Rail Freight by the Numbers: 2026 Industry Statistics

Rail freight does not make headlines the way trucking does. Most shippers know rail is cheaper than truck on long hauls, more fuel efficient, and better suited to bulk. What they often do not know is the actual scale of the network, the productivity gains over the last four decades, the safety record relative to the rest of US transportation, and the volume of capital the carriers reinvest every year just to keep the trains moving.

This is a working reference to the rail freight statistics that actually matter to shippers in 2026. The numbers come from the Association of American Railroads (AAR), the Bureau of Transportation Statistics (BTS), the Surface Transportation Board (STB), the Energy Information Administration (EIA), and the Federal Railroad Administration (FRA). Where 2026 data is still rolling in, we cite the most recent fully reported year and flag it.

The Network at a Glance

The US freight rail network is one of the largest privately funded transportation systems in the world. It is owned, built, and maintained almost entirely by the railroads themselves — not by federal or state governments — and the scale is easy to underestimate until the numbers are on the page.

• Roughly 140,000 route miles of freight rail track across the United States, according to AAR and BTS. The Class I carriers own and operate about 92,000 of those miles directly.
• 7 Class I railroads operate in North America: Union Pacific (UP), BNSF, Norfolk Southern (NS), CSX, Canadian National (CN), CPKC (formed by the 2023 Canadian Pacific and Kansas City Southern merger), and the Ferromex/KCSM operations on the Mexican side. The number dropped from 40 in 1980 to today's seven through four decades of consolidation.
• About 600 short line and regional railroads handle the first and last miles of much of the network, operating roughly 50,000 miles of track and originating or terminating around one in four carloads.
• Approximately 115,000 employees at the US Class I railroads — a fraction of the workforce the same industry employed in the 1980s, reflecting decades of productivity gains.
• ~1.7 million freight cars in the North American fleet, with a mix of railroad-owned, leased, and shipper-owned equipment.
• Over 25,000 locomotives in service across the Class I fleet, the vast majority diesel-electric.

The headline takeaway: a small number of carriers (seven Class Is plus the short line ecosystem) move an enormous share of US freight on a network they paid for themselves. That structural reality drives the consolidation, the pricing power, the capital intensity, and most of the operational dynamics shippers experience in 2026.

Carloads, Containers, and Volume

AAR publishes carload and multimodal traffic data every week — it is the most current data set in US transportation. Looking at the most recent fully reported year and recent weekly trends, the picture in 2026 looks like this:

• Roughly 11–12 million originated carloads per year on US Class I railroads. A carload is one freight car of merchandise — bulk grain, chemicals, coal, steel, lumber, or any other commodity moving in a railcar.
• Roughly 13–14 million container and trailer units per year moving on US Class I railroads as multimodal traffic. Container-on-flatcar volume has grown into the largest single category of rail freight measured in units.
• Combined annual volume of roughly 24–26 million carloads plus container units across the Class I network.
• More than 1.5 trillion ton-miles per year moved by US Class I railroads. A ton-mile is one ton of freight moved one mile — the industry-standard measure for output across modes.

Volume in 2024 and 2025 ran roughly flat to modestly positive year-over-year, with container traffic recovering from the 2022–2023 destocking cycle and bulk carload categories trending mixed. Coal continued its long secular decline; chemicals, grain, and motor vehicles held up; container volumes posted positive growth in most weeks of 2025. The first months of 2026 have followed the same pattern.

Modal Share: Where Rail Wins

Modal share is the share of total US freight movement carried by each mode — rail, truck, water, pipeline, and air. The number depends heavily on whether you measure by tons, ton-miles, revenue, or unit count. The ton-mile figure is the cleanest for comparing freight productivity across modes.

• Approximately 28 percent of total US freight ton-miles move by rail, according to BTS Freight Facts & Figures. Trucks carry the largest share of total ton-miles, followed by rail and then pipeline.
• Approximately 40 percent of US long-distance freight (over 750 miles) moves by rail. The longer the haul, the larger rail's share — because rail's cost advantage scales with distance.
• Roughly one-third of US exports by tonnage touch the rail network somewhere in the move, particularly for grain, coal, chemicals, and containerized goods reaching coastal ports.
• About 6 percent of US freight value moves by rail — far lower than the ton-mile share because rail dominates lower-value-per-pound commodities like bulk grain, aggregates, and coal.

The lesson behind the numbers: rail moves the heavy stuff, far. For a shipper deciding between rail and truck, the practical break-even is usually somewhere in the 500–1,000 mile range with reasonable volume, with a longer haul and a denser, lower-value commodity pushing the math harder toward rail. The full mechanics are covered in our rail vs. truck cost comparison.

Top Commodities by Carload

AAR breaks weekly carload data into commodity categories. The recent ranking — measured by carloads, not value — looks roughly like this:

• Containers and trailers (multimodal traffic): the largest category by unit count, growing year over year and now structurally larger than coal.
• Chemicals: consistently one of the top three bulk carload categories. Plastic resins, industrial gases, fertilizer inputs, and petrochemicals.
• Coal: historically the single largest category, now roughly 14–16 percent of carloads versus around 40 percent in 2008. The decline is structural and tracks utility-sector coal retirements.
• Grain: highly seasonal, with peaks tied to harvest. Wheat, corn, soybeans, and grain products. Our shipping grain by rail guide covers the operational details.
• Motor vehicles and parts: finished autos, auto parts, and the trains that move them. Sensitive to OEM production schedules.
• Stone, sand, and gravel (frac sand and aggregates): tied to construction activity and energy drilling cycles. See our aggregate rail shipping guide.
• Forest products: lumber, pulpwood, paper, and wood pulp. Tied to housing starts.
• Metallic ores and primary metal products: iron ore, finished steel, coil, and scrap. Our steel and metals rail shipping guide walks through the equipment and logistics.
• Petroleum and petroleum products: crude, refined products, and biofuels.
• Food and farm products: processed foods, ethanol, distillers' grains, and agricultural products beyond raw grain.

The commodity mix has shifted meaningfully over the last decade. Coal's secular decline has been offset by container traffic growth, frac sand cycles, and chemicals stability. Shippers in chemicals, plastics, grain, and aggregates have the deepest, most experienced carrier ecosystem available — rail has been moving their products at scale for decades.

Fuel Efficiency and Productivity

Fuel is the single largest variable cost on a freight rail move outside of crew. Every productivity statistic in the rail industry eventually traces back to ton-miles per gallon, and the numbers have improved dramatically over time.

• More than 470 ton-miles per gallon on US Class I railroads, on average. That metric — how far one ton of freight travels on one gallon of diesel — is the AAR's standard productivity benchmark.
• Roughly 3–4x more fuel efficient than long-haul trucking on equivalent ton-miles. The exact multiple varies by commodity and lane, but the order of magnitude is consistent across BTS, EPA, and AAR analysis.
• Fuel efficiency has improved by roughly 100 percent since 1980, reflecting longer trains, more efficient locomotives, distributed power, energy management software, and improved aerodynamics on the cars.
• One freight train can replace several hundred trucks on the highway, depending on train length and commodity. The single-train-replaces-X-trucks ratio is one of the most-cited rail statistics — and it is roughly accurate for a long unit train of bulk freight.

Fuel surcharges are the mechanism by which the diesel cost gets shared with the shipper. They are calculated off published indexes — most commonly the Department of Energy's On-Highway Diesel Fuel Price (HDF) — and move month to month with the diesel market. Our rail fuel surcharge calculation guide covers the mechanics. For shippers running cost models, the diesel-driven variability in landed cost is one of the few line items where the math has to be done carefully, not assumed.

Safety and Hazmat Performance

Rail safety statistics get less attention than they deserve. The headline number a shipper should know:

• More than 99.9 percent of hazardous materials shipments arrive without a release caused by a train accident, per AAR safety reporting.
• Train accident rate has fallen roughly 30 percent since 2000, with derailment rates and grade crossing collision rates both trending down over the long horizon.
• Lowest employee injury rate among the major modes of US transportation industry tracked by the Bureau of Labor Statistics.
• Substantially safer per ton-mile than truck for hazmat — by a factor commonly cited at 4–5x in federal analyses, varying by commodity class.

The headline safety incidents that do make the news — major derailments at high-profile locations — are real and consequential. But the underlying long-term trend, even after recent high-profile events, is steady improvement on a per-ton-mile basis. For shippers in chemicals, fuels, and hazmat-rated commodities, rail's safety performance is one of the strongest arguments for the mode and a regular topic in conversations with insurance and risk teams.

Capital Investment and Network Spending

The US freight rail network is one of the most capital-intensive industries in the country. Unlike the highway system, the locks and dams on inland waterways, or the airports, the rail network is privately built and maintained — every dollar of track, ballast, ties, bridges, signals, locomotives, and yards comes from the carriers' own balance sheets.

• Roughly $25 billion per year reinvested by the US freight rail industry in capital expenditures and maintenance, according to AAR figures.
• More than $760 billion invested cumulatively since 1980, the year Staggers partially deregulated the industry. The deregulated industry was able to fund the investment because it could finally earn an adequate return on capital.
• Around 19 percent of revenue reinvested in the network at a typical Class I, dramatically higher than most other US industries.
• Track-mile maintenance: the Class I carriers replace millions of ties and tens of thousands of miles of rail each year as part of ordinary maintenance. The work happens largely out of public view.

The capital intensity matters to shippers because it sets the floor on what the industry can charge. A network that requires this much reinvestment has to earn a return on it — which is why rate pressure has been steady since the early 2000s and is unlikely to reverse. Shippers building long-horizon rail programs should model rate trajectory accordingly. Our CFO-facing business case for rail covers how finance teams typically frame the math.

Pricing Trends Since Staggers

Rail pricing has two stories running in parallel. The long-run real-terms story is one of dramatic productivity-driven decline. The recent nominal story is one of inflation-pace rate increases.

• Real rail rates per ton-mile have fallen more than 40 percent since 1980, according to STB and AAR data, after adjusting for inflation. The decline was steepest in the 1980s and 1990s and has plateaued in the last decade.
• Nominal rates have risen with inflation in the 2020s, particularly during the 2021–2023 diesel and labor cost cycle. The 2024–2026 period has seen rate increases moderate as inflation has cooled.
• Fuel surcharges have moved with diesel across the same period. The HDF-indexed fuel surcharge can swing a per-car total cost by 10–25 percent depending on diesel direction, which is why long-duration contracts should be modeled with fuel scenarios, not point estimates.
• Contract rates differ materially from public tariffs. The vast majority of rail freight moves under negotiated contracts, not posted tariffs. The published "rate" most shippers see when they ask is an indicative estimate that becomes a binding number only when a contract is executed.

For a working framework on how rates are built, what is negotiable, and where the leverage points are, see our how rail freight rates work piece. For a shipper-side overview of how to actually request an estimate that comes back useful, our how to get a rail shipping quote guide is the right starting point.

PSR and Operating Metrics

Precision Scheduled Railroading (PSR) reshaped Class I operations across the 2017–2023 period and continues to define the operating model in 2026. The statistics tell the story of what changed.

• Average train length has grown from roughly 5,400 feet in 2009 to over 7,000 feet in recent years on multiple Class Is. Longer trains move more freight with the same crew and locomotive cost.
• Locomotive productivity — measured in gross ton-miles per available horsepower — has improved by double-digit percentages over the PSR era.
• Operating ratios at the major Class Is moved from the high 60s and 70s a decade ago into the 55–65 range, with PSR adopters compressing fastest. (Operating ratio is operating expenses divided by revenue — lower is better. A 60 OR means the carrier earns 40 cents of operating income on every revenue dollar.)
• Employment headcount at US Class Is fell from roughly 165,000 in 2014 to around 115,000 by recent years, reflecting the productivity reset.
• System velocity — average miles per hour for trains in the network — varies by carrier and quarter, but generally runs in the low to mid 20s mph for manifest service. Velocity is volatile and is a leading indicator of service problems when it dips.
• Terminal dwell time — how long a car sits in a yard between trains — typically runs in the 20–30 hour range and is the other half of the service quality picture. Lower is better.

PSR is not universally loved by shippers — the operating discipline has cut into service flexibility in some lanes, and the staffing model has been tested during volume surges. But the productivity gains are real, and the underlying economics of the rail industry have been reshaped by them. For a deeper read on the shipper-side implications, our short line vs. Class I piece touches on how the carrier landscape has shifted.

What the Numbers Mean for Shippers

A statistics post is only useful if a shipper can pull operational decisions out of it. The numbers in 2026 carry a few practical implications.

Rail's cost advantage is structural, not cyclical. The 40-plus-percent real rate decline since Staggers and the 100 percent fuel efficiency gain are not going to reverse. For shippers moving high-volume, low-value freight over long distances, the math on rail keeps working. Trucks remain faster on short and middle hauls and unbeatable for short-fuse, time-critical loads, but the long-haul economics of rail are not going to weaken.

The carrier count is small and concentrated. Seven Class Is and roughly 600 short lines is a much smaller carrier set than the trucking industry. Shippers should treat the carrier relationship as a long-term operational partnership, not a transactional one — there is not an unlimited bench to switch to. The corollary is that a strong logistics partner is one with documented relationships and lane experience across the Class I and short line ecosystem, which is the framework our guide to choosing a rail freight logistics partner walks through.

Volume mix is shifting. Coal is in secular decline. Container traffic, chemicals, motor vehicles, and grain are stable to growing. A shipper in a growth category has a healthy, attentive carrier ecosystem; a shipper adjacent to declining categories may find that the carrier's lane and equipment priorities are tilting elsewhere. Worth knowing when negotiating contracts and equipment commitments.

Service variability is real. System velocity, dwell, and on-time performance vary quarter to quarter. Shippers building rail programs should model transit-time variability into safety stock and inventory plans rather than assuming a smooth average. Rail transit is significantly longer than truck transit, and the variance is wider — both are facts of the mode that should be reflected in operations planning, not surprises that hit the first time a car gets held at an interchange.

The data is freely available. AAR's weekly traffic report, BTS Freight Facts & Figures, STB's quarterly statistical reports, EIA's diesel index, and FRA's safety database are all published openly. A shipper building or auditing a rail program should pull from the source data periodically, not rely entirely on what providers and carriers volunteer. Our free rail logistics courses walk through the metrics that matter and how to read them in the context of an actual program.

If you would like an indicative estimate on a specific rail lane — line haul, fuel surcharge, transload, dray, and equipment broken out the way the data in this post suggests it should be — send the origin, destination, commodity, and volume. We will return a clean, line-itemized estimate and a straight conversation about whether rail fits the freight. Contact our team or visit our rail logistics services page to start.