Spot held the $4,700 line through Geneva week

Gold spot closed at $4,703 on Thursday May 14, $15 above the prior day's close and $1,519 above the same date one year ago. The metal has now spent the better part of two months trading in a tight band roughly $200 wide, and is still about 16 percent off the January 28 all-time high of $5,602.22. The London PM fix on May 14 came in at GBP 3,474.82 per ounce, consistent with the dollar-side print after currency adjustment.

The pause in spot did not change the wholesale story for the trade. Kilo bar premiums stayed tight through the week. Scrap rates were active around the higher fix, particularly out of pawn-tier networks that were finally seeing meaningful inventory turn. My refining contacts did not loosen the queue. None of this is consistent with a near-term breakdown in the structural bid.

Q1 Council data confirmed the bid is institutional

The World Gold Council's Q1 2026 release, published in late April, gave dealers the year's cleanest snapshot of the buyer base. Central banks bought a net 244 tonnes in the first quarter, up 3 percent year over year. Bar and coin demand surged 42 percent to 474 tonnes. That is the second-highest quarterly figure for retail bar and coin demand on record. The Council's full-year 2026 central-bank guidance is roughly 755 tonnes. That figure is meaningfully below the last three years (each above 1,000 tonnes) but well above the pre-2022 baseline of 400 to 500 tonnes annually.

The structural read is that the central-bank rotation into gold did not pause when spot reset off the January high. The bid stepped down a tier but did not break. For dealers calibrating premium spreads going into JCK, that matters more than the day-to-day spot action. The full cross-asset wrap of Geneva auction week and the Q1 release is here.

What is moving on my floor

Three operational reads from this week. First, kilo bar premiums on the wholesale side are still tight enough that I am not discounting against spot for sub-quarter delivery, even on volume orders. Second, scrap conversion through the smelt cycle is moving faster than at any point since February, partly because pawn networks are finally letting go of inventory they had hoarded through the January spike. Third, the trade-in side at the boutique is the strongest it has been since the back half of 2024 because clients are converting older gold jewelry into new high-purity pieces and bullion at favorable margins.

The cross-product read is also worth noting. The Phillips Geneva XXIII hammer total of $96.3 million, the highest-grossing single watch sale ever held, did not happen in a vacuum. Trophy stores of value are running together. The watches breakdown is here. The same client base that bid the Patek 2523 South America to $10.2 million is also in the bullion and jewelry-melt market on the buy side, which is why kilo premiums are not loosening.

The diamond cross-read

The diamond market gave us the contrast this week. De Beers Sight 4 from April 27 to April 30 was dull and stagnant. RAPI April split between melee strength and one-carat softness. The store-of-value bid is concentrated in gold and trophy watches, not in the natural diamond complex right now. Richard's full diamond read is here.

What I am watching into JCK

Three things on my screen for the next two weeks: the FH April Swiss export print due in the back half of May, which will tell us whether the trophy bid in Geneva is translating to brand-side reorders; the Richemont fiscal-year print on May 22, which will flag any softening in jewelry buying that could feed back into gold-melt supply; and the kilo-premium tape through JCK, which is the cleanest weekly read on whether the structural bid is intact going into the second half. If JCK floor traffic holds and Council Q2 data shows central-bank net buying staying at or above the 244-tonne Q1 pace, $4,700 is a floor, not a ceiling. If both soften, we revisit the $4,400 to $4,500 range I called out in February. The week's read is constructive on balance, not euphoric.