India has spent roughly $59.68 billion of its reserves in six weeks defending the rupee, and the bill underneath it is denominated in crude. The Reserve Bank of India reported foreign exchange reserves of $726.027 billion as of October 9, down $8.578 billion on the week and down from a record $785.706 billion on September 4. That is the fifth consecutive weekly decline. The rupee has clawed back to 94 and change, stabilized by the simple expedient of a central bank selling dollars into the market every session it opens.
Here is why that belongs on a commodities desk rather than an FX desk. The proximate cause of the drain is an import bill inflated by energy. Brent was reported up 2.1% at $108.17 a barrel on Friday, European benchmark gas at €88.21 a megawatt-hour, and diesel stubbornly expensive despite efforts to add global supply. India refines far more crude than it produces. Petroleum Minister Hardeep Puri noted this week that the country has widened its sourcing from 27 countries to 41 in response to disruptions through Hormuz and the Red Sea. Diversifying the origin of the barrels does nothing about the price of them. More suppliers, same invoice.
Inside that reserve decline sits a line item that should interest anyone long bullion: RBI gold holdings fell $1.319 billion on the week to $105.094 billion. Part of that is valuation rather than tonnage, since the dollar price of the metal has been soft. The broader point survives anyway. A reserve manager burning through a defense fund is not a reserve manager adding ounces. One report this week put the RBI as having already spent about half of the more than $140 billion it raised through extraordinary measures earlier in the year. Central bank accumulation has been a structural bid under this market; when the world's second-largest gold-consuming economy has its central bank in liquidation mode and its currency at 94 to the dollar, the domestic retail price of a tola is doing something quite different from what the screen in New York shows.
Gold now accounts for roughly 14.5% of India's total reserves — and it is the component a defending central bank values down first.
Which brings us, at last, to the price. XAU/USD closed at $4,164.73, down $3.43 or 0.08%. It opened at $4,168.22, a nickel above the prior close of $4,168.17, printed a high of $4,169.45 and a low of $4,163.48, and finished in the lower third of a range spanning all of $5.97. Six dollars. On a day when a G20 central bank disclosed it had spent the better part of sixty billion holding its currency together, the premier monetary metal traded a band you could cover with a rounding error. I have seen quieter sessions. Not many, and not with this much in the background.
What the quiet is actually telling me
The five-session change is -0.1%. The thirty-session change is -4.6%. Against the 90-day range of $4,027.92 to $4,659.66, the close sits 10.6% below the high and 3.4% above the low. That is a market that has stopped falling without starting to rise — compression following a drawdown, with nobody willing to underwrite the next leg in either direction.
| Symbol | Close | Change | Day range | 52-week range |
|---|---|---|---|---|
| XAU/USD | $4,164.73 | -3.43 (-0.08%) | $4,163.48–$4,169.45 | $3,886.49–$5,595.46 |
I owe readers an accounting. Earlier this month I argued a three-session bounce would carry gold above $4,350 inside a week. It never came close, and the reason matters: I underweighted the demand side. The reflex is to treat inflation and currency stress as bullish for bullion. But hot U.S. producer prices, driven substantially by a 20.3% monthly jump in gasoline producer prices, are an energy-cost story, and energy-cost inflation taxes the physical gold buyer rather than recruiting him. A rupee at 94 against triple-digit crude means the Indian jeweler faces a dearer local gold price and a squeezed customer in the same breath. Demand destruction at the physical level is a cap, and it has been functioning as one. My subsequent call for a close below $4,120 resolved correctly, which was less satisfying than it sounds — same mechanism, read the right way around on the second attempt.
So, the test. If the physical squeeze is real, a close below $4,120 within ten sessions puts the 90-day low back in play, and $4,120 is where this argument lives or dies. What would change my mind is a close above $4,200 in that same window. That would tell me the monetary bid is reasserting itself over the physical-demand drag, and that the reserve drain is being read as dollar-negative rather than consumer-negative. I am genuinely unsure which way it breaks. The UN's warning this week that an energy and fertilizer shock is pushing agricultural prices toward a food crisis is precisely the sort of development that has historically brought sovereign buyers back toward bullion rather than away from it.
There is a counterpoint I cannot dismiss. If the intervention works — if the rupee holds 94 and the weekly bleed slows — the Indian physical buyer returns at a friendlier local price, and the marginal demand that has been missing comes back without the dollar price needing to move at all. That would be the bullish case arriving through the service entrance. The reserves have now fallen five weeks running; the question is whether the sixth week shows a central bank that has stopped the bleeding, or one that has simply run short of things to sell.