What Actually Moves Gold and Silver Prices, and It’s Not Supply

Gold dropped roughly $1,000 across approximately three trading sessions near its 2025 peak. For months before the correction, the dominant explanation for rising prices centred on supply constraints, mine output trends, and deficit arithmetic. None of those factors changed during the selloff. The macro backdrop remained supportive. The fundamentals, as most investors understood them, pointed the same direction they had for a year.

So what actually moved the price?

CPM Group, a U.S.-based precious metals research firm with historical data stretching back to the mid-1960s, offers a framework that answers the question directly: investor sentiment and investment flows, not mine supply or fabrication demand, are the primary force setting gold and silver prices. This is not a fringe contrarian take. It is a data-grounded model built on six decades of market-structure analysis.

Here is the framework CPM Group uses to explain not just what happened in 2025, but why the same events keep happening in every precious metals cycle, and how you can use it to read the next one before it arrives.

Why the price of gold and silver is not set in the mine

CPM Group’s analytical structure breaks the precious metals market into three components:

• Total supply: mine output plus recycled metal
• Fabrication demand: jewellery, industrial applications, and technology
• The residual: the metal left over after fabrication demand is subtracted from total supply, which must be absorbed (or sold) by investors

That third component is where prices are actually determined. Investment demand on the residual is the price-setting margin. When investors want more of it, prices rise. When they take profits and sell, prices fall.

CPM Group’s core assertion: Investment demand is the single most important factor lifting gold prices, driven by political, economic, and social anxieties and gold’s safe-haven role.

This might sound counterintuitive if you have spent years watching mine production forecasts. But the logic becomes clear when you understand the stock-to-flow relationship.

The mechanism: how sentiment moves price when supply cannot

Gold’s above-ground stocks (all the gold ever mined and still in existence) dwarf annual mine production. That structural feature means the willingness of existing holders to buy, hold, or sell has far more impact on price than any plausible change in yearly mine output. Watching production forecasts to anticipate gold price moves is structurally disconnected from what actually sets the price.

CPM’s historical data shows that investors have been net buyers of gold in all but roughly three or four years since approximately 1965-1966. Those net investment figures represent the balance of simultaneous gross purchases and gross sales, not one-directional flows. Investors are simultaneously the largest buyers and the largest sellers of gold and silver.

In practice, the cycle works like this. Macro fear, falling real yields, and currency weakness draw new investors into gold and silver through ETFs and futures, pushing prices higher. At elevated price levels, earlier buyers take profits through the same channels, causing sharp corrections. This pattern, not any shift at the mine gate, is what CPM documents as the primary source of precious metals price cycles.

Gold’s 2025 cycle as a live case study in flow-driven pricing

Gold started 2025 at approximately $2,660 per ounce. By mid-year it had surged past $3,500. The catalysts at each stage were not supply shocks or sudden changes in mine output. They were investment catalysts: geopolitical risk, interest-rate expectations, and broad macro anxiety that drove capital into gold through ETFs and futures.

The ETF channel tells the story most clearly.

GLD, the largest gold ETF, took in approximately $7.5 billion through early July 2025, after recording outflows the prior year. Gold and Bitcoin funds combined attracted approximately $7 billion over five recent trading sessions during the peak inflow period.

The distinction between gold ETF vs physical gold matters directly here: ETF structures are what make rapid institutional entry and exit possible, and that liquidity is precisely why ETF flow data functions as a leading indicator of price direction rather than a lagging one.

The roughly $1,000 decline across approximately three sessions near gold’s 2025 highs occurred while macro fundamentals remained supportive. Nothing changed in the mine supply picture. What changed was that investors who had accumulated gold since 2019 began taking profits at elevated levels, and the same ETF and futures channels that had driven prices higher enabled rapid exits.

Non-traditional investors, including retail participants, high-net-worth individuals, and institutional buyers, had entered the gold market in large numbers starting around August 2024. Their accumulation drove the rally. Their profit-taking drove the correction. The fundamentals stayed the same in both directions.

What this tells you is straightforward: monitoring weekly ETF flow data gives you a more actionable real-time signal for short-cycle gold price movements than waiting for quarterly mine production reports. The flows are where the price-relevant information appears first.

Reassessing the silver industrial deficit narrative

If you follow silver markets through retail commentary, you have almost certainly encountered the industrial deficit argument. The logic is compelling on its surface: solar panel manufacturing and technology applications are consuming more silver than mines can produce, creating a structural deficit that will mechanically force prices higher.

CPM Group’s published position is that this narrative is wrong.

Their data shows that silver is currently in surplus, not deficit. Above-ground inventories remain at or near record highs. Honest analysis, using CPM’s methodology, shows surpluses rather than the deficits that dominate retail discussion.

The disagreement comes down to methodology. Two frameworks, two different conclusions:

• The popular framing: mine output minus industrial demand equals a deficit, which must force prices higher
• CPM Group’s framework: total supply (mine output plus recycled silver) minus total fabrication demand equals a residual absorbed by investors, and that residual currently shows a surplus, though a relatively small one

Under CPM’s approach, fabrication demand, while healthy (especially from solar and other industrial uses), is not the primary driver of recent silver price strength. Investment demand, which has risen sharply since 2023, is the main underpinning. Large institutional investors function simultaneously as both the largest buyers and the largest sellers of silver, and their shifting positions have the greatest price impact regardless of the industrial supply-demand balance.

CPM Group’s position: Fabrication demand is not the primary driver of recent silver price strength. Investment flows are.

For you, this means that if you bought silver specifically because you believed a physical shortage would force prices higher, you were acting on a premise that CPM Group’s data does not support. That does not necessarily make silver a bad position, but it means the analytical basis needs reassessing.

For investors reassessing silver positions after reading CPM Group’s investment-flow framework, our full explainer on silver price scenarios examines how different institutional flow trajectories and sentiment regimes translate into distinct 2026 price outcomes, including the downside cases that optimistic deficit-based models typically exclude.

Silver’s 2025 volatility through the investment-flow lens

Silver’s 2025 price behaviour illustrates the same investment-flow dynamic that CPM documents in gold. During June and July 2025, silver market participants became heavily pessimistic. Some forecasters projected sharp further declines that never materialised. Dealers imposed significant discounts on non-standard silver products, including 100-ounce bars and medallions, during this pessimism period.

Silver’s mid-year trough sat in the $55-$58 range according to CPM Group’s own interview materials, though aggregated research sources place the trough in the high-$30s. (The discrepancy likely reflects differences in data sources or venue-specific pricing. Both figures are preserved here with their respective sourcing.) By late 2025, silver had recovered to approximately $67-$68 per ounce, a figure broadly consistent across sources.

The rebound followed the pattern CPM documents repeatedly: pessimistic sentiment and heavy selling preceded the recovery. CME Group data cited by CPM shows precious metals trading volume rose more than 50% in the first half of 2025, with the 100-ounce silver futures contract processing exceptionally high volumes. COMEX contract rolls and futures-related trading added short-cycle volatility distinct from physical supply dynamics, reinforcing that you need to monitor the investment channel, not just the physical market, to understand silver’s price movements.

Reading the signals that actually move gold and silver prices

CPM Group’s framework converts into a specific set of signals worth monitoring. If you currently track mine output reports and industrial demand data to assess your precious metals exposure, CPM’s analysis suggests redirecting that research time toward the variables where price-relevant signals actually appear first.

1. Weekly ETF flow disclosures: ETF inflows and outflows are the most immediately informative real-time indicator for marginal investment demand. In 2025, GLD’s flow reversal from outflows to billions in inflows tracked the price surge more closely than any production metric.
2. Central bank reserve data: Central bank gold accumulation provides the long-term structural underpinning beneath the sentiment-driven cycle. CPM projects elevated investor and central bank gold accumulation over at least the next 10 years, a floor of demand that is separate from and more durable than retail or institutional ETF flows.
3. Sentiment extremes: Both optimism peaks (where profit-taking risk is highest) and pessimism troughs (where rebounds have historically followed) are the practical output of CPM’s framework for allocation decisions. Silver’s mid-2025 washout, followed by a recovery to near-record levels, is the most recent documented instance.
4. Futures positioning data: COMEX positioning and contract roll activity generate short-cycle volatility that can obscure the underlying investment trend. Tracking net speculative positioning helps you distinguish between flow-driven moves and noise.

CME Group metals volume and open interest reports provide daily positioning data across gold and silver futures contracts, giving you a direct window into the speculative and institutional flows that generate the short-cycle volatility CPM Group identifies as distinct from underlying physical supply dynamics.

CPM Group projects elevated investor and central bank gold accumulation over at least the next 10 years, providing a long-term demand floor beneath the sentiment-driven cycle.

Central bank gold demand operates on a longer cycle than ETF flows, but CPM Group treats it as the structural floor beneath the sentiment-driven moves, a source of persistent net buying that has been durable across multiple interest-rate environments and geopolitical regimes.

Gold’s long-run price path illustrates the scale of investor-driven appreciation: approximately $250 per ounce in 1999, rising through approximately $700, then approximately $1,700, and reaching approximately $4,600 by late 2025 (per CPM Group interview materials). At each stage, the move was powered by investment demand shifts, not sudden changes in mine supply. The same documented pattern of profit-taking has repeatedly capped or temporarily reversed rallies even when macro drivers remained supportive.

What the 2025 cycle changes about how to assess precious metals exposure

The analytical inversion at the centre of CPM Group’s framework is this: the price-relevant question for gold and silver is not “what is the supply-demand balance?” but “what is the current state of investor sentiment and flow direction?”

That reframing does not eliminate uncertainty. CPM’s framework does not provide precise price targets, and it does not tell you when the next correction will arrive. What it does is redirect your analytical effort toward the variables that actually governed 2025’s dramatic moves, and away from the supply-deficit arithmetic that failed to explain the correction when it happened.

Investors have been net buyers of gold in all but roughly three or four years since the mid-1960s. The investment-flow framework is not a 2025 phenomenon. It is the pattern that has governed every major gold and silver price move since the modern bullion market emerged.

The same investment channels that accelerated 2025’s rally, primarily ETFs and futures, are the same channels that enabled the rapid corrections. Understanding the mechanism matters more than predicting the direction. When the next cycle arrives, the mine supply picture will be largely irrelevant to the timing and magnitude of the move. The flows will tell you first.

Institutional gold price forecasts for 2026-2027 broadly reflect the CPM framework in practice: the most credible forward projections weight investor sentiment and central bank accumulation more heavily than mine supply changes, and their divergence from supply-side models is widest at exactly the price levels where corrections have historically occurred.

If you currently hold precious metals positions based on deficit logic, this does not mean your position is wrong. It means the reason it works, if it works, is probably not the reason you think. Recalibrating toward the investment-flow framework gives you a more durable analytical lens, one grounded in six decades of data rather than a narrative that CPM Group’s own market-structure analysis does not support.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results. Financial projections and price figures referenced are subject to market conditions and various risk factors.