Gold demand: flat volume, record value, ETF outflows

The World Gold Council reported total demand including over-the-counter activity unchanged year on year at 1,269 tonnes for the second quarter. First-half demand reached 2,522 tonnes, up 2 per cent, with a record half-year value of about 380 billion dollars.

The record value is a price statement dressed up as a volume statement. Anyone quoting the dollar figure as evidence of demand growth is conflating the two.

The over-the-counter component of the headline figure also deserves caution. It is the least transparent line in the World Gold Council's methodology, estimated as a residual rather than measured directly, and it has previously been revised by meaningful amounts. Treat the 1,269 tonne figure as a reasonable estimate of total demand rather than a precise count.

Exchange-traded funds shed roughly 45 tonnes over the quarter under selling pressure as the price consolidated. That is the weak leg of the demand structure and the one most sensitive to the rate path.

Fund flows are the marginal, price-setting buyer in most quarters. Their absence explains why the metal consolidated despite official-sector strength: central banks set the floor, funds set the direction.

It is possible to read the outflow more benignly: profit-taking after a strong run is a normal feature of investment vehicles and does not by itself indicate a change in the medium-term view among fund holders. The distinction that matters is whether outflows are concentrated among short-term tactical holders or are starting to include longer-duration allocators, and the World Gold Council's data does not cleanly separate the two.

Price-sensitive jewellery demand in India and China continues to absorb the adjustment through lower volumes at higher value. Recycling has picked up in the same markets, which caps rallies at the physical level even while investment demand pulls the other way.

This dynamic means the demand data is internally offsetting: strength in one category is frequently a function of weakness in another, mediated through price. A reader looking for a single clean demand narrative in this release will not find one, and that absence is itself the finding.

Coverage of this release has generally led with the record dollar value, which is the least analytically useful number in the report because it mechanically follows from price rather than telling you anything new about buyer behaviour. The more informative comparison, tonnage growth by category against the prior four quarters, received comparatively little attention despite showing the clearest divergence between price-insensitive official buying and price-sensitive investment flows.