September’s construction slowdown was concentrated in commercial projects. Starts in that category fell 18.4%, a sharp move that outweighed stronger activity in education, highways and other institutional work. The headline decline in total construction was much smaller, but that gap is the point: a modest change across the full market can conceal steep weakness in a sector that accounts for large projects. The monthly result needs to be read by category, not treated as a single verdict on construction demand.
Total construction starts fell 1.5% from August to a seasonally adjusted annual rate of $1.33 trillion, Dodge Construction Network reported. That followed August’s 24.8% monthly drop, which the report attributed to a slowdown in megaprojects. The two months tell different stories about the same measure: August’s retreat was tied to fewer megaproject starts, while September’s sector mix showed commercial weakness alongside gains elsewhere. Starts can move sharply when the timing of a small number of large projects changes. That makes a monthly reading useful as a snapshot, but a poor basis for assuming that every part of the market is moving in the same direction.
Commercial weakness sets the pace
Nonresidential starts declined 6.3% in September, with commercial construction driving the fall. Commercial starts were down 18.4%, while institutional starts rose 16.3%. Those opposing movements matter because a broad category can look relatively steady even when its parts are pulling in sharply different directions. Education and highway projects helped offset the commercial decline and limit the total retreat. The gains didn’t erase the weakness; they changed how much of it showed up in the headline figure.
“Construction starts edged lower in September, but gains in institutional and nonbuilding construction helped cushion the decline,” said Sarah Martin, Dodge’s director of economic research. She said those sectors countered weakness in commercial and single-family starts. Her account points to a market with competing signals rather than a uniform downturn. For readers tracking construction, the distinction between the direction of total starts and the direction of specific sectors is practical: the headline captures the net result, while the category figures show where the pressure is concentrated.
Institutional gains also deserve a measured reading. A rise in starts for education and other institutional work suggests that those categories supported September’s activity, but one month doesn’t establish a durable trend. The same caution applies to highways and other nonbuilding work, which helped cushion the decline. Starts measure projects entering the pipeline; they don’t by themselves say how quickly those projects will proceed or how much activity will appear in later periods. The data here support a comparison of sectors and months, not a claim about the eventual pace of construction.
Why the mix matters
Commercial projects can be large enough that changes in their timing have an outsized effect on a monthly total. Similar measures of project starts often reflect a mix of many smaller jobs and a smaller number of large ones, so the composition can matter as much as the direction. That’s why August’s reported megaproject slowdown and September’s sharp commercial decline should be kept in view together, while still being treated as distinct explanations. The report links August’s drop to megaprojects; it doesn’t establish that the same factor explains September’s commercial weakness.
There’s a difference between saying construction starts fell and saying construction broadly contracted. The first is supported by September’s 1.5% monthly decline in the total. The second would require a stronger claim about how widespread and persistent the weakness is. In September, institutional activity rose even as commercial and single-family starts weakened, so the evidence points to uneven movement. That unevenness isn’t a technical footnote. It changes what the data can reasonably tell a reader about the market.
This is more significant than it looks: the smaller headline decline can make the month seem calm, even as commercial starts drop sharply. At the same time, the commercial figure alone shouldn’t be stretched into a prediction about future conditions. One month can identify where a change occurred; it can’t settle whether that change will last. (And this is the part most people overlook.) The report gives a clear account of September’s mix, but persistence is still an open question.
What to watch next
The next useful signal is whether commercial weakness persists or whether September proves to be another uneven month in a volatile stream of large projects. A subsequent month with similar weakness would make it harder to dismiss the decline as a short-lived fluctuation, while stronger commercial starts would change the picture. Neither outcome can be inferred from this report alone. The comparison should also account for institutional and nonbuilding activity, since those categories helped limit September’s retreat.
If you’re working in this space, the practical takeaway is to keep the aggregate and sector readings side by side. The total starts figure describes the net monthly change; the category figures show which activity lifted or pulled down that result. A single headline can blur those differences. September’s data are mixed, and that’s the most defensible reading.
For now, the evidence is narrow but useful. Total starts slipped after a much steeper August decline, and commercial construction supplied the clearest point of weakness in September. Education, highways and other institutional work offered a counterweight. Whether that balance holds will depend on the next readings. Not a settled trend.
