Why Did US Durable Goods Orders Surge 1.8% in September? What the Data Really Shows

 

The Census Bureau reported on October 27 that September new orders for durable goods rose 1.8% to $345.4 billion, the sixth increase in seven months, against a market expectation of 0.6%.

 

That headline is almost entirely an aircraft story. Transportation equipment orders rose 4.6%, an increase of $5.2 billion, while the total increase derived from the published figures comes to roughly $6.1 billion. Transportation equipment alone accounted for roughly 85% of the total gain. Excluding transportation, orders rose just 0.5%. The driver within transportation was commercial aircraft orders, up 28.4%.

 

The divergence across capital goods definitions makes the point more sharply. Nondefense capital goods orders rose 6.2% to $107.8 billion, while core capital goods orders excluding aircraft rose just 0.5% to $88.0 billion. On this article's calculation, roughly 93% of the $6.3 billion increase in nondefense capital goods came from the aircraft-related portion.

 

That does not amount to weak business investment, for two reasons. Core capital goods shipments rose 1.2%, ahead of the 0.5% in orders, and shipments are what feeds directly into equipment investment in gross domestic product. Core capital goods orders also rose 10.9% year over year, a figure that reflects the trend better than a single month's 0.5%.

 

One technical note is worth knowing before citing any of these figures: the Census Bureau states in this report that "statistical significance is not measurable for this survey," because the estimates are not based on a probability sample.

 

 

 

Total orders rose 1.8% to $345.4 billion, against an expectation of 0.6%. This was the sixth increase in the last seven months.

 

Transportation equipment accounted for roughly 85% of the total gain. It rose $5.2 billion against a total increase of roughly $6.1 billion, leaving orders excluding transportation up just 0.5%.

 

Commercial aircraft orders rose 28.4%. Nondefense capital goods orders rose 6.2%, but core capital goods excluding aircraft rose only 0.5%.

 

Core capital goods shipments rose 1.2%, ahead of orders. Shipments are what feeds directly into equipment investment in gross domestic product; orders do not.

 

Core capital goods orders rose 10.9% year over year. That figure reflects the business investment trend better than the single month's 0.5%.

 

Defense capital goods orders fell 1.8%. Orders excluding defense rose 2.2%, above the 1.8% headline, making defense a drag.

 

The Census Bureau states that statistical significance is not measurable for this survey. The estimates are not based on a probability sample, so sampling error cannot be measured and confidence intervals cannot be computed.

 

 

 

Transportation's contribution can be calculated directly. Total orders rising 1.8% to $345.4 billion implies an increase of roughly $6.1 billion, while transportation equipment alone rose $5.2 billion, a share of roughly 85%. That calculation is made here from the published figures.

 

Defense ran the other way. Orders excluding defense rose 2.2%, above the 1.8% headline, while defense capital goods orders fell 1.8% for a decrease of $0.4 billion, making defense a drag this month rather than a contributor.

 

August total durable goods orders rose 0.1%. Note that the revision table at the end of the report covers total manufacturing including nondurables rather than durable goods alone, with new orders revised from $663.5 billion to $664.9 billion and shipments from $658.6 billion to $659.5 billion, both marginally higher; those figures should not be mixed with the durable goods series.

 

 

The Census Bureau states in this report that "statistical significance is not measurable for this survey," explaining that the Manufacturers' Shipments, Inventories and Orders estimates "are not based on a probability sample," so sampling error cannot be measured and confidence intervals cannot be computed.

 

This needs stating precisely, because it differs from other Census data products. Some surveys, housing starts for instance, publish 90% confidence intervals that allow a judgement on whether a given change is statistically significant. The durable goods report is not in that category, so it is equally wrong to say that a 0.5% change "falls within the margin of error" and to say it is "significantly different from zero" since the survey supplies no computable margin and neither claim has a basis.

 

The practical implication is that small monthly changes should not be used as precise values, particularly a near-zero reading such as the 0.5% in core capital goods orders. The reason the 10.9% year-over-year figure is more reliable is precisely that it spans twelve months and does not depend on single-month precision.

 

 

The report contains three capital goods measures that are easily confused, and their divergence this month is exactly what makes the point.

 

 

The roughly $19.8 billion difference between nondefense capital goods and core capital goods is the aircraft and related portion. On this article's calculation, nondefense capital goods rose from roughly $101.5 billion to $107.8 billion, an increase of roughly $6.3 billion, while core capital goods rose from roughly $87.6 billion to $88.0 billion, an increase of roughly $0.44 billion. So roughly 93% of the increase in nondefense capital goods came from the aircraft-related portion, with roughly 7% from every other equipment category combined.

 

That is precisely why the core capital goods definition exists. Aircraft orders are extremely lumpy: a single wide-body order can run into billions of dollars, so one airline's procurement decision can change an entire month's figure, and what it reflects is an airline's fleet planning rather than broad business appetite for equipment. This month's 28.4% rise in commercial aircraft orders is a textbook display of that lumpiness.

 

So in describing business investment, the figure to cite is core capital goods at 0.5%, or the more robust 10.9% year over year, rather than nondefense capital goods at 6.2% or total orders at 1.8%. Those latter two principally reflect aircraft this month.

 

 

These two measures are frequently treated as the same thing, yet they enter the economic data in entirely different ways.

 

Orders are a leading indicator and do not enter gross domestic product directly. New orders reflect how much equipment businesses have contracted for future delivery, which signals future production activity without constituting output in the period the order is signed.

 

Shipments are the current-period measure and feed directly into the equipment investment component of gross domestic product. Core capital goods shipments are among the principal inputs for estimating equipment investment.

 

This month the two pointed the same way at different magnitudes: core capital goods orders rose 0.5% while core capital goods shipments rose 1.2% to $86.1 billion, making shipments the stronger of the two. Total durable goods shipments rose 0.9% to $336.8 billion, an increase of $3.0 billion.

 

That combination carries a specific implication for the forthcoming third-quarter gross domestic product figure: what determines this quarter's equipment investment contribution is the 1.2% in shipments, not the 0.5% in orders. Reading only the orders figure therefore understates the support this data provides to current-quarter growth. For the week's macro calendar overall, see our compilation on this Federal Reserve decision and what markets are watching.

 

Unfilled orders offer a third angle. That measure rose 0.8% to $1,623.6 billion, an increase of $13.6 billion, meaning new orders flowed in faster than shipments consumed them and the backlog grew. That supports future production, though it may also reflect lengthening delivery times, and public data cannot distinguish between the two causes.

 

Inventories rose 0.3% to $610.0 billion, a smaller increase than the 0.9% in shipments, indicating inventories did not accumulate relative to demand this month.

 

 

The brief lists AI equipment investment as an analytical angle, and one definitional point needs establishing first: the Census Bureau report contains no data category for artificial intelligence.

 

What the report publishes is data by industry classification, within which computers, machinery and motor vehicles were listed as contributors to this month's durable goods increase. Attributing the core capital goods increase to AI buildout is an inference economists draw from the computers and electronic products component, not a classification disclosed in the data.

 

The analytical views in circulation include Reuters attributing the core capital goods increase to the AI boom, Bloomberg economist Eliza Winger saying AI-related investment "continued to power capital spending through the quarter," and a Citi Wealth report dated October 7 holding that the AI buildout sits in its early-to-middle stages, citing semiconductor demand, core capital goods orders and AI-related exports as evidence.

 

Those views are reasonable, and the 10.9% year-over-year growth in core capital goods does require a structural explanation. But citations should mark this as analytical inference rather than a data classification, because the report supplies neither a dollar figure for AI-related spending nor any data centre breakdown.

 

Winger also raised a qualification worth noting: some firms say their investment plans depend on "the outcome of the midterm elections" and the path of Federal Reserve policy. With the US midterm elections only days away, that means the decision environment reflected in this month's capital goods data may shift within weeks. Put differently, the strong 10.9% year-over-year figure rests on a policy premise that has not yet been settled.

 

For the financing structures and scale behind AI infrastructure investment, see our earlier analysis of CoreWeave's financing structure.

 

 

For third-quarter gross domestic product, watch the 1.2% in core capital goods shipments. That measure is among the principal inputs for estimating equipment investment and came in ahead of the 0.5% in orders, with total shipments up 0.9% for an increase of $3.0 billion. Reading only the orders figure understates durable goods' contribution to current-quarter growth.

 

For industrials and the aerospace chain, aircraft and other equipment need separating. Commercial aircraft orders rising 28.4% drove transportation equipment up 4.6%, but that reflects airlines' fleet procurement decisions, which are lumpy and do not represent broad equipment demand. Orders excluding transportation rose just 0.5%. This month's data is therefore more informative about the aerospace chain than about general industrial equipment.

 

For corporate financing costs, unfilled orders and the year-over-year rate are the steadier references. Unfilled orders rising 0.8% to $1,623.6 billion indicates a growing backlog, and core capital goods up 10.9% year over year indicates durability in equipment investment demand. But with some firms already saying investment plans depend on the policy path, changes in financing costs may feed directly into subsequent order flow.

 

For risk assets including crypto, transmission is indirect. Durable goods data does not act on crypto markets directly; its influence runs through economic growth and rate expectations, where business investment data shapes the market's reading of growth strength, which shapes policy path expectations and the broad pricing environment for risk assets. Every link carries uncertainty, and this release itself divides between a strong total, a weak core and stronger shipments, so it supports no directional judgement. To observe how prices actually respond around macro releases, view live BTC/USDT quotes and depth on MEXC.

 

 

For James Mitchell, this release is a textbook demonstration of why the headline figure in macro data is almost always the one least worth citing.

 

The 1.8% came in at more than triple the 0.6% expected, reading like a strong signal for business investment. But break the increase apart: total orders rose roughly $6.1 billion, of which transportation equipment alone contributed $5.2 billion, about 85%. Strip transportation and 0.5% is what remains. And the driver within transportation was commercial aircraft orders at 28.4%. So this month's strong durable goods orders are, in substance, a few airlines signing procurement contracts.

 

The capital goods divergence is even blunter. Nondefense capital goods rose 6.2% while core capital goods excluding aircraft rose only 0.5%. On my arithmetic, of the roughly $6.3 billion increase in nondefense capital goods, roughly 93% came from aircraft, leaving every other equipment category contributing roughly 7% between them. That is the entire reason the core capital goods definition exists: a single wide-body order can run into billions, one fleet-planning decision can rewrite a month, and none of it bears on whether businesses are buying machine tools, servers or trucks.

 

I would not write the conclusion as weak business investment, though, for two specific reasons.

 

The first is shipments. Core capital goods shipments rose 1.2%, ahead of the 0.5% in orders. That matters for tomorrow's third-quarter gross domestic product, because what enters equipment investment is shipments, not orders. Orders are a leading indicator signalling future production; shipments are current output. Reading only that 0.5% understates the support this data gives current-quarter growth. The distinction is omitted from reporting almost without exception.

 

The second is the year-over-year rate. Core capital goods orders rose 10.9% year over year. That is far more reliable than a single month's 0.5%, and the reason it is more reliable is exactly the technical note I come to next.

 

The Census Bureau wrote a sentence in this report that is almost never cited: statistical significance is not measurable for this survey, because the estimates are not based on a probability sample, so sampling error cannot be measured and confidence intervals cannot be computed. This has to be understood precisely. It does not mean the 0.5% "falls within the margin of error," since that claim requires a computable margin the survey does not supply. Nor does it mean the 0.5% is significantly different from zero. It means neither judgement can be made. So for a near-zero monthly reading, the right response is not to argue about significance but to stop treating it as a precise value and look instead at the year-over-year figure, which spans twelve months and does not depend on single-month precision.

 

On the AI attribution, one definitional caution. This report has no artificial intelligence data category. What it gives is industry classifications, with computers, machinery and motor vehicles listed as contributors. Attributing core capital goods growth to AI buildout is an inference economists draw from the computers and electronic products component. It is a reasonable one, since 10.9% year over year does require a structural explanation, but it is inference rather than disclosure. The report supplies no dollar figure for AI-related spending and no data centre breakdown. When a write-up says durable goods data confirms AI investment, one step in that evidence chain is conjecture.

 

The last point strikes me as the most underrated. Bloomberg economist Winger noted that some firms say their investment plans depend on the midterm election outcome and the Federal Reserve's policy path. The midterms are days away. That means the strong 10.9% figure rests on a policy premise that will change within weeks. For anyone using this release to project equipment investment over coming quarters, that qualification matters more than any component figure.

 

 

 

The Census Bureau reported September new orders for durable goods up 1.8% month over month to $345.4 billion, the sixth increase in seven months, against a market expectation of 0.6%. The data was released on October 27, 2026 at 8:30 a.m. Eastern.

 

 

Because transportation equipment alone accounted for roughly 85% of the total gain. Total orders rising 1.8% implies an increase of roughly $6.1 billion, while transportation equipment rose $5.2 billion, driven within that category by commercial aircraft orders up 28.4%. Excluding transportation, orders rose just 0.5%. That share is calculated here from the published figures.

 

 

That conclusion does not follow, for two reasons. Core capital goods shipments rose 1.2%, ahead of the 0.5% in orders, and shipments are what feeds directly into equipment investment in gross domestic product. Core capital goods orders also rose 10.9% year over year, a twelve-month figure that reflects the trend better than a single month. The Census Bureau also states that statistical significance is not measurable for this survey, so a near-zero monthly change should not be used as a precise value.

 

 

Core capital goods is nondefense capital goods excluding aircraft. September nondefense capital goods orders were $107.8 billion, up 6.2%, while core capital goods were $88.0 billion, up 0.5%, with the roughly $19.8 billion difference being aircraft and related items. On this article's calculation, roughly 93% of the $6.3 billion increase in nondefense capital goods came from aircraft. Aircraft orders are extremely lumpy, with a single wide-body order running into billions, so business equipment investment should be gauged using the core definition.

 

 

It depends on the purpose. Orders are a leading indicator reflecting how much equipment businesses have contracted for future delivery, signalling future production without constituting current output. Shipments are the current-period measure, and core capital goods shipments are among the principal inputs for estimating equipment investment in gross domestic product. This month orders rose 0.5% while shipments rose 1.2%, so judging this quarter's growth contribution means reading shipments while judging production over coming quarters means reading orders.

 

 

The report itself has no artificial intelligence data category; it publishes data by industry classification, within which computers, machinery and motor vehicles were listed as contributors. Attributing core capital goods growth to AI buildout is an economists' inference, held by Reuters, a Bloomberg economist and Citi Wealth among others, and the 10.9% year-over-year growth does require a structural explanation. But the report supplies no dollar figure for AI-related spending and no data centre breakdown, making this inference rather than disclosure.

 

 

Because the Census Bureau states in the report that the estimates "are not based on a probability sample," so sampling error cannot be measured and confidence intervals cannot be computed, and it states explicitly that "statistical significance is not measurable for this survey." This differs from other Census data products that publish 90% confidence intervals. The practical implication is that one can neither say a small change falls within the margin of error nor say it is significantly different from zero, and near-zero monthly readings should not be treated as precise values.

 

 

The key figure is core capital goods shipments up 1.2% to $86.1 billion, which is among the principal inputs for estimating equipment investment and came in ahead of the 0.5% in orders. Total durable goods shipments rose 0.9% to $336.8 billion, an increase of $3.0 billion. Reading only the orders figure understates durable goods' support for current-quarter growth. Unfilled orders rose 0.8% to $1,623.6 billion, indicating a growing backlog, though public data cannot distinguish whether that reflects demand support or lengthening delivery times.

 

 

The information above is provided for general macroeconomic and market information only and does not constitute investment advice, financial advice, legal advice, tax advice or a recommendation to buy or sell securities. The data cited comes from the Census Bureau's advance report on durable goods manufacturers' shipments, inventories and orders for September 2026, released on October 27, 2026, together with related public reporting, verified on October 28, 2026. That release is an advance report with data seasonally adjusted but not adjusted for price changes, and subject to subsequent revision. The Census Bureau states in the report that the estimates are not based on a probability sample, so sampling error cannot be measured and confidence intervals cannot be computed, and that statistical significance is not measurable for this survey; accordingly no small monthly change in this article should be characterised either as falling within a margin of error or as significantly different from zero, and near-zero monthly readings should not be treated as precise values. Figures identified as this article's calculations, including transportation equipment's share of the total gain, the aircraft-related share of the increase in nondefense capital goods, and the derived increase amounts, are not figures published by the Census Bureau; they are calculated from published data and actual results may differ on a different basis. The revision table at the end of the report covers total manufacturing rather than durable goods alone and those figures should not be mixed with the durable goods series. The report contains no data classification for artificial intelligence, and content attributing growth to AI buildout reflects analytical inference by economists and research firms rather than data disclosure, with the report supplying no dollar figure for AI-related spending and no data centre breakdown. The statement that business investment plans depend on policy premises comes from a third-party economist's view. Transmission between macroeconomic data and risk asset prices is indirect and subject to multiple uncertainties, this release itself divides between total, core and shipments measures, and no single data point supports a directional judgement. Crypto asset prices are highly volatile and leveraged products can produce losses exceeding the initial outlay. Readers should consult the original data release before making any decision, judge against their own financial circumstances, investment objectives and risk tolerance, and consult a qualified professional where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from the use of this information.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise: Technical Analysis, Market Trends and Cycles, Trading Strategies, Bitcoin and Altcoin Analysis, Risk Management.

 

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)