Flat is the generous reading. U.S. cement shipments, including imports, came to an estimated 8.94 million metric tons in May, according to the U.S. Geological Survey’s latest Mineral Industry Survey, posted Sept. 28. That’s 0.3% more than May 2025. It’s also 6.8% below the five-year May average of 9.59 million tons, and the second-lowest May in the six years USGS tabulates, ahead of only 2025.
Year to date the picture is better. Shipments through May reached 39.5 million tons, up 4.0% from the same stretch of 2025.
Type IL now dominates U.S. cement shipments
The bigger story is the mix. Of all blended cement reported this year, 23.9 million tons, or 96%, was portland-limestone cement (Type IL). Blended cement accounted for roughly 24.9 million of the 39.5 million tons shipped through May, which puts Type IL at about six of every ten tons of cement moving in the country. Texas was the top destination.
That shift has happened fast enough that specs haven’t all caught up. Plenty of project manuals still call for ASTM C150 Type I/II by default, and the supplier’s answer is increasingly a C595 Type IL submittal. If you’re a GC or a concrete sub, it’s worth checking whether your standard specs, mix submittals and set-time assumptions reflect what’s actually in the silo.
Where cement demand is weakest
Compared with its own five-year May average, the West Coast was soft. Northern California shipments ran 26.8% below, Washington 22.7% below, southern California 17.4% below and Colorado 16.8% below. Georgia was down 13.3%. A few markets went the other way: western New York was 18.4% above its average and Illinois outside Chicago was 11.0% above.
The top five consuming states (Texas, California, Florida, Ohio and Illinois) took 37.5% of May shipments. Texas, Missouri, California, Florida and Michigan led production.
Masonry cement is where the weakness shows most clearly. Shipments fell 14.1% from a year earlier to 163,000 tons, and are down 9.4% year to date at 779,000 tons. Florida, Texas, North Carolina, California and Georgia took 53.8% of it. Masonry cement follows block walls and stucco, which is to say residential and light commercial, and it’s consistent with the 27% drop in housing completions Exchange reported last week.
Clinker output is outrunning cement demand
Domestic kilns are running harder than shipments would suggest. Clinker production, excluding Puerto Rico, hit 6.67 million tons in May, up 3.2%, and 26.5 million tons year to date, up 6.2%. Imports of cement and clinker rose 4.1% in May to 2.61 million tons, but just 2.2% for the year at 10.0 million tons.
Clinker growing at 6.2% while shipments grow at 4.0% points to domestic producers taking a larger share of supply, or building inventory. The USGS data can’t tell you which. It also carries caveats: the figures are preliminary estimates, and clinker trucked in from Canada on informal entry into the Detroit and Seattle districts isn’t counted, which matters given the recent changes to Canadian import treatment.
The full tables are on the USGS cement statistics page. The number to watch in the June report is that 6.8% gap to the five-year average, and whether Type IL’s share of blended tonnage goes any higher than 96%.