Gold as the neutral asset of a splitting monetary order. The war premium left — and gold rose anyway. Why two rival systems are buying the same metal, and why it earns a place in a portfolio, not a moment on the tape.
Before the evidence, our claim. Gold is being repriced not as a commodity, and not merely as an inflation hedge, but as the one reserve asset that is nobody’s liability — and it is being bought, at the same moment, by two blocs that agree on almost nothing except a distrust of the dollar they are both trapped inside. In the West, confidence in the Fed is fraying; in the East, reliance on Washington is being quietly hedged. The two roads run in opposite directions and arrive at the same metal. What follows walks each road, meets the strongest objection head-on, and lands where an owner has to land — on the question of how much to hold, not when to trade.
For two years gold traded off two things: the path of US real interest rates and confidence in the Federal Reserve. On that logic the summer sell-off made sense — the sharp rise in real yields after the Iran conflict was a genuine headwind, because a higher real return on bonds raises the opportunity cost of holding an asset that yields nothing.
On the breakout, the composition of the bid showed through. Real yields had eased only marginally — the 10-year inflation-protected yield slipped from about 2.44% to 2.40%, a move too small to be the cause of a 4.3% day — and nominal yields had come off their war-time peak as Hormuz reopened. The larger move was in the dollar, which slid to its weakest since early June. One session is not a regime, and we do not build a thesis on a single tape; but the session crystallised a shift the longer-lens data had already been showing: the marginal price of gold is no longer set by the opportunity cost of bonds so much as by the dollar — and, underneath the dollar, by confidence in the institution that issues it.
That the shift is structural rather than a one-day quirk is the industry data’s own point: on a multi-year lens gold has climbed through historically restrictive real rates since 2023, because a different, price-insensitive buyer — central banks and Asian demand — has been overriding the opportunity-cost arithmetic. Elevated real rates, on this evidence, have not been the obstacle the textbook insists they must be. To understand the bid, then, look past the tape to the two very different buyers now reaching for the same metal — and why.
The Western bid is a credibility trade. A new Fed chair has commissioned task forces on, among other things, the central bank’s inflation framework, and has floated examining a broader set of data than the current target measure. Read charitably, that is housekeeping; read as the market is inclined to read it, it looks like preparing to move the goalposts on inflation rather than clear them. Longer-run inflation expectations — the measure the Fed watches most closely — remain anchored near 2%; but households’ expectations three years out are the highest since 2022, business surveys of prices paid are elevated, and the AI build-out is itself a near-term inflationary force through its demand for power, chips and critical materials.
None of this is a forecast that inflation reaccelerates. It is something subtler and, for gold, more durable: a widening doubt about whether the institution charged with controlling inflation will define the problem honestly and then act on it. When that doubt grows, the asset that depends on no institution’s good faith earns a premium. The nuance matters, because it disciplines the claim: gold and inflation are not joined at the hip. Below roughly 4%, gold largely ignores the inflation rate; it is above that level, when inflation begins to feel like a policy mistake, that gold’s interest ignites. Core PCE sits near 3.3% today — close enough to that zone to be watched, not yet inside it. The Western bid is less about the next inflation print and more about faith in the referee — and that faith is thinning.
The Eastern bid is a different animal, and it is the one most Western commentary misses. Over the past decade China has built, piece by piece, an alternative to the dollar’s grip on oil: its own cross-border clearing network, a yuan-priced crude contract in Shanghai, and — the piece that matters most — the fact that an exporter paid in yuan can convert those proceeds into physical gold on the Shanghai Gold Exchange (and its international board) and through Hong Kong. This is convertibility, not a gold peg: there is no ‘oil-for-gold’ contract, and claims of one are a myth. But the exit it offers is real, and it eases the problem that had always capped the yuan’s international reach — no reserve manager wants to hold a currency they cannot take home. China’s answer was not to make the yuan freely convertible; it was to make it convertible into gold.
Here it is worth being precise about what carries the argument. The case does not rest on divining Beijing’s motive; it rests on observable behaviour, and the behaviour is not in doubt. Three things are documented: official-sector gold buying has run at record levels for two years, with central banks saying they intend to keep adding (World Gold Council); China has cut its US Treasury holdings to an eighteen-year low and keeps selling (US Treasury data); and the plumbing of a yuan-based, gold-convertible alternative to dollar oil settlement is in place and in use. Point those three in one direction and the conclusion — a large, price-insensitive official bid for gold, running alongside a deliberate retreat from the dollar — follows from the facts, not from a theory.
What remains interpretation is the why: the reading that a share of that buying functions specifically as the collateral that lets a yuan oil trade settle. It is a compelling explanation, but it is an interpretation, not a proven mechanism, and the argument does not depend on it. If it is right, the official bid is stickier still; if it is merely diversification, the bid is still large and still price-insensitive. Either way, the leg stands. One recent episode fits the pattern: through the 2026 Hormuz disruption, China drew on its reserves and cut crude imports by roughly 40% to a decade low (US EIA; China customs), riding out a supply shock with an autonomy from the seaborne dollar market that few large importers could manage.
The strain is visible on the Western side of the same plumbing too. To arrest the yen’s slide, the US and Japan intervened jointly for the first time in roughly fifteen years — and the revealing detail is the financing: the US Treasury has been pressing the Federal Reserve to expand an emergency facility (the FIMA repo line) so that Japan can fund yen-buying by pledging its Treasuries to the Fed rather than selling them, which would push US yields higher. A reserve-adjacent major currency that its own central bank cannot defend without drawing on the Fed’s balance sheet is itself a statement about the system’s stress; and the intervention is already struggling, with the yen resuming its slide within days. Even Washington’s own answer to de-dollarisation — encouraging privately issued dollar stablecoins as a new source of demand for Treasuries — is, read plainly, an admission that the old, automatic source of that demand is fading. Two blocs, both now openly managing the dollar.
The best counter-argument deserves to be put at full strength. It is that there is, quite simply, no alternative. A currency earns reserve status by meeting a demanding and well-understood set of conditions — deep and liquid markets, free convertibility, a government-bond market large enough to absorb the world’s savings, and a sovereign willing to run the external deficits that supply the world with its currency. On every one of those the United States still stands alone; China, with a managed rather than convertible currency and a surplus rather than a deficit, meets almost none. Reserve accumulation, on this view, is not a vote of confidence but an arithmetic necessity — a world that sells more to America than it buys must park the dollars it earns somewhere, and the Treasury market is the only pool deep enough to hold them. The dollar’s dominance is therefore secure for years, whatever its flaws; the de-dollarisation of settlement, though real, is slow.
We think that argument is correct — and that it does not weaken the case for gold; it explains it. The no-alternative argument is a claim about the reserve currency. Gold is not a candidate to be the reserve currency; it is the neutral reserve asset that both blocs accumulate precisely because there is no alternative currency — because they are trapped in a dollar whose management they increasingly distrust, the West doubting the Fed’s resolve and the East doubting Washington’s restraint. If a credible alternative currency existed, capital would flow to it and gold would matter less. None does — so the flow finds the one reserve asset that is nobody’s liability and no government’s to freeze. The absence of an alternative is not gold’s problem. It is gold’s reason.
What follows is a statement about weighting, not timing. Gold earns a place as a strategic, structural holding in a sophisticated portfolio — insurance against a monetary order fragmenting at both ends, held by owners who trust neither the Fed’s resolve nor Washington’s restraint, and underpinned by a central-bank bid that has become part of the plumbing rather than a matter of preference. That case has strengthened, not weakened, through the summer.
On timing we are deliberately unhelpful, and the tape is the reason. Gold has risen about 14% in a fortnight to roughly $4,300, and the move now carries the marks of a crowded one: the daily relative-strength index pushed into overbought territory (around 72) on the breakout; speculative futures positioning is stretched and one-sided, with none of the capitulation that usually clears the way for a durable low; and the exchange-traded inflows leading the advance are the fast, price-sensitive money, not the sticky official bid the case rests on. None of that refutes the argument — it dates the entry. The discipline follows: hold the strategic allocation, and add into the pullbacks a firmer dollar or a Fed surprise will provide, not into a vertical.
Name the risk plainly, because it is not the obvious one. The cleanest threat is not gently higher rates; it is a credible Fed — a genuine tightening the market does not price — because that would land a double blow: it revives the real-rate opportunity cost and, by restoring the Fed’s anti-inflation credibility, drains the Western debasement premium at the same time. That single scenario undoes both near-term legs at once, and it is the honest reason to size the allocation patiently rather than press it here. The structural supports should keep gold underpinned once something in the macro picture breaks and long-dated yields fall — though not necessarily by repeating the outsized gains of last year. The register, in short: constructive on the holding, patient on the price.
What to watch. Whether real yields stay contained and the dollar’s slide extends — the two near-term drivers of the breakout; confirmation that the structural central-bank and Asian bid continues; and, the cleanest thing that would revive the old headwind, any sign the Fed genuinely tightens. The near-term US jobs and inflation prints are the swing.
What to be wary of. Chasing a roughly 14% move compressed into a fortnight; treating gold as a tactical trade when the case here is a strategic allocation; and mistaking the no-alternative argument — which is true — for a reason gold cannot rise, which does not follow.
What we are not doing. This is general market commentary, not advice or a recommendation on any security. Gold, the currencies and the mechanisms described here are set out as conditions, not as a trade. Anything that touches your own portfolio is a suitability conversation — one that needs a person, not a page. We are always glad to have it.
Gold price, demand and positioning — World Gold Council (Gold Market Commentary, July 2026; Gold Demand Trends Q2 2026) and market data. The rotation from real rates to the dollar and the Fed-credibility reading draw on Bloomberg market commentary (6 August 2026). Real yields — US 10-year TIPS. China’s strategic-reserve scale — US EIA estimate (China does not officially disclose). China’s Q2 crude-import decline — US EIA and the American Petroleum Institute. China’s US Treasury holdings — US Treasury TIC data. The joint yen intervention and the FIMA repo facility — CNBC, Bloomberg and Reuters (1–5 August 2026). Figures are as reported at the date of writing and may be revised; where a figure is an estimate it is described as such.
按教科书,黄金本应走软:霍尔木兹海峡重启、春季的战争溢价消退、油价回落至 80 美元下方,而美国实际利率仍处于多年高位——对一项不生息的资产而言,本是最不利的环境。七月金价一直被压在约 4,027 美元、年内下跌约 8%。然而 8 月 5 日,金价单日上涨 4.3%(今年二月以来最强),升至约 4,300 美元。地缘溢价已经离场,黄金却照涨不误。这正是本文的切入点:不是黄金无视了某一天的数据,而是在战争溢价消退、实际利率仍高之际它依然获得买盘——以及这对一个投资组合应如何持有黄金意味着什么。
我们的核心判断:黄金正被重新定价——不是作为商品,也不仅是通胀对冲,而是作为“无人负债”的中立储备资产;并且在同一时刻被两个几乎毫无共识、却都不信任其所被困之美元的阵营同时买入。在西方,对美联储的信心正在磨损;在东方,对华盛顿的依赖正被悄然对冲。两条路方向相反,却通向同一种金属。
西方的理由(贬值)。这是一笔“信用交易”。新任美联储主席就通胀框架等设立多个工作组,并暗示将参考比现行目标“更广泛的数据”——市场倾向于将其读作“移动球门”而非达标。长期通胀预期仍锚定于 2% 附近,但居民三年期预期升至 2022 年以来最高,企业调查中的价格压力偏高,AI 建设本身在近端亦具通胀性。要点在于:黄金与通胀并非“形影不离”——通胀低于约 4% 时黄金基本无视之,唯有当通胀开始显得像“政策失误”时,黄金的兴趣才被点燃(当前核心 PCE 约 3.3%,接近该区域但尚未进入)。西方的买盘,关乎的是对“裁判”的信心,而这份信心正在变薄。
东方的理由(去美元化)。过去十年,中国逐块搭建了绕开美元“石油霸权”的替代体系:自有的跨境清算网络、以人民币计价的上海原油合约,以及——最关键的一块——以人民币收款的出口方可在上海黄金交易所(及其国际板)与香港将所得兑换为实物黄金。这是可兑换性,而非金本位挂钩:并不存在“以油换金”的合约,声称有者是一种误传;但这一“出口”是真实的,它缓解了长期制约人民币国际化的难题——没有储备管理者愿意持有一种“拿不回家”的货币。本论点不依赖揣测北京的动机,而立足于可核实的行为:官方黄金购买连续两年处于历史高位(世界黄金协会);中国持有的美国国债已降至十八年低位并持续减持(美国财政部数据);人民币计价、可兑金的替代石油结算体系已然运行。将三者指向同一方向,结论便自明:一股庞大且对价格不敏感的官方黄金买盘,与一场蓄意的“撤离美元”并行。
最强的反驳(TINA)——以及为何它是在“解释”而非“驳倒”本文。反方最有力的论点是:别无选择。储备货币需满足严苛条件——深度与流动性、自由兑换、足以吸纳全球储蓄的国债市场,以及愿意承担相应外部赤字的主权国;美国独一无二地满足全部,中国几乎全不满足。因此美元的主导地位在数年内仍稳固,结算层面的去美元化虽真实但缓慢。我们认同这一判断——但它并不削弱黄金的理由,反而解释了它:TINA 说的是储备货币;黄金并不参选储备货币,它是两大阵营都在增持的中立储备资产——正因为没有替代货币,它们被困在一种其“管理方式”日益不被信任的美元里。若真有可信的替代货币,资金会流向它、黄金将不那么重要;正因没有,资金才流向唯一“无人负债、无人可冻结”的储备资产。没有替代品,不是黄金的问题,而是黄金的理由。
我们的结论:这是一个“配置”问题,而非“择时”。黄金应在成熟投资组合中占有一席战略性、结构性的配置——对一个两端同时碎裂的货币秩序的保险。至于择时,我们刻意不给答案:金价两周内上涨约 14% 至约 4,300 美元,日线 RSI 升至超买(约 72)、投机头寸偏多且拥挤、领涨的是快钱而非黏性的官方买盘——这些并不推翻论点,只是提醒“入场点已变差”。纪律因此而来:持有战略配置,在更强美元或美联储意外带来的回调中逐步加仓,而非追高。真正的风险不是温和的加息,而是一个可信的美联储——它会同时抬高实际利率、并抽走西方的“贬值溢价”,一击而中两条近端逻辑;这正是从容配置、而非此处追高的理由。
本摘要为节选,非全文翻译。本刊为一般性市场评论,不构成投资建议或对任何证券的推荐。如与英文版本存在任何歧义,概以英文版本为准。