The Clock Nobody Agreed On: U.S. Detention, Dwell Time and Who Actually Pays in 2026

A broker books a dry van into a grocery DC outside Columbus with a 6:00 a.m. appointment. The carrier's rate confirmation gives two hours of free time and $75 an hour after that. The driver checks in at 5:40, gets a door at 9:15 and is released at 11:50. The carrier invoices four hours of detention, the broker pays it to keep the relationship, and then bills the shipper. The shipper declines. Its contract with the broker starts the clock at the scheduled appointment, not at arrival, requires notice to the shipper's transportation desk before the end of free time, and accepts only facility in and out times signed on the bill of lading. The broker has a driver's text message, an ELD log showing on duty, not driving, and a BOL with no out time. It eats the full amount.

The scenario is illustrative, but every piece of it happens daily. The money was lost when two contracts defined the clock differently and nobody's workflow noticed. With dwell rising in 2025 and no federal detention rule coming, the 2026 question is which detention you will bill, absorb or pass through, and whether your evidence matches the contract you are billing against.

Dwell moved the wrong way in 2025

An industry research institute's operational costs update, released in July 2026 and based on a February to May 2026 survey of for-hire fleets running more than 182,000 tractors, found average dwell time rose in 2025 to 1 hour and 49 minutes per stop. That is 11 minutes short of the industry-standard two-hour threshold. Refrigerated and flatbed fleets already average more than two hours per stop because live loading is common, and tanker dwell jumped 24% year over year.

The same report put the cost in operating terms. The average operating cost was $106.69 per truck hour in 2025 including fuel, and $84.68 excluding it. Adjusted for inflation and longer waits, it estimated truckload carriers lost $5,392 per truck to detention in 2025 and drivers lost $962 each, even after detention pay. It also found that every additional 30 minutes of average dwell per stop was associated with a 5.4 percentage point increase in driver turnover. Compare those hourly costs to the detention rates on your recent rate confirmations. Even collected detention can be priced below what the hour costs.

One definitional detail matters. That survey measures dwell from scheduled arrival (or actual arrival if late) to departure, a convention some shipper contracts also use. Many carrier rate confirmations start at check-in. If your two contracts differ, every early arrival is a dispute you have already lost.

Washington is studying, not ruling

There is still no federal requirement that shippers or receivers pay detention, and nothing in 2026 changed that.

FMCSA's ELD-based study, "Impact of Driver Detention Time on Safety and Operations," is collecting 12 months of ELD, GPS, telematics, pickup and delivery, and safety data from up to 80 carriers and 2,500 drivers. OMB approved the collection on May 22, 2024, with an expiration of May 31, 2027, and its terms require peer review and OMB review of the final report before publication. FMCSA says earlier studies could not separate normal loading time from true detention. No findings had been published as of early November 2026.

In Congress, a Congressional Research Service report updated September 11, 2026 framed the question as what federal role, if any, there should be in reducing excessive detention. It highlighted a researcher's suggestion that facilities be publicly rated the way cities grade restaurants. Surface transportation authorization now runs to December 11, 2026 after an extension enacted September 2, 2026. The House reauthorization bill, H.R. 8870, was introduced and ordered reported in May 2026, and its introduced text contains no detention provision.

FMCSA's split duty period pilot, scheduled for 2027, may change how dock time hits the 14-hour window. It does not change who pays. Detention remains a contract and evidence problem, so plan 2027 contracts on that basis.

Evidence: what each timestamp actually proves

Most disputes are not about whether the truck waited. They are about whether your proof meets the standard in the contract being billed.

ELD records. Under 49 CFR 395.2, on-duty time includes attending a vehicle being loaded or unloaded and remaining in readiness to operate, so dock time belongs in on duty, not driving. But the ELD duty status is a driver input, and the federal ELD specification only requires position accuracy on the order of plus or minus one mile, with the device clock allowed to drift up to 10 minutes from UTC. An ELD log proves the driver was on duty somewhere near the facility. It does not prove gate arrival, and a shipper with a truck stop across the road knows it.

Geofence and telematics events. A tight geofence around the facility gives arrival and departure events with far better precision, generated without driver input. It is usually the strongest carrier-side evidence, if the geofence is drawn on the actual gate and the contract does not name facility records as the sole source.

Driver-reported times. Texts and app check-ins establish notice. Alone, they rarely survive a shipper audit.

Facility records. Gate logs, dock scheduling timestamps and signed in and out times on the BOL are what shipper contracts commonly treat as controlling. A BOL with no out time is a common way good claims die.

The working rule: capture at least two independent timestamps for arrival and departure, and know which one each customer contract treats as controlling before the truck arrives.

Rate confirmation terms that decide the outcome

Detention language is usually one sentence. It should be five decisions.

• Clock start. Scheduled appointment, actual check-in, or the later of the two. Early arrivals should never start the clock unless the facility accepts the early load.

• Free time. Two hours is common, but live-load reefer and multi-stop freight need different defaults.

• Rate and increments. Per hour, per quarter hour, or rounded down.

• Caps and conversion to layover. A daily cap, and the point where detention becomes layover, written so nobody invents it after the fact.

• Notice and submission windows. When the carrier must notify the broker, when the broker must notify the shipper, and how many days after delivery the claim must be submitted with evidence.

The last item is where broker margin disappears. If your shipper requires notice before free time expires and your carrier rate confirmation only requires an invoice within 30 days, you will pay claims you can never pass through. Make the carrier's window shorter than the shipper's, with slack for your team to act.

Federal law adds an outer limit for motor carriers. Under 49 U.S.C. 13710, a carrier that wants to collect charges beyond those originally billed must issue the additional bill within 180 days of receipt of the original bill, and a shipper contesting charges must do so within 180 days of receiving the bill. Contract carriage can waive some of these rights in writing, so read your master agreements.

Asset, broker and hybrid: different defaults

Asset carriers carry the cost directly: the truck hour, the driver's 14-hour window and the next load that misses its pickup. Because detention counts against on-duty time, three hours at a receiver can turn a same-day reload into a next-day reload. That lost load often costs more than the detention bill. Paying drivers detention automatically on a clear threshold, regardless of collection, is increasingly defensible given the measured turnover link.

Brokers sit in the middle with two contracts and two clocks. The core decision is a pass-through policy: pay carrier detention only when it is collectible upstream, pay all documented detention and absorb the gap as a cost of capacity, or something in between by customer tier. Each is defensible. Deciding load by load, by whoever answers the phone, is not. Brokers also own the evidence gap: the proof sits in carrier systems and the controlling records sit at the shipper's dock.

Hybrids get the hardest version. The same customer may see asset-hauled loads and brokered loads into the same DC, often with different detention terms. If the asset side absorbs detention while the brokerage side bills it, the customer notices and the precedent weakens both. Hybrids need one customer-level detention policy with explicit exceptions for each mode.

Facility scorecards: price the dock, not the lane

Detention is a facility attribute more than a lane attribute. A useful scorecard tracks, by facility and appointment type:

• average and 90th percentile dwell, measured from the same clock start your contracts use

• share of stops exceeding free time

• detention billed, collected and paid out, with collection rate and average days to pay

• disputes lost and the reason (no notice, missing out time, clock start mismatch)

• impact on the next load: missed pickups and resets attributed to that stop

With that data, decisions get specific: a dwell premium on quotes into the worst facilities, drop trailers where live loads run late, numbers brought to the shipper's quarterly review, and declined freight where collection rates make the load unprofitable at any realistic rate.

Encoding your detention policy

None of this is complicated, but every operation is specific. A reefer fleet with dedicated grocery accounts, a spot-heavy dry van brokerage and a hybrid running both need different free time defaults, notice rules, pass-through policies and approval thresholds.

Shops that collect reliably encode those decisions into their workflow rather than a policy document. Customer and facility records that carry their own clock start rule, free time, rate, cap and notice window. Load statuses that open a detention event automatically when a geofence or check-in shows free time is about to expire, and alert the right desk to notify the shipper before it does. Evidence requirements a claim must meet before it can move to billing. Permissions that control who can approve paying carrier detention that is not collectible upstream, and who can waive it for a key account. Billing rules that link the carrier payable to the customer receivable so margin on every detention event is visible.

Detention will stay a contract problem until federal data or legislation changes it, and neither looks likely before 2027. The operators who come out ahead will be the ones whose systems know, before the truck checks in, which clock this customer agreed to.

FAQ

Is there a federal rule requiring shippers to pay truck detention in 2026?

No, there is no federal requirement that shippers or receivers pay detention, so terms are set by contracts and rate confirmations. FMCSA is still running an ELD-based detention study.

What is the average dwell time per stop for U.S. trucks?

An industry research institute's July 2026 operational costs update found average dwell rose to 1 hour and 49 minutes per stop in 2025. Refrigerated and flatbed fleets averaged more than two hours.

Are ELD records enough to prove detention?

Usually not on their own, because ELD duty status is entered by the driver and federal ELD specifications only require position accuracy of roughly plus or minus one mile. Pair ELD data with geofence arrival and departure events and facility in and out times on the BOL.

How long does a carrier have to bill detention after delivery?

Your contract usually sets the practical deadline, often a set number of days after delivery with evidence attached. For motor carriers, 49 U.S.C. 13710 also requires bills for charges beyond the original invoice within 180 days of the shipper's receipt of that invoice, unless validly waived in a written contract.

Should brokers pay carrier detention they cannot collect from the shipper?

That is a policy choice that should be made by customer tier, not load by load. Some brokers absorb documented detention as a cost of capacity, others pay only what is collectible, but either way the rules, notice windows and approval permissions should be explicit.