Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- Recalculate your US inland budget on a USD 2.19 per mile dry-van assumption for the remainder of 2026, and stop assuming a seasonal softening in October and November.
- By 30 September, model converting your longest lanes above 800 miles to intermodal and price it at a 15% to 20% discount to truck before assuming rail capacity is available.
- Review every index-linked inland clause now, since a market 21% above the seasonal average automatically moves those clauses against you at the next reset.
- Separate drayage from long-haul in your contracts so a tight truck market cannot reprice your short port runs on the same trigger.
- Track the weekly load-to-truck ratio and set a written trigger: if it falls for four consecutive weeks, reopen long-haul contracts early rather than waiting for the annual renewal.
DAT reported dry-van spot linehaul around $2.19/mile excluding fuel in its Sep 1 update, about 32.4% above a year earlier and 21% over the nine-year seasonal average. AAR data for the week ending Aug 29 show total US rail traffic up 4.1% YoY, with intermodal +5.7% to 303,191 units. With carrier exits and disciplined fleet growth removing excess capacity, shippers face a capacity-driven market rather than a demand boom.
The Analysis
DAT's September 1 update puts dry-van spot linehaul at roughly USD 2.19 a mile excluding fuel, about 32.4% above a year earlier and 21% above the nine-year seasonal average. AAR data for the week ending August 29 shows total US rail traffic up 4.1% year on year, with intermodal up 5.7% to 303,191 units. The read that matters is in the last line of the release: carrier exits and disciplined fleet growth have taken the excess capacity out, so this is a capacity-driven market, not a demand boom.
That sentence is the whole story, and it is the one most people skim past. A rate that rises because demand is strong has a natural ceiling, because high prices eventually choke off the buying. A rate that rises because trucks have left the road does not have that ceiling. There is no wave to wait out. I have watched inland markets do this before, and the tell is always the same: the rate holds flat on a quiet week instead of softening.
If your landed cost model still treats the US inland leg as a stable input that tracks the peak season, this quarter will break that model.
Three numbers anchor everything. Dry-van spot around USD 2.19 a mile excluding fuel, up about 32.4% year on year. That same figure sits 21% above the nine-year seasonal average. And rail intermodal moved 303,191 units in the week to August 29, up 5.7%, with total rail traffic up 4.1%.
Start with why the increase happened now, because the reason changes what you should do about it. A rate spike on strong demand and a rate spike on thin supply look identical on a chart and behave completely differently over the following two quarters. Capacity-driven tightening comes from the supply side: carriers have exited, and the ones still standing have been disciplined about adding tractors and drivers. Freight rates run on a cycle, and when the cycle turns, the excess capacity that normally floods back and caps the price simply is not there to flood back. That is why the same spot rate can climb 32% year on year while nobody can point to a boom. The volume supports it, but the volume is not the cause.
The 21% above the seasonal average is the number I would put in front of a board. Seasonal averages exist because rates follow a rhythm. Being a fifth above that rhythm in early September means the normal September softening either did not arrive or has already been spent. Whichever it is, you cannot plan against a seasonal dip that the market has decided not to deliver.
The rail number is not a footnote. Intermodal up 5.7% while dry-van spot is up 32% tells you shippers are already voting with their feet and pushing long-haul freight onto the rails. That is the rational response, and it is also the thing that will bite the people who do it late. Rail intermodal capacity is not elastic week to week either; when enough shippers convert at the same time, the rail option tightens too, and then you have given up the truck alternative without gaining a free one.
Who this lands on. A shipper moving full truckload on long hauls through the Midwest or the Southeast feels almost all of it, because they are competing for exactly the capacity that left. A shipper doing short regional runs is cushioned, because the capacity that exited mostly served long-haul lanes and the local market never tightened the same way. A shipper who already runs a meaningful share of intermodal has a hedge and may not need to change much, though they should expect their rail rates to follow. And the exporter using inland drayage to reach a port is exposed from a direction people forget: drayage competes for the same drivers, so a tight truck market pushes up the short haul to the terminal even when the long haul is not yours.
On timing, spot transmits to your next load, effectively immediately. Contracted truckload rates reprice on an annual or semi-annual cycle, so the shipper under contract is insulated for now and will get a shock at renewal if the tightness persists. Inventory is the buffer that hides it for a month or two, and then the cost surfaces all at once at the quarterly review.
The point the coverage tends to skip is that capacity-driven markets do not correct on the calendar. Demand-driven tightness ends when buyers stop buying. Capacity-driven tightness ends only when carriers find the money to re-enter, and re-entry takes quarters, not weeks, because it needs ordering, hiring and confidence that the rate will hold. That is why I would treat this as a floor rather than a spike. The related point nobody says: the seasonal average is what most contracts are indexed against, so a market sitting 21% above it will make every index-linked clause this autumn resolve in the carrier's favour, quietly, without anyone negotiating anything.
Let me quantify with stated assumptions. Assume 300 dry-van loads a month on an average haul of 600 miles. At USD 2.19 a mile, that is USD 1,314 a load, or about USD 394,000 a month. Assume the year-ago rate was USD 1.65 a mile, or about USD 992 a load, so the year-on-year gap is roughly USD 322 a load and about USD 96,500 a month. Over twelve months that is close to USD 1.16 million of extra inland spend. The 600-mile haul and the 1.65 year-ago rate are my assumptions; the 2.19 spot and the 32.4% increase are the reported ones. If you could convert 40% of that volume to intermodal at a 20% discount, you would recover roughly USD 9,000 a month without changing anything else.
What would flip this. If industrial demand weakens enough that carriers start competing for whatever freight exists, even a thin fleet will undercut itself, and the rate comes down fast from a high base. Watch the load-to-truck ratio rather than the headline rate. It turns before the rate does.
↑ Back to the key points