The struggle showed up in gold and silver prices. A higher long-term base is also a real sentence in a real bank note. Those two facts are not the same fact. On September 28, 2026, BMO Capital Markets cut its near-term gold and silver averages and raised its long-run averages. Kitco, reporting that note the same evening, had spot gold near $4,135, down almost 3.5 percent on the day, and spot silver near $61.20, down more than 4.5 percent. The struggle was on the screen. The higher base was in a model. Mixing them is how a cut to the next quarter gets sold as a floor under this week.
The numbers have to sit next to each other. BMO’s fourth-quarter gold average is now about $4,650, down 2 percent from $4,750. The bank does not see $5,000 again until the second quarter of 2027. Its long-term gold average moved from $3,100 to $4,000, a rise of 29 percent. Silver’s next three months are now about $67.40, down from $71.40. The first quarter of 2027 is about $69.30, down from $73.50. The long-term silver average moved from $36 to $47, a rise of 31 percent. Read those pairs slowly. The near term was cut. The far term was raised. A headline that keeps only the raise is an edit.
This piece has one theme. The higher base BMO just published sits under the price, or, in silver’s case, far under it. It is not this week’s floor. Spot gold near $4,135 is already standing on the new $4,000 long-run average. Spot silver near $61 is about 30 percent above the new $47 long-run average. A base you have not fallen to is not support. A base you are standing on can still give way. The struggle is the gap between a live price and a near-term average the bank itself just lowered. Calling that struggle a base being built skips the only investor-friendly change, and parks it in the wrong year.
This is news and context. It is not investment advice. It is not a gold price forecast or a silver price forecast from this desk. BMO’s averages are BMO’s. They are not a personal target, and they are not a reason to buy or sell gold stocks, silver stocks, or the metals. A model can be thoughtful and still be wrong. This week is already a reminder that the path and the print are different objects.
One theme
Hold the theme against every later paragraph. Is a higher long-term base forming for gold and silver? In BMO’s model, a higher long-run average is now on the page. Forming, in the price, is a separate test. Gold would have to keep living above $4,000 when real yields are ugly. Silver would have to keep living above $47, which it is doing easily, and that ease is the trap. Easy is not proof. A long-term average is a center of gravity in a model. It is not a bid in the pit.
What is driving the long-term gold price outlook, in this note, is not the yield. BMO says the raise looks wrong if you only watch TIPS and hiking cycles. The bank thinks monetary debasement and Chinese official buying are now larger than the old yield driver, and that the People’s Bank of China will backstop prices at a higher level. That is the driver. It is a view. It is not a law. Could gold and silver build a higher price floor? They can, if that backstop shows up in ounces when the price falls, and if the opportunity-cost trade does not overpower it. This week the opportunity-cost trade was winning the day. The backstop is a forecast about the decade. Days and decades should not share a verb.
What the struggle actually was
Start with the tape, because the tape is the struggle and the model is not. Monday, September 28, was a hard down day. Kitco’s dispatch put gold near $4,135 and silver near $61.20 while both were still falling. Other series put Monday’s gold low near $4,111 to $4,115 and the close near $4,115 to $4,133, after Friday near $4,285. Silver’s Monday close on one series was $60.64, down about 5.7 percent. Tuesday, September 29, brought a bounce of tens of dollars in gold and a silver price still near $61. A bounce is not a repair. Friday is still overhead. The gold silver selloff of this week is a real event. It is not yet a verdict on a long-run average.
The map above the price is the near-term cut. A fourth-quarter gold average of $4,650 sits roughly $500 above a $4,135 print. That is about 11 percent. Silver’s $67.40 three-month average sits roughly $6 above a $61 print. The bank lowered both and the market is still under both. So the struggle is not a quarrel with a blogger. It is a quarrel with the bank’s own next quarter. BMO gold price forecast, in the quarter that is about to start, was revised down because the headwinds are real. Pretending the revision was a raise is how people get hurt with a PDF.
Kitco tied the day’s pain to oil, the Middle East, fresh inflation fear, and aggressive rate expectations. It said the 10-year yield had pushed to a fresh 20-year high above 5.20 percent, and that investor demand was struggling because the opportunity cost of a non-yielding asset had gone up. Check that yield on a live Treasury page before you carve it into a speech. The direction is the point BMO itself makes. Bond yields and gold are in their old fight. Interest rates and gold are in the same fight. When the yield spikes, the metal that pays nothing has to justify itself every hour. Some hours it cannot.
There is a second map, from the day before, that should not be glued to this one. A Goldman precious-metals note had spot breaking the 50-day and the 100-day, with those averages near $4,266 and $4,319, and the 200-day still under the price near $4,046. A Société Générale note, as reported the same day as BMO’s, talked about a pivot near $4,315 and supports near $4,095 and the June-July troughs around $3,960 to $3,940. Averages differ by feed. Do not staple two banks’ lines into one indicator and call it gold price support. Support is a place that held, or a number in a model. Name which one you mean.
The long-term number is a floor in a model
Now the number people will quote without the rest of the page. BMO Capital Markets gold forecast for the long run is an average around $4,000, up from $3,100. That is a 29 percent raise in a long-run assumption. It is the biggest change in the note, and it is easy to misuse. A long-term gold price forecast of $4,000 is not a forecast that the next stop is $4,650, and it is not a forecast that $4,000 cannot be visited. An average is a center. Prices spend time on both sides of a center. If the new center is $4,000 and the price is $4,135, the price is not being “built” above a mysterious new era. It is already in the neighborhood of the era the bank just wrote down.
Do the silver math with the same coldness. A long-run average of $47, up from $36, is a higher base inside the model. Spot near $61 is not standing on it. Spot is floating well above it. Silver price support, if you mean BMO’s long-run average, is more than $10 under the market, about 23 percent under a $61 print. A gold silver price correction that actually went looking for those long-run averages would be a small further dip in gold and a large further dip in silver. That is not a prediction that it happens. It is the distance. People who hear “higher base” and picture a trampoline under today’s price have not measured the trampoline.
The January bull case should not be smuggled into this note. In January, BMO published a thought experiment, not this base case. It stretched central-bank buying and fund flows and got a bull case near $6,350 gold by the fourth quarter of 2026 and near $8,650 by the fourth quarter of 2027, with a silver sketch near $160 and $220 if the gold-silver ratio sat at 40 to 50. That exercise is not the September update. The September update cuts the next quarter and sets a $4,000 long-run gold average. Anyone who cites $8,650 in the same breath as Monday’s note is stapling a boom scenario to a caution. Stapling is not analysis.
What is driving the long-term gold price outlook
BMO’s own answer is the one to use, and then to test. The analysts said the raise looks counterintuitive with TIPS yields spiking and with central banks taking tentative steps into a hiking cycle. Gold, they said, has been surprisingly resilient in the face of hawkish repricing. They read that resilience as evidence that monetary debasement and Chinese demand are now larger than the traditional yield driver. They see Chinese buying as highly countercyclical. They expect People’s Bank demand to be strong through the forecast period, backstopping prices at a higher level.
That is what is driving the long-term gold price outlook in this document. Not a chart pattern. Not a mining-stock story. Two slow forces, set against one fast force. The fast force is the yield. The slow forces are the fear that money and debt are being debased, and the official bid from China. Central bank gold demand, in this telling, is the backstop. It does not have to win every Monday. It has to show up often enough, in size, that the long-run average belongs at $4,000 rather than $3,100. If the ounces do not show up, the raise was a story about a buyer who stayed home.
The same note refuses to throw out the yield. It says the path into year-end is a contest. On one side, the debasement trade: rising rates and a rising term premium as a sign that fiscal sustainability is in doubt, which can be bullish for gold. On the other, the opportunity-cost trade: rising rates as a reason not to hold a metal that pays nothing. BMO thinks a hawkish Federal Reserve has bought some credibility for now and helped flatten the term premium. It also thinks it will not take much for debasement fear to return. Price risk, the bank says, is still skewed up, with that $4,650 fourth-quarter average, unless oil spikes further. Federal Reserve gold is not a program the Fed runs. It is this contest, priced every week.
US dollar and gold sit inside the same contest. Dedollarisation is one of the slow forces BMO names, beside debasement, as a reason the metal might drift higher in a more orderly way than last year. A firm dollar in a given week is the fast weather. Weather can last a season. A multi-year doubt about the dollar’s reserve role can still be the bank’s view. Holding both without averaging them into a slogan is the job. This week’s weather was yields up and metals down. The long-run view is that Chinese buying and debasement fear put a higher floor under the next several years. The floor is $4,000 in the model. The weather is a 10-year yield that Kitco put above 5.20 percent. Both can be on the same page. Only one of them sets tomorrow’s margin call.
Silver is the metal the bank does not trust as much
BMO does not give silver the same long-run speech. It says higher gold prices should drag silver higher. It also says, in plain words, that faster thrifting and slower solar deployment are loosening the industrial balance, and that it expects the metal to underperform gold over the coming quarters. The silver price outlook for the next two quarters was cut. The silver market outlook is not “$80 because a different bank said so.” It is “a three-month average of $67.40, a first-quarter average of $69.30, and a long-run average of $47, with gold doing the pulling and solar doing less of the eating.”
The solar point is specific, and it should stay specific. Before 2025, BMO says, much of the solar industry thought the big gains from using less silver had already been booked. The spike above $120 in January 2026 tested that comfort. Chinese manufacturers answered with narrower lines, less paste, more copper substitution, and silver-coated copper pastes. Those efforts, the bank says, are now bearing fruit, and the long-run silver intensity of a panel is meaningfully lower. Inflation and silver can still rhyme when the investment bid is hot. The industrial rhyme is getting thinner. A silver price forecast that leads with solar and never says thrifting is using last year’s chorus.
Put the two long-run averages side by side and the gold and silver outlook stops being one trade. Gold’s new long-run center is $4,000, and the price is near it. Silver’s new long-run center is $47, and the price is near $61. If both averages are honest centers, silver has more air under it than gold does. Air is not a crash. Air is the distance a correction can travel before it meets the number the bank calls a base. Could gold and silver build a higher price floor? Gold’s candidate floor was just rewritten to a place the price already occupies. Silver’s candidate floor was rewritten to a place the price would have to fall a long way to touch. Same adjective. Different basements.
Inflation and gold are the fear that feeds the debasement side of BMO’s contest. Inflation and silver are messier, because silver is still, in this bank’s older comments and in this solar discussion, an industrial metal wearing a monetary costume. A hot inflation print can lift both. A hot inflation print that also lifts oil, hike odds, and the 10-year yield can hurt both, and hurt silver more, because the industrial user starts thrifting at the same time the investor faces a higher opportunity cost. That double brake is why a gold silver market correction does not have to be symmetrical. This week it was not. Silver fell harder. Beta did that. Thrifting may do it again, more slowly.
Safe haven demand did not show up on the scary day
Safe haven gold demand is the phrase people reach for when oil jumps and a strait is in the news. This week it failed the day test. The scary headlines arrived with higher yields, not instead of them. Gold fell. Silver fell more. A haven that only works when the shock is not also a reason for tighter policy is a part-time haven. BMO’s debasement trade is a different haven. It is the fear that the bonds themselves are the problem, that rising yields mean the debt is less trusted, not that the central bank is more credible. The bank thinks the hawkish Fed bought some of that credibility back, for now. For now is the whole near-term forecast. It is why the quarter was cut and the decade was raised in the same document.
Global economic uncertainty is not a blank check for either metal. A war headline, a hiking cycle, a Chinese buyer, and a solar factory learning to use copper are four uncertainties. Only some of them bid gold. Only some of those bid silver. Lump them into one cloud and you will buy the costume of the week. The precious metals outlook in the BMO note is narrower than the cloud. Gold is near the top of the bank’s preference list. Silver is expected to lag for several quarters. A precious metals market outlook that cannot say that sentence is smoothing a disagreement the bank left in the open.
What the raise does, and does not do, for miners
Gold stocks are a lever on the metal, and a lever does not know the difference between a long-run average and a bad Monday. Gold mining stocks can fall hard while a bank raises a decade-long center from $3,100 to $4,000, because the shares price the next year of metal and the fear that the next year gets worse. A $4,000 long-run average, if a company truly believes it, changes mine plans and reserve prices over time. It does not change Monday’s margin call. Mining stocks 2026 are still a 2026 object. The long-run raise is a later object. Do not pay 2026 prices for a 2030 assumption unless the filing shows the assumption is already in the mine.
The cost test is the only one that fits this note. Gold mining companies whose all-in costs sit well under $4,000 still have a wide margin at the new long-run average, and a wider one at $4,135. The margin is thinner than it was near $5,400. It is not a crisis at the mine. Silver mining companies face a stranger gap. A long-run silver average of $47 is a planning price. A spot price of $61 is a cash-flow price. A company that builds a mine on $61 and then meets $47 has built the wrong mine. A company that can live at $47 and happens to sell at $61 has a cushion. The note does not tell you which company is which. Silver mining stocks and gold mining stocks have to be read one cost curve at a time.
Junior gold stocks and junior silver stocks do not have the producer’s cushion. They have a treasury and a drill. A higher long-run average does not fill the treasury. A gold silver selloff can close the financing window even if the long-run model just got kinder. Canadian mining stocks, Canadian gold stocks, and Canadian silver stocks include both kinds of company, on the main exchange and on the venture market. TSXV gold stocks are, on average, closer to the next placement and further from a mill. Precious metals mining stocks is a label, not a single risk. The label hides the split. The split is the risk.
Gold stocks to watch, and silver stocks to watch, are not a buy list in this piece. They are names with a dated test. Does the company publish a cost that still works at $4,000 gold or at $47 silver, not only at January’s highs? Does it need to sell shares before the October Fed meeting? Is its largest asset priced on a bull case the bank itself separated from the base case? Those questions are a watchlist. They are not a recommendation. This desk names no winner. A winner is a cost, a share count, and a country. A long-run average does not do the reading.
People also asked
Is a higher long-term base forming for gold and silver?
Is a higher long-term base forming for gold and silver? Inside one Canadian bank’s model, yes, and the amounts are knowable. Gold’s long-run average went from $3,100 to $4,000. Silver’s went from $36 to $47. Forming in the market is the harder question. Gold is already trading near the new gold average, after a violent year that included a January closing high near $5,405 on one series and a drawdown of about a quarter. A base “forming” at a price you are already at is a claim that the price will stop going down here. BMO did not promise that. It promised a higher center of gravity and a lower next quarter. Silver’s base, at $47, is not forming under the ticks. It is sitting downstairs. The price can build a floor near $61. It has not done so just because a model moved a long-run average that the price does not touch.
What is driving the long-term gold price outlook?
What is driving the long-term gold price outlook is the pair BMO named, not the pain of this week. Debasement fear, and Chinese official buying that the bank expects to be strong and countercyclical. The yield is the opposing driver, and the bank thinks the yield has won some credibility back for the Fed without winning the decade. If you want a one-line driver, use theirs. The new themes are larger than the old yield theme, in their judgment, which is why the long-run average rose even as the quarter was cut. Judgment can be early. Early looks like this week.
Could gold and silver build a higher price floor?
Could gold and silver build a higher price floor? They could. The conditions are not a slogan. For gold, the floor holds near $4,000 if the Chinese backstop buys the dips and if the debasement trade keeps the opportunity-cost trade from becoming the only trade. A break under $4,000 would say the new average was early or wrong. For silver, a higher floor than $36 can be true at $47 and still leave a long fall from $61. A floor near the high $50s or low $60s would be a market fact, earned by buyers, not a fact inherited from the long-run cell in a spreadsheet. Buyers earn floors by showing up twice. One bounce does not count as twice.
Two clocks
The note runs on two clocks, and the mistake is to wear both at once. The near clock says the fourth quarter was revised down, $5,000 gold waits until the second quarter of 2027, silver lags gold for several quarters, thrifting is working, and oil plus yields can still knock the metal over. The far clock says the center of the next many years is $4,000 gold and $47 silver, higher than the old centers, because debasement and Chinese demand moved the furniture. A gold and silver forecast 2026 that uses only the far clock will buy a dip that the near clock just widened. A forecast that uses only the near clock will declare the long-run raise a joke because Monday was red. Both clocks are in the same PDF. Read both.
Orderly is the word BMO used for the drift it still expects. Higher, but more orderly than last year. Orderly is not what the January spike to $120 silver felt like, and it is not what a 4 percent down day feels like. The bank is trying to describe a market that can rise without repeating the panic bid. Markets do not have to cooperate with the adverb. They can be disorderly on the way to a higher average, and they can be orderly on the way through it to the downside. The adverb is a hope about the path. The averages are the hope about the destination. Hopes are labeled. They are not levels.
The gold market outlook for the next two quarters is therefore allowed to be dull and harsh at the same time. Dull, because the bank still has gold near the top of its list and still sees upside skew if oil does not spike again. Harsh, because spot is under the quarter’s average, yields are the headwind the bank calls critical, and the metal has to decouple from real yields to make the nice path. Decoupling is the claim. It has not been completed. This week was coupling. Yields up, gold down. A gold price outlook that says “decoupled” in the present tense is ahead of the evidence. “Trying to decouple” is the tense the note earns.
Three ways the two clocks can resolve
These are paths, not predictions.
One. The backstop shows up. Chinese official buying and the debasement bid absorb this drawdown above $4,000. Gold grinds back toward the $4,650 fourth-quarter average without a new mania. Silver lags, as the bank expects, and spends time between the low $60s and the high $60s while thrifting keeps the industrial balance loose. In this path the higher base was real, and it was gold’s base. Silver’s higher long-run average never had to be touched, because the near-term average did the work. The struggle was a passage, not a basement.
Two. The opportunity-cost trade keeps the wheel. Yields stay high, oil stays a hike scare, and gold goes looking for $4,000 and possibly through it. The new long-run average becomes a battleground instead of a floor. Silver, leveraged to gold and loosened by thrifting, falls farther in percentage terms and makes $47 look less theoretical. In this path the raise was early. Early is a polite word for a loss if you bought the raise as if it were a bid. The note allowed this path. It said the yield contest is live, and it said a further oil spike sits outside the upside skew.
Three. Chop between the Monday low and the mid-$4,000s while everyone waits on the Fed and on whether official buying is real in the next reserve data. Silver chops with a downside bias, which is what “underperforming gold” looks like when gold is not trending. This path punishes anyone who needed a verdict this week. The long-run raise can still be right in 2028 and useless in October. Useless in October is not the same as false. It is the same as early, and early is expensive if the position is leveraged. Precious metals mining stocks in a chop get the worst of the boredom and the gap moves. Producers with cash survive boredom. Juniors finance it, or they do not.
How to read the next revision
This is a filter, not a plan.
Read which number moved. If the long-run average rises again and the quarter falls again, the bank is repeating this note, not confirming a bottom. If the quarter is revised up toward the spot price, the struggle is fading in the model. If the long-run gold average is cut back toward $3,100, the higher base was withdrawn. The base is not a mood. It is a cell. Watch the cell.
Read the Chinese bid in ounces, not in adjectives. Countercyclical and strong are the adjectives. The test is whether official buying is large on the weeks the price is weak. A buyer who only appears in the narrative is not a backstop. Central bank gold demand is a table. Tables get revised. Until the table agrees with the adjective, the adjective is a thesis.
Read silver’s industrial balance beside gold’s monetary story. If thrifting keeps lowering silver use per panel, the drag the bank named is still on. If solar deployment speeds back up and the copper pastes disappoint, the drag lightens and the underperformance call gets harder. Silver price support from industry is a use-per-panel number. Get the number. Do not borrow gold’s debasement story and paint it silver.
Read miners after the clock is clear. A producer that lives at $4,000 gold is a different risk from a venture listing that needs $5,000 to look funded. Canadian gold stocks are not one trade. Junior silver stocks are not a call option on BMO’s far clock unless the treasury says they can wait for it. Gold stocks to watch means filings that answer the cost question at the new long-run average. It does not mean the names that bounced the most on Tuesday.
What this does not mean
It does not mean the low is in. A higher long-run average published on a down day is a coincidence of calendars until the price stops making new lows. BMO did not say the Monday print was the floor. It said the floor in the model is higher than it used to be, and the next quarter is lower than it used to be. Quote both halves or quote neither.
It does not mean $4,000 is a promise of support. An average is not a wall. Gold can trade at $3,700 and still average $4,000 over a long window if it also trades at $4,300. The raise makes the center higher. It does not forbid a visit below the center. Anyone who buys “because $4,000 is the base” has turned a mean into a stop. Means are not stops.
It does not mean silver is cheap because the long-run average went up. Silver is expensive relative to that average. It is cheap relative to January’s $120 spike. Both comparisons are true, and they point at different trades. The gold silver market correction from the January highs is already enormous. A further trip toward $47 would be another leg, not the first. Do not let a 31 percent raise in a long-run cell hide a price that is still double the old cell and far above the new one.
It does not mean you should own the metal, the miners, or the bank’s preference list. Gold at the upper end of a bank’s list is the bank’s portfolio language. It is not your allocation. Shares add costs, politics, and dilution. The metals add opportunity cost every day the yield stays high. Global economic uncertainty does not refund either one.
It does not mean this desk has a target. The conditional sentence is the whole forecast. The higher base holds, as a description, if official buying and the debasement bid keep gold from living below $4,000. It fails, as a description, if the opportunity-cost trade pushes the price through that average and the buyer the bank counted on does not arrive. Silver’s near-term path lags gold unless thrifting stops. That is a reading of BMO. It is not a silver price prediction with a date on it from here.
The close
Gold and silver are struggling. On September 28, the prints were near $4,135 and $61, with Monday’s lows and closes in the same neighborhood, after a week that took gold down from the high $4,200s and took silver down harder. BMO, the same day, cut the fourth-quarter gold average to about $4,650 from $4,750, pushed $5,000 out to the second quarter of 2027, cut silver’s three-month average to about $67.40 and the first-quarter average to about $69.30, and said silver should underperform while solar thrifting loosens the balance. It also raised the long-run gold average from $3,100 to $4,000 and the long-run silver average from $36 to $47. The raise is the higher base. The cuts are the struggle. They were published together on purpose.
Is a higher long-term base being built? A higher center is now written down, and the reason the bank gave is monetary debasement plus Chinese buying that it expects to backstop dips, even while TIPS and a hiking cycle argue the other way. That is what is driving the long-term gold price outlook. It is not yet a floor you can point to on a chart and say it held. Gold is already near $4,000. Silver is nowhere near $47. Could gold and silver build a higher price floor? Only if buyers earn it, twice, at a level you can name. For gold the candidate is the new average itself, and it is close enough to touch. For silver the candidate in the model is a long way down, and the candidate in the market has not been proven by one bounce.
The theme is the refusal to mix the clocks. A long-run raise is not this week’s support. This week’s selloff is not the death of the raise. Spot is under the quarter the bank just cut and, for gold, already on top of the decade the bank just lifted. That is an uncomfortable address. Uncomfortable is not the same as a base. A base is a number that holds when the yield is screaming. The yield, this week, was screaming. The number has not been asked the question yet.
Important information
This article is for information and education only. It is not investment advice and not a recommendation to buy or sell gold, silver, gold stocks, silver stocks, precious metals mining stocks, Canadian mining stocks, junior gold stocks, junior silver stocks, or any other asset. Gold stocks to watch and silver stocks to watch, here, mean names with a dated public test. They are not buy lists. Bank averages are not guarantees. Prices are volatile. Futures can cost more than the cash posted against them. Miners face cost, political, operational, and dilution risk. Past prices do not indicate future results.
The BMO figures in this piece — fourth-quarter gold near $4,650, down from $4,750; $5,000 gold deferred to the second quarter of 2027; a long-term gold average near $4,000, up from $3,100; silver near $67.40 over three months, down from $71.40; a first-quarter 2027 silver average near $69.30, down from $73.50; a long-term silver average near $47, up from $36 — come from BMO Capital Markets’ early fourth-quarter 2026 update as reported by Kitco on September 28, 2026, including the analysts’ remarks on debasement, Chinese buying, the yield contest, and solar thrifting. The January bull-case sketches near $6,350, $8,650, $160, and $220 are from a separate BMO scenario reported at the time, not from the September base case. Kitco’s same-day market levels, including gold near $4,134.80, silver near $61.19, and a 10-year yield described as a 20-year high above 5.20 percent, are that outlet’s figures and should be checked against live data. Other September 28 and 29 prints cited here, including Monday lows near $4,111 to $4,115, a silver close near $60.64, and a January gold closing high near $5,405, come from separate market series that do not match to the cent. Moving-average levels near $4,266, $4,319, and $4,046 follow a September 28 note citing Goldman Sachs. Société Générale levels are as reported the same day and are not the same indicator. Readers should read the BMO note and check live quotes. This article does not consider any person’s goals or finances.