US output grew 3.4%. Private buyers tell a different part of the story

What the quarter delivered

The US Bureau of Economic Analysis published its first estimate on 29 October at 08:30 Eastern daylight time. Real output grew at a seasonally adjusted annual rate of 3.4%, after 2.2% in the preceding quarter. In ordinary quarterly terms BEA estimated growth of about 0.8%. Dividing an annualized headline by four gives a quick approximation. Taking the fourth root of 1.034 and subtracting one gives about 0.84% using the rounded headline.

That distinction matters if a commodities note says the economy produced 3.4% more in three months. The output increase was appreciably smaller. The annualized figure expresses the pace that would result if the quarterly change continued for four quarters.

Source notes: [1] US Bureau of Economic Analysis · Q3 GDP advance estimate

Follow the goods into the warehouse

Consider a hypothetical domestic manufacturer that makes machinery in September and places it in inventory. Its production can count in third-quarter GDP even if a buyer takes delivery in October. When that buyer purchases the machine, inventory is run down. Counting the full machine as new output again in October would double-count the same production.

This is worth checking in the actual release because BEA identified inventory accumulation and equipment as leading sources of the increase in investment. Its narrower measure of final sales to private domestic purchasers includes household spending and private fixed investment, excluding changes in inventories. That measure grew at a 4.9% annual pace, up from 4.6%. Growth in both series is evidence against an explanation that relies exclusively on unsold stocks. The 1.5-point difference between the two Q3 rates is not an inventory contribution. The measures cover different components and have different bases.

Source notes: [1] US Bureau of Economic Analysis · Q3 GDP advance estimate

The link to gold runs through an expectation

Suppose an analyst believes strong private purchases will persuade investors to expect tighter monetary policy. The observation to seek is a repricing of the expected rate path relative to immediately before publication. An advance estimate can show accelerating growth yet leave expectations untouched if traders had already priced it. It can even disappoint an expectation for faster growth. An archived market consensus would help distinguish those cases. This article does not assign a numerical surprise.

Why the first price tick cannot settle attribution

BEA also scheduled September personal income and outlays at 08:30 Eastern on the same day. A gold move in the following minute would therefore span two releases. To argue for a GDP-specific effect, preserve forecasts for both releases and identify which components surprised. At the second estimate scheduled for 25 November, compare private final demand and inventory contributions separately. A revision concentrated in inventories would change the production story more than the evidence about buyers.

Source notes: [1] US Bureau of Economic Analysis · Q3 GDP advance estimate · [2] US Bureau of Economic Analysis · News release schedule