NextFin News - Gold is rallying again, but the move looks less like a simple fear trade than a test of who now sets the floor in the market. Fresh July inflows into physically backed gold exchange-traded funds, renewed official-sector buying in May and a second quarter in which total demand held steady even after prices came off record highs all point to the same tension: is this another cyclical rebound driven by macro nerves, or evidence that a deeper class of buyers is turning every correction into a re-entry point?
The numbers suggest the answer is both, but not equally. The latest monthly reserve data show official gold reserves increased by a net 41 tonnes in May 2026, according to the World Gold Council’s compilation of IMF and central-bank disclosures through 30 June. July then brought $3 billion of net inflows into global gold-backed ETFs, lifting assets under management 1% to $530 billion and collective holdings by 23 tonnes to 4,068 tonnes as of 31 July. In the broader physical-and-financial market, second-quarter total gold demand including over-the-counter activity was unchanged from a year earlier at 1,269 tonnes, while first-half demand reached 2,522 tonnes and a record value of $380 billion. Prices had eased from the first quarter’s extremes, but demand did not break with them.
That is the key market fact. Gold still behaves like a cyclical asset over short windows. ETF flows turn quickly, macro hedging appetite rises and falls, and investor conviction remains sensitive to the path of the dollar, real yields and global risk. But the deeper support under the market increasingly looks structural. Central banks are still buying. Reserve managers still expect to keep buying. Supply is not responding quickly enough to force a reset. The result is a market in which rallies may still be cyclical in timing, but corrections increasingly meet a class of buyers whose motives are strategic rather than speculative.
As-of anchor: all fund-flow, holdings and demand figures in this article are drawn from World Gold Council releases published through 31 July 2026, with monthly official reserve data compiled through 30 June 2026.
What the Latest Demand Data Actually Say
The most immediate reason gold is rallying again is easy to see. July reversed a soft second quarter for ETF demand. World Gold Council data show global gold-backed ETFs were under selling pressure in the second quarter, with outflows of 45 tonnes as weaker gold prices, firmer inflation expectations and a stronger dollar weighed on North American demand. Then July broke that pattern: investors added $3 billion, raising global holdings to 4,068 tonnes. That rebound matters because ETF money remains the fastest part of the gold complex. It does not tell you everything about the market, but it tells you when tactical investors are returning.
Yet the July reversal matters most in combination with what did not weaken. Total second-quarter gold demand including over-the-counter activity was 1,268.9 tonnes, effectively unchanged from a year earlier. First-half demand reached 2,522 tonnes, up 2% from the same period of 2025, and the value of that demand hit a record $380 billion. Central banks made significant purchases in the quarter, with demand of 289 tonnes, while the monthly reserve data showed a further net 41-tonne increase in May alone. If gold were only a momentum trade, the cooling in ETF demand during the second quarter should have produced a more obvious breakdown in aggregate demand. It did not.
The supply side reinforces that point. Second-quarter total supply was also 1,268.9 tonnes. Mine production rose 2% year on year to 965.6 tonnes, but recycled gold fell 6% to 326.1 tonnes. This matters because a classic cyclical rally should, over time, draw out more scrap, invite producer hedging and eventually create the supply response that caps the move. Instead, even with elevated prices, the market saw only modest mine growth and a weaker recycling response. Prices eased from the first-quarter peak, but that did not trigger the sort of supply adjustment that would quickly loosen the balance.
There is a second piece of context that makes the current rally more revealing than it first appears. The World Gold Council’s second-quarter report said gold ETFs came under selling pressure during Q2, but total demand still held steady because over-the-counter activity and central-bank buying offset the weakness. That is not a trivial detail. It means the market no longer needs every traditional source of demand to fire at once. One pocket of money can step back while another one keeps the overall balance tight. In older gold cycles, that kind of substitution was less visible. In the current cycle, it is central to how the market is functioning.
That combination is why this story is more important than a headline about another week of haven buying. The tactical rebound is real. So is the strategic support underneath it. One is about timing. The other is about market structure. Confusing the two is how investors end up treating a durable repricing as if it were just another short-lived panic bid.
The Cyclical Part of the Rally Is Real, and It Should Not Be Ignored
Start with the short-term truth: gold still rallies faster when tactical money comes back. ETF inflows are the visible sign of that re-engagement. The second quarter proved how quickly that cohort can reverse. Gold ETFs saw outflows of 45 tonnes during the quarter, and the World Gold Council explicitly tied those outflows to weaker prices and, especially in North America, to upward adjustments in inflation and interest-rate expectations alongside a stronger US dollar. Then July delivered a reversal. That is the definition of cyclical behavior. The same investor base that backed away when macro conditions turned less friendly came back when the setup improved.
There are at least three historical cycle comparisons that support calling the recent leg a cyclical rally in timing rather than denying that cyclical forces exist. The first is the Q2-to-July reversal itself: a 45-tonne ETF outflow in the quarter followed by a 23-tonne increase in holdings in July. The second is the February peak in ETF holdings. July’s 4,068 tonnes remained below the record 4,176 tonnes reached on 27 February 2026, which means the latest rebound has repaired some but not all of the earlier retreat in investment demand. The third is the repeated pattern in gold markets more broadly: periods of tactical ETF or futures buying can accelerate a move, then partially unwind when policy expectations or the dollar move against them. The July bounce fits that template.
A fourth comparison helps sharpen the point. In full-year 2025, gold demand including over-the-counter activity exceeded 5,000 tonnes for the first time, and the metal set 53 new all-time highs during the year. Even in that extraordinary environment, ETF demand and consumer demand did not move in a single straight line. Investment flows strengthened at some moments, jewelry demand weakened under the weight of high prices, and official-sector buying remained resilient rather than euphoric. The market was already learning to trade with multiple demand engines moving at different speeds. That same pattern is visible again in 2026, with ETFs recovering after a weak quarter even as official demand stays firm.
That matters because it keeps the analysis honest. Not every rally in gold is a regime shift. Some rallies are simply macro hedges getting rebuilt after a correction. If the article stopped there, the conclusion would be straightforward: fast money came back, and gold rose with it. But that conclusion would fail the second-order test. It would explain the speed of the move without explaining why the floor under the market keeps looking higher than traditional cyclical models would imply.
The cyclical channel works like this. Softer confidence in competing assets, a more attractive hedge profile, a less punitive carry backdrop for a non-yielding asset, or a rise in geopolitical uncertainty brings tactical capital back into gold. That capital expresses itself quickly through ETFs and futures. Price then moves faster than physical demand alone would justify. This is the part of the story everyone sees. It is real, and it often dominates the headlines. But it is not the full market anymore.
What is different this time is not that tactical money has stopped mattering. It is that tactical money now appears to be trading on top of a thicker strategic base. That means the cyclical rallies can still be sharp and the cyclical pullbacks can still happen, but the market may be mean-reverting to a higher equilibrium than in older cycles. Short-term volatility survives. The center of gravity shifts.
The Structural Part Is the More Important Story
The structural case begins with central banks, because official demand is the clearest evidence that gold’s buyer base has changed. World Gold Council data show 2024 central-bank buying totaled 1,045 tonnes, the third consecutive year above 1,000 tonnes. In 2025 that total moderated to 863 tonnes, but even that lower number remained far above the 2010-2021 annual average of 473 tonnes. In 2026, monthly statistics showed official reserves increased by a net 41 tonnes in May, while second-quarter central-bank demand reached 289 tonnes. Those are not the numbers of a market whose strategic buyers have disappeared.
More important than the tonnage is the persistence. In a cyclical framework, demand that rises sharply when risk spikes should fade as prices rise or the original catalyst cools. Official-sector buying has not behaved that way over the last several years. It persisted through 2024, when annual buying surpassed 1,000 tonnes again. It remained strong in 2025 even as the annual total slipped below that threshold. And it is still visible in 2026 while ETFs and other investor channels continue to swing with the macro backdrop. That persistence is the hallmark of structural demand. It is responding to reserve-allocation decisions, not to a weekly chart.
The survey evidence points in the same direction. In the World Gold Council’s 2026 central-bank survey release, 89% of reserve managers said they expect global central-bank gold holdings to increase over the next 12 months. A record 45% said they expect their own institutions to add to holdings. Another 83% said gold would account for a higher share of total reserves five years from now, up from 76% in the prior survey. In the same release, 90% cited gold’s performance during times of crisis as a reason to hold it. Those numbers do not prove that every central bank will buy aggressively every month. They do show that the official sector increasingly views gold as a live reserve asset rather than a historical relic.
“The World Gold Council’s annual Central Banks Gold Reserves Survey reveals that 89% of reserve managers expect global central bank gold holdings to continue increasing over the next 12 months,” the World Gold Council said in its 16 June 2026 survey release.
The quote matters less as rhetoric than as evidence of motive. Strategic reserve managers are not trying to maximize a quarter’s carry. They are responding to concentration risk, sanctions risk, crisis hedging needs and a broader reassessment of what belongs in official reserves. The same survey release noted that gold had recently overtaken US government bonds as the top reserve asset in the eyes of respondents, while 90% cited crisis performance, 84% long-term store-of-value characteristics and 82% diversification as reasons to hold the metal. That motivation stack is structural almost by definition. It does not disappear because a monthly payrolls report or inflation print changes market pricing for the next central-bank meeting.
Here the cyclical-versus-structural call becomes explicit. The rally is cyclical in its trigger because tactical flows have clearly turned back toward gold. The support underneath it is structural because the official-sector motivations are not self-correcting in the way speculative flows are. A hedge fund buying gold because it fears a policy surprise can sell it once the fear fades. A central bank adding gold because it wants more diversification and a larger geopolitical hedge does not automatically reverse when the next month’s macro headlines improve. One motive mean-reverts. The other does not.
There is also a practical market consequence. If central banks and other strategic buyers are less price-sensitive than jewelry consumers or momentum funds, then corrections do not clear the market the way they once did. They become transfer points. The market moves metal from weaker hands to stronger hands, then tries to rally again. That is why the latest bounce deserves more than a superficial explanation. It is happening in a market whose ownership base is already changing.
The Transmission Mechanism Runs Through Supply Elasticity, Not Just Fear
A lot of gold commentary stops at direct causality: buyers appeared, so prices went up. That is not enough. The more important mechanism is what this buyer mix does to supply elasticity and therefore to the durability of price moves. In the second quarter, mine production increased 2% year on year, but total supply did not meaningfully expand because recycled gold fell 6%. That means higher prices still did not produce a broad enough supply response to cool the market.
Why not? Mine production is structurally slow. Even with better margins, bringing new supply to market takes time, capital and permitting. Recycling is faster, but it depends on willingness to sell. When gold is being accumulated as a strategic reserve asset or a long-duration hedge, holders can become less responsive to spot price strength than simple textbook models assume. In effect, strategic demand does not just add tonnage on the way in. It can also reduce the amount of metal that comes back out.
That is the second-order implication many cyclical read-throughs miss. The first-order effect of fresh official or ETF buying is obvious: more demand pushes price higher. The second-order effect is more powerful: persistent strategic buying can make the market less efficient at generating its own balancing supply response. If the metal is increasingly held by entities with long horizons and low sensitivity to carry, then every correction has less available inventory to work through than it would in a purely speculative market. That does not remove volatility. It makes sustained downside harder to achieve.
The third-order implication is an expectation gap. A market that still prices gold mainly as a short-term macro hedge may repeatedly underestimate how shallow pullbacks can become when strategic buyers are waiting below the market. That is exactly why July matters. A quarter of ETF outflows did not lead to a collapse in aggregate demand, and a modest softening in price was enough to draw ETF investors back in while official demand remained intact. The market corrected, but the ownership structure did not reset.
Read the full second-quarter demand report through that lens and a pattern emerges. Investment is expected to remain the primary driver of demand growth through the rest of 2026, supported increasingly by over-the-counter activity and Asian buying, while Western ETF flows remain sensitive to rates and the dollar. Central banks are on course for another strong year even if annual buying ends below 2025. Jewelry volumes remain under pressure from high prices, yet overall demand is holding because the buyers that matter most right now are not the traditional price-sensitive consumer segments. That is not a normal cyclical-only market. It is a market being held up by a different stack of buyers.
The mechanism also helps explain why high prices have not been enough to shake demand loose. In many commodity markets, a strong price simultaneously invites more supply, reduces end-user demand and eventually forces a reset. Gold is different because some of its most important buyers are not using it as an industrial input. They are using it as a reserve asset, a balance-sheet hedge or a store of value. That means demand destruction can happen in jewelry or some retail channels while strategic demand stays intact. The price is high, but the reason people hold the asset has changed. That is what keeps the market tighter than a superficial reading would suggest.
The Strongest Counter-Thesis Is Serious, and It Could Still Win
The cleanest objection to the structural-bid thesis is that analysts may be overstating the role of official demand and understating how much of the latest move still depends on tactical investors. Under that view, gold remains primarily a macro and momentum asset. Central banks help, but they do not set the marginal price in the way ETF investors, futures traders and broader cross-asset positioning do. If macro conditions turn decisively against gold, private flows could still overwhelm the official bid and expose the latest rally as another cyclical move dressed up as a strategic shift.
This is not a strawman. It has evidence behind it. July’s ETF rebound came right after a quarter that saw 45 tonnes of outflows. Holdings at 4,068 tonnes remain below the 4,176-tonne peak from late February. The World Gold Council’s own outlook says central-bank buying in 2026 is likely to finish below the 2025 total, even if it remains strong. Jewelry demand is under pressure because high prices are constraining affordability. And one of the most striking facts in the second-quarter report is that consumer demand and investment demand are moving in opposite directions. That is a reminder that tactical money still matters a great deal for where the price clears in the near term.
That counter-thesis deserves space because it attacks the foundation of the bullish structural reading. If official buyers are supportive but not decisive, then the current market can still reverse hard when the tactical cohort turns. If the tactical cohort is the true price setter, the structural-floor argument becomes an elegant story built on the wrong marginal buyer. That is a serious challenge, not a debating prop.
My judgment is that the counter-thesis is directionally right about the short term but wrong about the medium-term floor. Tactical investors still set the speed and often the visible direction of the market. They do not, on the latest evidence, seem to be the only cohort determining where corrections stop. The proof is not philosophical; it is in the data sequence. Aggregate demand held steady in Q2 despite ETF outflows. Central-bank demand reached 289 tonnes in the quarter. Monthly official reserve data still showed net buying in May. July then brought ETF investors back before holdings had even fully repaired the earlier damage. That is not how a purely tactical market typically behaves.
The falsifying signal therefore needs to be concrete. The structural-floor thesis would weaken materially if three things happened together: first, monthly official reserve data turned into repeated net selling or an extended run of negligible buying; second, global gold ETF holdings fell well below current levels rather than stabilizing after pullbacks; and third, recycling and mine supply rose enough to push total quarterly supply materially above demand. A practical threshold would be two consecutive quarters in which central-bank demand falls markedly below the recent run rate, ETF holdings continue to contract, and total supply outgrows demand by a clear margin. If that combination emerges, this rally should be reclassified as mainly cyclical.
What Comes Next: Three Horizons, Three Different Risks
In the short term, gold still trades like a sentiment-and-liquidity asset. ETF flows will remain the fastest visible signal, and they can add torque in both directions. A continuation of July-style inflows would tell you that tactical investors are willing to rebuild positions after the second-quarter pause. A renewed run of outflows would say the latest rally still lacks staying power on its own. That is the horizon where macro headlines, the dollar and the broader appetite for hedges matter most.
In the medium term, the more important question is whether official-sector demand and over-the-counter buying continue to offset any unevenness in Western flows. The base case from current data is that they will. Central banks remain on course for another strong year of net purchases even if 2026 does not match the most extreme prior annual totals. July ETF inflows showed that tactical investors are still willing to re-enter on weakness. Supply growth remains modest. Put together, those conditions point to a market whose corrections are more likely to be contained than to become self-reinforcing.
In the long term, the thesis rests on whether gold’s role inside reserves is actually changing. The evidence says yes: 2024 central-bank demand topped 1,000 tonnes for a third straight year, 2025 demand remained high at 863 tonnes, and the 2026 survey showed both rising willingness to add gold and rising willingness to assign it a larger share of reserves. If that regime persists, gold is being repriced not simply as an emergency hedge but as a more permanent component of reserve architecture. That is a structural shift. If it fades, the market goes back to being far more dependent on cyclical investors for direction.
The scenario split follows naturally. The base case is a structurally supported market with cyclical rallies and corrections, where official demand keeps floors relatively high and tactical money determines how violent the swings become. The upside case is that July’s ETF rebound broadens, over-the-counter and Asian investment stay firm, and official buying continues at a pace strong enough to tighten the market further. The downside case is that macro conditions turn against gold at the same time official buying cools, ETF holdings resume shrinking and recycling finally responds more aggressively to high prices. Those are observable triggers, not abstract narratives.
Gold’s latest rally therefore says less about fear alone than about market ownership. The short-term move can still reverse. The longer-term floor looks harder to shake. In this market, the correction is cyclical, but the bid beneath it increasingly looks structural.
Explore more exclusive insights at nextfin.ai.
