US Q2 GDP Preview (July 2026): Core Capex Shipments Jump 1.9%, the Most Since 2021 — but GDPNow Slipped to 1.6% Before the July 30 Print. What It Means for the Dollar
The last major hard-data input before Thursday's advance GDP estimate landed on Monday 27 July, and it pointed three ways at once. Headline durable goods orders rose just 0.3% in June, badly missing a consensus near 2.5%. Underneath, orders for nondefense capital goods excluding aircraft — the business-investment core — rose 0.9% with May revised up to 1.9%, and core capital goods shipments, the line that feeds straight into the GDP accounts, surged 1.9%, the biggest monthly gain since December 2021. And yet the Atlanta Fed's GDPNow nowcast slipped to 1.6% from 1.7%. Three signals, three directions — and the Bureau of Economic Analysis publishes the verdict at 8:30 a.m. Eastern on Thursday 30 July, barely eighteen hours after the Federal Reserve decides.
This is a case study in why a fundamental read of a currency beats a price-only one. A headline-watcher saw "durable goods 0.3% versus 2.5% expected" on Monday and read a growth scare. That reading is wrong in a specific, instructive way: the headline series is dominated by lumpy aircraft and defence orders that have almost nothing to do with the pace of the economy, while the component that actually enters the GDP calculation went the other way and hit a four-and-a-half-year high. The growth factor that moves the dollar keys off underlying demand and what it implies for the rate path — not off a top line that tariffs and aircraft cycles have been whipsawing for three quarters. Decompose the report, and 30 July stops being a single number and becomes a read on whether the US economy is still outgrowing its peers.
- The advance estimate of Q2 2026 GDP lands Thursday 30 July at 8:30 a.m. ET — the first of three estimates, and the market-moving one.
- June durable goods (27 July) missed at +0.3% against a ~2.5% consensus, after a revised −4.0% in May — but the miss sat in transportation, not in demand.
- Core capital goods shipments rose 1.9%, the largest gain since December 2021; core orders rose 0.9% (vs 0.8% expected) and are 9.3% higher year-on-year.
- GDPNow fell to 1.6% on 27 July from 1.7% on 17 July, as the nowcast for real private domestic investment growth was cut from 6.0% to 5.5%.
- Forecasters sit higher: the Philadelphia Fed's SPF and a Reuters poll both see 2.1%, matching Q1's final pace. The market's band is roughly 1.6–2.1%.
- Timing is the hook: the Fed decides 29 July with no Q2 GDP in hand and no new dot plot, so the number arrives the next morning as the first verdict on that stance.
- See how the growth and interest-rate factors are scoring the dollar right now on the live meter.
What actually landed on 27 July: the last hard input
The Census Bureau's advance report on durable goods is the final substantive piece of source data before the BEA closes its books on the quarter, and June's edition was unusually two-faced.
The top line disappointed. New orders for manufactured durable goods increased 0.3% to $334.8 billion, against expectations clustered near 2.5%, following a revised 4.0% decline in May. Strip out transportation — the category where a single wide-body aircraft order can swing the aggregate by a percentage point — and orders rose 0.6%. Excluding defence, they rose 0.3%. Computers and electronic products led the gainers, up 3.1% to $31.1 billion. Unfilled orders for transportation equipment rose $4.1 billion, or 0.4%, to $1.002 trillion, a reminder that the backlog is still building even in the month the headline stumbled.
The core told a different story. Orders for nondefense capital goods excluding aircraft — the standard proxy for what businesses are committing to spend on productive equipment — rose 0.9% in June, ahead of the 0.8% economists polled by Reuters had expected, and May's increase was revised up to 1.9% from an initially reported 1.4%. On a year-over-year basis, core orders stood 9.3% higher.
Most important for Thursday's arithmetic: core capital goods shipments jumped 1.9% after a 0.2% rise in May. Shipments, not orders, are what the BEA feeds into the equipment-investment line of GDP, and a 1.9% monthly gain is the strongest since December 2021. Economists reading the report expected another quarter of double-digit annualised growth in business equipment spending. Christopher Rupkey of FWDBONDS argued corporate capital spending is "keeping the economy afloat" while other sectors stay cautious; Bernard Yaros of Oxford Economics noted the strength is not confined to the artificial-intelligence build-out but also reflects a rebound in company vehicle purchases; and Priscilla Thiagamoorthy of BMO Capital Markets flagged the double edge — strong capex supports activity and productivity, but can also sustain price pressure.
Why GDPNow still fell: composition, not contradiction
Here is the part that trips up a quick reading. Despite the strongest core capex shipments in four and a half years, the Atlanta Fed's GDPNow model cut its Q2 estimate on 27 July, to 1.6% annualised from the 1.7% it had carried since 17 July. Its commentary attributed the move to the day's Census releases, which lowered the nowcast for second-quarter real gross private domestic investment growth from 6.0% to 5.5%.
That is not the model disagreeing with the capex story. GDPNow aggregates every component of the accounts, and the same batch of Census data that carried the strong equipment numbers also carried inventory and trade detail. Equipment is one line inside gross private domestic investment; inventories are another, and an inventory contribution marked lower can more than offset a stronger equipment read. The headline GDP figure is the sum of all of it.
For the dollar, the distinction matters enormously. A 1.6% print built on strong business equipment and soft inventories describes a fundamentally healthier economy than a 1.6% print built on weak demand — inventories unwind and rebuild, while capital spending signals what firms believe about the next several years. This is the same decomposition logic that separates headline from core inflation, applied to the growth side: the top line is an accounting identity, and the factor read lives underneath it.
What the nowcasts say now: GDPNow 1.6% versus forecasters 2.1%
The trackers have drifted half a point apart, which is itself informative.
| Tracker | Q2 2026 estimate | As of | What it captures |
|---|---|---|---|
| Atlanta Fed GDPNow | 1.6% | 27 July (from 1.7% on 17 July) | Mechanical nowcast; full trade and inventory drag priced in |
| Philadelphia Fed SPF | 2.1% | Q2 survey | Economist survey; smooths one-off swings |
| Reuters poll of economists | 2.1% | 27 July | Median forecast, matching Q1's pace |
| Q1 2026 (third estimate) | 2.1% | Final | The baseline the Q2 print is measured against |
GDPNow, being purely data-driven, has fully absorbed the trade and inventory drags. The forecaster surveys lean on core demand and smooth through one-off swings, which is why they sit a few tenths higher. The practical anchor for Thursday is the roughly 1.6–2.1% band: a print inside it barely touches the rate path, while a number outside it — say sub-1%, or above 2.5% — is the surprise that moves the dollar.
When it lands, and why the timing is the story
The BEA publishes the advance estimate on Thursday 30 July at 8:30 a.m. Eastern, per its GDP release schedule. It is the first of three passes at the quarter — a second estimate in late August, a third in late September — and the advance is the one markets trade, because it is the earliest official measure of April-to-June output.
What makes this release unusual is the calendar around it. The Federal Open Market Committee announces on Wednesday 29 July at 2:00 p.m. Eastern — a meeting mapped in the July FOMC preview — and the committee decides without the Q2 GDP number, which does not exist until the next morning. It is also a non-projection meeting, so there is no fresh dot plot. The market therefore trades the Fed's guidance on Wednesday afternoon, then receives the first hard read on whether the economy justified that guidance on Thursday at the open. When the two agree, GDP merely confirms the rate path; when they clash, the dollar reprices twice in a day.
The Q1 baseline and the tariff distortion
Q2 will be measured against a first quarter that was itself a moving target: the advance estimate came in at 2.0% on 30 April, was revised down to 1.6% on 28 May, then settled back at 2.1% in the third estimate. Three prints, a half-point range — a reminder that the advance figure is an early draft, not a final verdict.
Composition was the bigger story, and it is where the tariffs left their fingerprints. Ahead of the duties, importers front-loaded shipments; because imports subtract in the GDP accounts, net trade became a large, misleading drag that understated how much the domestic economy was actually spending. As that front-loading unwound, the trade line swung back the other way. For three quarters the headline has been pushed around by flows that say more about tariff timing than about demand — which is why analysts lean on final sales to private domestic purchasers, consumer spending plus business fixed investment, stripped of trade and inventories, as the cleaner gauge of underlying momentum. Monday's core capex numbers feed directly into that measure, and they went up.
From GDP to the dollar: the growth factor
For the dollar, this release transmits through the growth factor — one of the five fundamental factors a currency-strength model tracks — with the interest-rate factor amplifying it.
Stronger-than-expected growth supports the dollar on two channels at once: it keeps the Fed's path higher for longer, widening the expected rate gap versus peers, and it signals an economy that attracts capital. Weaker growth runs the film backwards. What matters on Thursday is not the level but the gap versus what is already discounted — and, after Monday, the quality of that gap. A beat carried by business equipment now has corroborating evidence behind it; a beat carried by inventories does not.
The context the number arrives into
The print does not land in a vacuum. The run of data into it has been mixed rather than uniformly soft: June nonfarm payrolls shocked at just 57,000 with prior months revised down, detailed in the 57K payrolls breakdown; June CPI cooled to 3.5% headline with core at 2.6%, covered in the June CPI report; and retail sales slowed to 0.2% — but capital spending, as Monday showed, has not rolled over. Pulling the other way is the 1 August tariff cliff, mapped in the August tariff-cliff breakdown, an upside risk to prices landing two days after GDP. The same Thursday morning also brings June core PCE, previewed in the core PCE preview, plus the Bank of England decision and weekly jobless claims, so cross-currents will be live.
The three scenarios for July 30
| Scenario | What it looks like | Growth/rate-factor read | Likely dollar reaction |
|---|---|---|---|
| Hot print (≥2.5%) | Beat led by firm consumer spending and the equipment line Monday flagged | Reinforces a hawkish hold; September hike kept live | USD firm; front-end yields rise |
| In line (~1.6–2.1%, base case) | Steady growth near consensus; trade and inventory drag as expected | Rate path unchanged; confirms the Fed's stance | Muted; dollar trades composition and Fed tone |
| Cold print (<1%) | Sharp slowdown reaching core demand, not just inventories | Pulls first-cut expectations forward | USD softer; growth premium fades |
The asymmetry worth noting: after a soft labour-market run and with the Fed already expected to hold, the market is arguably better braced for a weak number than a strong one — and Monday's capex data quietly raised the odds that the upside surprise is the one that lands. A hot print clashing with a cautious Wednesday hold would be the larger shock relative to positioning, and it would arrive the day after the decision, when there is no press conference left to soften it.
What to watch when the number drops
Read it in this order. First, the headline versus the ~1.6–2.1% band — that gap sets the initial move. Second, and more important, final sales to private domestic purchasers, the core-demand gauge that tells you whether a surprise is real momentum or trade-line noise. Third, the business equipment line, where Monday's 1.9% shipments jump should show up, and the inventory contribution, the component GDPNow marked down. Fourth, the consumer-spending line, the economy's largest engine and the one the softening jobs and retail data put in question. Then read all of it against Wednesday's Fed tone: the market's verdict on whether the growth data confirms or contradicts the committee is where the dollar's larger move usually happens.
None of this is a trade signal or a forecast dressed up as certainty. It is a map: which outcome tips which scenario, and how each reads through the growth and interest-rate factors that drive the dollar. On 30 July the map turns into a data point — and the market's read of whether the US is still outgrowing its peers, far more than the single headline number, is what the euro, the pound, the yen and the rest will trade against.
For more on how currency strength is built from fundamentals rather than price, see the about page, and for the dollar specifically, the USD currency page.
Educational macro context only — not investment advice.