The May trade blowout: one number, two stories, and only part of it round-trips

On 26 June the Census Bureau's advance report landed with a thud: the U.S. deficit in goods trade jumped to **$105.8 billion** in May, a 14-month high, up $22.7 billion (27.4%) from April's $83.0 billion. Economists had penciled in about $85 billion. The wire copy wrote itself: a surge in imports blew out the gap.

That sentence is true and almost beside the point. "Deficit hits 14-month high" is a description. The question that changes what you should *do* about it is narrower: is May's blowout a temporary pull-forward that mechanically reverses, dragging this quarter's GDP only to hand it back next quarter, or a structural deterioration that sticks? The economy ran almost exactly this experiment a year ago. The twist this time is that the answer is not the same on both sides of the ledger, and getting it right means stripping out a swing in oil prices that quietly moves both.

Start with a fact sitting in plain sight in the release and almost unmentioned: of the $22.7B the gap widened, the *drop in exports* (-$11.8B) was a **larger single mover than the rise in imports** (+$10.9B). The story everyone told, "imports surged," describes the *smaller* of the two forces.

The story everyone told, imports surged, describes the smaller of the two forces. The bigger mover was exports falling.

Now the imports. If May's jump is companies racing to beat a known deadline, that is front-running: you buy in May what you would have bought in August, so the calendar, not the economy, moves. Front-loaded imports subtract from this quarter's GDP and add it straight back next quarter. An accounting head-fake. The tell is *where* the imports land.

Read the imports (top four bars) and the fingerprint is clear. They rose most in goods you can warehouse and sit on, autos (+6.3%) and consumer goods (+5.7%), and **least in capital goodscapital goodsMachinery and equipment that businesses buy to produce other goods; usually ordered to specification and hard to stockpile in advance. (+0.4%)**, the machinery firms order to spec and cannot stockpile in advance. A broad, demand-driven boom would lift everything. This is concentrated in exactly the storable categories you pull forward to beat a cost increase. Not proof, but the signature of front-running rather than underlying strength.

**1. The tarifftariffA tax a government charges on imported goods, which raises their cost to domestic buyers. cliff on 24 July.** The administration's Section 122Section 122A U.S. trade law that lets the president impose a temporary universal tariff of up to 15% to address a trade deficit, but only for 150 days unless Congress extends it. "universal" 10% tariff, which a buyer pays as a tax on imported goods, took effect 24 February 2026 under a statute that caps such measures at 150 days. The clock runs out **24 July** (24 February plus 150 days), and the President cannot extend it without Congress. The 10% is not expected to simply vanish: the U.S. Trade Representative has described Section 122 as a "bridge" to country-specific Section 301Section 301A U.S. trade law used to impose tariffs on specific countries deemed to trade unfairly; here, country-specific rates expected to replace the expiring universal Section 122 tariff. tariffs expected on or around the same date, some of them above 10%. So for goods from those countries, the rational move is to land the container now, at 10%, before any step-up. A textbook front-run with a hard, dated catalyst.

**2. A war still hot in May.** Conflict around Iran disrupted the Strait of HormuzStrait of HormuzA narrow shipping channel between Iran and the Arabian Peninsula through which a large share of the world's seaborne oil passes; disruptions there can spike oil prices., the narrow channel that carries a large share of the world's seaborne oil, through the spring, and survey sponsors attributed part of the import surge to firms stocking up to avoid shortages and price spikes tied to the fighting. By mid-to-late June the U.S. and Iran had signed a framework to end the war, Gulf oil flows were recovering toward roughly three-quarters of prewar levels, and Brent crudeBrent crudeA global benchmark price for oil, widely used as a reference for crude traded internationally. (the main international oil benchmark) had fallen back near $72 a barrel, its lowest since February. That motive was draining away. Both reasons share a feature: they are about *timing*, not appetite. When the calendar passes them, the buying they pulled forward leaves a hole, not a trend.

Here is where the tidy "two stories" framing has to be disciplined. If a spike in oil prices had inflated the dollar value of *industrial-supplies imports* without more barrels actually crossing the border, then the *same* oil move cuts the other way on exports. Petroleum and refined products are a major U.S. export. As Brent round-tripped lower into late June, the dollar value of industrial-supplies exports fell 7.0 percent even if the physical volume of barrels shipped barely changed. That slice of the export drop is a *price* effect, not a *demand* effect, and like the import side it can reverse when prices settle. Call it the export half of the oil illusion.

Strip that piece out and a cleaner signal remains: **consumer-goods exports fell 9.2 percent**, the steepest drop of any major export category. Consumer goods are not priced off a swinging commodity benchmark, so a fall that size is harder to wave away as an accounting artifact. And note what was *not* working against these exports: the dollar in 2026 has been weaker, not stronger, than a year earlier, and a weaker dollar makes American goods cheaper abroad. Exports falling anyway, with the currency tailwind at their back, is the kind of number that can reflect genuine softening in foreign demand, whether because trading partners are slowing or because retaliation against U.S. tariffs is biting. One month does not tell you which. But it is the residual the Fed cannot simply look through.

"The widening trade deficit is bad news for national income growth, and it suggests that net exportsnet exportsExports minus imports; because imports are subtracted in the GDP formula, a jump in imports mechanically lowers measured GDP even if nothing else changes. might drag down real GDP growth too," Carl Weinberg, chief economist at High Frequency Economics, told Reuters after the release, capturing the worry beneath the front-run noise.

Why does a trade gap move GDP at all? In the GDP formula, imports are subtracted and exports added, so net exports (exports minus imports) feed straight into the growth number. A front-loaded import surge therefore drags measured growth in the quarter the goods arrive, then flatters it the next quarter when imports fall back. That is not a forecast; it is arithmetic, and the economy demonstrated it last year.

In early 2025, a record rush to beat tariffs made net exports a record drag on growth, knocking Q1 GDP to -0.6 percent. The very next quarter the same flow reversed: imports fell back, net exports swung to a near-equal *boost*, and GDP printed +3.8 percent. A near-complete round-trip. Crucially, one cushion sat underneath it: goods that pile up unsold still count as output, so rising inventoriesinventoriesThe stock of unsold goods businesses hold; rising inventories can partly offset a trade drag in GDP because goods produced for stock still count as output. can offset part of a trade drag inside GDP even before the reversal arrives. The May data points the same direction, but we deliberately plot no Q2 2026 bar, because no official contribution figure exists yet and we will not invent one.

Forecasters reacted the way the arithmetic implies. After the report, the Atlanta Fed's GDPNowGDPNowThe Atlanta Federal Reserve's running model estimate of current-quarter GDP growth, refreshed through the quarter as fresh data comes in. nowcastnowcastA real-time estimate of current-quarter GDP that updates as new data arrives, such as the Atlanta Fed's GDPNow., a model estimate of current-quarter growth that updates with each data release, was cut to 2.5 percent from 3.0 percent a week earlier; Morgan Stanley trimmed its Q2 call to 2.1 from 2.5 percent, and Goldman Sachs to 2.2 from 2.4.

Here the two stories split on policy. The import surge is the kind of noise a central bank is supposed to see through: it drags one quarter, reverses the next, and tells you little about the trend. But a genuine softening in export demand is exactly the signal the Fed should *not* look through, because it can persist. The same $105.8 billion headline therefore carries opposite instructions depending on which half you weigh, and the only way to tell them apart is to strip the common oil-price swing out of both.

So the answer to the opening question is: it depends which side you mean. The import blowout looks like a datable front-run with a hard 24 July deadline and a now-fading war premium, the sort of move that should largely round-tripround-tripA move in the data that reverses itself, ending roughly where it started; here, purchases pulled forward into one quarter that drop back the next, cancelling out. out of the data by autumn, just as the 2025 episode did. The export drop is split: the oil-driven part is a price illusion that will reverse with crude, but the 9.2 percent fall in consumer-goods exports is a cleaner read on demand that may not. That residual, small in dollars but stubborn in meaning, is the part worth watching.

A final discipline: these are advance Census figures that will be revised, and a single month is a print, not a trend. The honest read of May is not "the trade picture is deteriorating" or "it is all noise." It is that one number contains a piece that reverses on schedule and a smaller piece that might not, and the work, for the Fed and for everyone else, is keeping the two apart.