Calling the subtraction a mistake is Zhu's opinion. It is a serious opinion. It is not a receipt.
There is another subtraction, and it is memory. The $129 billion erased after 2011 was not a metaphor. It was deals and builds that looked wise at the top. Investors who will not pay 9.9 times again are not necessarily using an outdated lens. They may be using the last lens that saved them money. An outdated lens would ignore net cash, ignore the cost gap, and price every miner as if it were still the 2013 version of itself. A current lens would say the finances are better and the multiple should not return to a boom peak until the better finances have lived through a flat tape. Between those two lenses is the entire argument. Five stocks do not settle it. Five filings might.
Reserve scarcity is Zhu's longer claim, and it should stay long. He cites a scarcity of major discoveries since the 1990s, and a path from discovery to production that has stretched from about 6.4 years in 1999 to about 18 years. If that is right, the mines that already exist are harder to replace than the multiple implies. Scarcity does not make a given share cheap. It makes a failed replacement expensive. A company that uses its record cash to buy a bad ounce, because ounces are scarce, can destroy the balance sheet Zhu just praised. Hathaway's "M&A is just around the corner" is the moment that risk shows up. Scarcity is a reason to be careful about deals, not a reason to cheer them.
The fund is not the research
Sprott Gold Equity Fund, as of June 30, 2026, did not look like the five names above. Its top holdings included gold bullion at about 10.7 percent, then G Mining, IAMGOLD, Coeur, DPM Metals, i-80 Gold, OceanaGold, Agnico, Eldorado, and Ramelius. The fund can hold bars and smaller miners at the same time the research note talks about majors and net cash. That is allowed. It is also a warning against reading a firm as a single mind. Zhu writes about an index. Hathaway talks about lazy cash and darts. The equity fund, on that date, had a large bar position and a set of names a dart could also hit. Performance through September 30 showed a strong three-year number and a weak September, with the fund down about 8.8 percent in the month against a gold-and-silver stock index down about 11 percent. A good three years and a bad month can both be true. Neither is a stock pick.
SGDM is closer to the majors, and it still holds a streamer and a royalty in its top ranks. If you buy the ETF because you agreed with Zhu, you did not buy his sentence. You bought a rules-based basket, an expense ratio around 0.46 percent, and whatever the index does on a day yields jump. On October 7 the basket was down more than 3 percent. Gold miners have operating leverage. The leverage does not pause because a research note was published. Stronger balance sheets reduce the chance that a down quarter becomes a bankruptcy. They do not reduce the chance that a down week becomes a down share price.
What to watch, if the verb is serious
Watch the use of cash, not the existence of cash. A balance sheet can be strong in four ways, and only one of them is automatically good. It can hold net cash because the mine is so far above its cost that money arrives faster than it can be wasted. That is Agnico's neighborhood, if the cost stays near $1,459. It can hold cash because the plan was written for a higher gold price and spending has not caught up. That is the question sitting on Newmont's $4,500 deck. It can show strength that the cost line is quietly spending. That is the question on Barrick's walk from $1,581 to $1,866. It can show cash and a smaller future at the same time. That is Kinross's cut beside its $1.9 billion. Or it can be strong because the business never owned the pit. That is Franco-Nevada, and it is a different promise.
Watch the cost against a gold price that is no longer the second-quarter average. The record earnings were not a secret method. They were price minus cost, and the price was extraordinary. Harris's $1,900 average all-in cost, if it is even roughly right for the sector, leaves a huge margin at $4,100 and a smaller one at $3,500 and a problem near $2,200. You do not need Sprott to do that subtraction. You need the company to keep reporting the cost in a way you can audit. All-in sustaining cost leaves things out. Use it as a trend. Do not use it as a moral score.
Watch whether anyone pays you. Hathaway's complaint about lazy cash is a shareholder's complaint. Zhu's praise of distributions is the other half. If dividends and buybacks are at highs, the cash is not all lazy. If a particular company is the exception, the sector sentence does not cover it. Read the capital return, then read the project budget. A company can raise the dividend and approve a mine that only works at $5,000. Both lines will be in the same release. The release is the watch. The headline is the trailer.
Why the bull case does not need Eric
The bull case Sprott the firm is making does not require Eric Sprott's latest placement. It requires three things that can be checked without him. First, that costs stayed behind the gold price, so earnings really did hit records rather than merely getting talked about. The index EPS path and the major miners' 2026 cost prints are the evidence, and they point the same way for the quarters already reported. Second, that balance sheets used the windfall to retire debt instead of to relive 2011. Net cash at many producers is the evidence Zhu cites, and it is checkable company by company. Third, that the multiple stayed low anyway, so you are not paying peak-cycle prices for peak-cycle earnings. The 5.7 times figure is the evidence, and it is an index figure with a date of September 30, 2026.
The bear case does not require a villain either. It requires Hathaway's calendar. Gold in the mid-$4,000s, costs up high single digits or worse, margins thinner in year two than in the photograph. It requires the memory of $129 billion. It requires the chance that M&A, which he welcomes, is how the memory repeats. It requires a 10-year Treasury yield near 5.3 percent, which makes a dollar of miner cash in 2028 worth less than it was when yields were lower, no matter how clean the balance sheet looks today. Yields do not care that the industry grew up. They reprice every long-lived claim.
Eric Sprott can be bullish for a fourth reason that fits neither paragraph. He can think a small company, funded now, will be worth more if ounces stay scarce and the majors need to buy something in a few years. That is a venture bet riding on Zhu's scarcity point and Hathaway's deal point. It is not a bet on record earnings at Newmont. If you came for the headline, you came for the majors' cash. If you came for Eric, you came for a placement. Do not use one as the reason for the other.
What would make Sprott wrong
Zhu is wrong if the earnings were only the gold price, and the gold price gives back enough of the January move that the fourfold EPS jump collapses toward the old $50 world. He is wrong if the net cash is spent on acquisitions that the next downturn writes down, and the $129 billion gets a sequel. He is wrong if investors are not using an outdated lens but a correct one, because they have watched costs walk up at Barrick and guidance walk down at Kinross and decided the "new era" is a few companies wide, not an industry wide. A sector sentence is always a little false. The falsehood becomes expensive when you buy the average.
Hathaway is wrong if the cash is not lazy, because the projects on the books are good, the buybacks are large, and two or three years of a flat gold price still leave margins the equity market has refused to capitalize. He said the stocks could work even without a higher gold price, through discipline, deals, and operating leverage. That claim dies if the deals are bad or the leverage only works upward. Operating leverage on a falling gold price is not a bonus. It is the reason miner funds fall more than the metal, as they did on October 7, when GDX was down about 3 percent and the metal was down less.
Eric is wrong, on any single check, the way a placer is often wrong. A junior can fail, sit, or get taken out. His bullishness is a portfolio of those outcomes, not a research conclusion about balance sheets. Quoting him as the reason the majors are safe is a category error. The majors are safer than they were in 2013 if the net cash is real. They are not safe because a famous buyer likes a different part of the market.
The close
Sprott says gold miners have stronger balance sheets and record earnings. Kenny Zhu said it, about an index, on October 7, 2026, and he paired it with a multiple that has not kept up. John Hathaway said the profitability is the best he has seen since 1998, and that the cash becomes a trap if it sits there or if costs catch the price. Eric Sprott has been buying a different shelf of the market. The headline that turns all of that into five stocks to watch is a headline. The watch is the use of the cash.
Agnico asks whether a low cost can stay low. Newmont asks whether scale will tell the truth about a $4,500 plan in a $4,100 market. Barrick asks whether a rising cost will spend the balance sheet the earnings just built. Kinross asks whether net cash and a production cut can be true in the same year, which they can. Franco-Nevada asks whether you wanted a miner at all. None of these is a buy because Sprott exists. Each is a filing. The filing is the era, if there is a new one. The slogan is the lens.
A note on sources and limits
The October 7, 2026 Sprott note is "Gold Mining 2.0: Are Investors Looking Through an Outdated Lens?" by Kenny Zhu. The earnings path from about $49.60 at the end of 2023 to $219.77 on September 30, 2026, the forward EV/EBITDA of 5.7 times versus 9.9 times in June 2014, the description of net cash, lower debt, and high distributions, the $129 billion of value erased from 2012 to 2017, the discovery timeline from about 6.4 years to about 18 years, the second-quarter miner drawdown of about 25.3 percent from March 2 to July 16, and the August rally of about 30 to 32 percent are that note's figures, tied there to the NYSE Arca Gold Miners Index and to sources including McKinsey and S&P Global. They are not this article's forecasts.
John Hathaway's comments are from his conversation with Paul Harris of Kitco at the 2026 Precious Metals Summit in Beaver Creek, published by Sprott in late September 2026. The "lazy balance sheets," the billion dollars of cash, the line about profitability since 1998, the warning that margins compress if gold marks time in the mid-$4,000s, the 8 to 10 percent cost pressure, the M&A remark, and the dart image are his. The average all-in cost near $1,900 and the margin of $2,000 to $2,400 were Harris's figures in that interview, not a Hathaway calculation.
Eric Sprott's 2026 placements in names such as Cerrado Gold, Silver Storm, and Goldgroup are taken from those companies' disclosures. They are not a published investment strategy and not a recommendation. SGDM holdings and the October 7 net asset value near $71.53 are from Sprott's fund page and change daily. The Gold Equity Fund's June 30 top holdings, including bullion near 10.7 percent, are from the fund's materials as of that date. September performance is the fund's own table.
Company cost figures are second-quarter 2026 reports as compiled in a September comparison: Agnico about $1,459, Newmont about $1,621, Barrick about $1,866, against a quarterly average gold price near $4,512. Kinross's revised production band, higher cost band, net cash near $1.9 billion, and free-cash-flow return target are from a late-September account of company guidance. Newmont's $4,500 planning assumption and the October 7 share prices are from an October 8 market note. Gold near $4,111 on October 7, and the January highs near $5,589 intraday and $5,405 at the close, are from market reports that week. The 10-year yield near 5.3 percent is the early-October tape. All of these can be revised by the next print.
This is not investment advice and not a solicitation to buy or sell any security. Sprott is an interested commentator because it manages gold products. Balance sheets can be spent. Costs can rise. Gold can fall further. Readers should read the company filings and the Sprott note itself, and speak with a licensed adviser, before any decision.