Gauging the growth potential of coast-to-coast mergers — analysis by Larry Gross

After many years of being regarded as regulatory kryptonite, the concept of a coast-to-coast rail network is back on the table in a big way. A key component of the merger pitch is growth, specifically conversion of freight from highway back to rail. Is it realistic to expect a merger to help claw back freight off the highway and into carload or intermodal?

First, some historical context. Figure 1 compares long-term carload growth against a couple of benchmarks: U.S. real GDP and U.S. long-haul truckloads. The use of the real GDP figures knocks out the effects of inflation and gives us a true measure of the increase in economic activity over this time frame. The truck numbers are dry van and refrigerated truck moves of 500 or miles in length, as provided by Noel Perry of Transport Futures, a noted transportation economist and trucking expert. The carload figures are from the Association of American Railroads and cover carloads originated by the U.S. Big Four (BNSF Railway, Union Pacific Railroad, Norfolk Southern Railway and CSX) excluding coal (since the secular decline of that crucial commodity is beyond the industry’s control) and intermodal (which is treated separately later).

In each case, the performance in the year 2016 is set at an index of 100 and the chart records the four-quarter moving average right through the end of 2024. During this 18-year period, GDP grew just under 42%, a compound annual growth rate (CAGR) of about 2% per year. Long-haul truckload growth trailed GDP, growing just 29%, a CAGR of 1.4% per year. Possible reasons for the growth deficit versus GDP include: the increasing importance of services in our economy, which generally means less freight per percent of GDP growth; a general trend toward shorter hauls; and better utilization of available truck capacity through optimized packaging and better loading practices.

But even though truckload growth trailed GDP, it still out-performed rail carload by miles. U.S. Big Four rail carloads excluding coal finished 2024 12% lower than in 2006, a loss of 0.7% per year. To be fair, however, those carloads were doing a bit more work in 2024 than they were in in 2006, to the extent that the percentage of 286K capacity cars in the fleet increased. So, perhaps we should add a couple of percentage points to the carload numbers in 2024 to account for that. That still leaves us with a big loss over 18 years.

The merger proponents say that the elimination of interchanges (aka “touches”) will enable rail to capture more highway freight. But over this 18-year period, the percentage of AAR Big Four “received” loads (i.e., loads received by these railroads in interchange) versus loads originated by these railroads actually increased from 38.5% in 2008 to 43.2% in 2024. In other words, the loss of single-line volume was worse than interchange volume. This implies that single-line service is far from a magic bullet for growth.

For reference, I have added the UP and NS specific numbers, as well. These include both originated and received loads (again, excluding coal and intermodal). UP’s numbers have held up better than average, but even here, they have seen a fractional decline over 18 years. NS volume has trailed the average, while the combined railroad volume has tracked the average closely.

In truth, the boundaries between truck and rail carload are well established. There has been some migration, but except for a brief period in 2014 and 2015, it has been one-way — from rail carload to truck. Is there a deep reservoir of current truck users in the watershed markets that would be ready to convert to rail carload if transit time was a day or two faster than the current offering? Doubtful. Perhaps they might convert at a lower price, but that would require the merged railroad to share the operational savings with the shipper. Again, doubtful.

That leaves the possible growth to intermodal, and here the story is quite a bit better. Figure 2 displays a similar analysis to that which was performed on carload, but with an intermodal spin. A third benchmark is added — that of loaded import TEUs (twenty-foot equivalent units, the universal measurement of international containerized shipments), arriving in the U.S. ports. The benchmarks are compared to the performance of U.S. intermodal (revenue moves originating and terminating within the United States), as measured by the Intermodal Association of North America’s monthly ETSO database. In addition to the total intermodal numbers, additional lines show U.S. intermodal’s performance in the international (ISO container) and domestic (domestic containers and TOFC) sectors.

Total intermodal activity grew 19% from 2006 to 2024, a CAGR of 1%. This is less than long-haul truck, but growth, nonetheless. Results were dragged down by the international intermodal sector, which shrank by 8% despite growth in inbound TEUs of 52%. The main culprit here wasn’t poor intermodal performance, but rather coastal share shift. In 2006, 44% of all import TEUs arriving in U.S. ports came in through the intermodal-friendly U.S. west coast. By 2024, as importers diversified their routing options, adding East and Gulf Coast options, U.S. West Coast share had shrunk to just 34%. Intermodal share is far lower off the East Coast and Gulf Coast ports, because the typical length of haul is shorter and much more truck friendly.

The good news is the performance of domestic intermodal — up 57% from 2006 to 2024, a CAGR of 2.5%. This far exceeded the growth in long-haul dry van and reefer truckloads. A true success story! However, there’s a catch. Most of the growth in domestic intermodal occurred during the “glory days,” prior to 2015. Since then, the volume story has become much more erratic. The problem is clearly displayed in Figure 3, which performs the same analysis as before, but with the year 2015 as the starting point.

Over the past 10 years, while GDP increased 24%, long-haul truck by 19%, and import TEUs by 37%, U.S. intermodal was up just 4.2%. The high point for intermodal was in 2018, when the implementation of electronic logging devices (ELDs) on heavy-duty trucks created a temporary shortage in truck capacity. Subsequently, PSR changes and the pandemic took their toll on intermodal. Volume briefly spiked during the early days of the post-pandemic surge but then retreated.

There are those who ascribe domestic intermodal’s current difficulties to over-capacity in the truck sector, and indeed, this has been an extended period of soft pricing and slow growth in truckload — some three years. But domestic intermodal hasn’t really grown in the past ten years. Therefore, the problem is more than the fact that we are not currently seeing the ideal “goldilocks” conditions in the trucking market that allow for intermodal market share gains.

The situation is summarized by Figure 4, which displays my estimates for the share of U.S. long-haul movement of truck-type freight by domestic intermodal. This measure hit 5.9% in 2025 Q2 — the lowest point since 2011!

Is a transcontinental railroad merger an answer to intermodal’s woes?

In 2024, a review of the AAR intermodal numbers shows that of the 13.8 million intermodal loads originated by the U.S. Big Four railroads, only 2.3 million intermodal loads were interchanged between connecting lines; 84% of intermodal loads were single line hauls. We don’t have a national intermodal network. Rather, we have a bunch of regional networks.

Figure 5 looks at the composition of U.S. truck and U.S. intermodal freight by mileage range for the year 2024. These are rough approximations. The truck numbers are from Transport Futures. The intermodal numbers are from IANA ETSO data, assuming one location per zone. For example, the length of haul for domestic loads moving between the Midwest and the Northeast is based on the highway miles between Chicago and New York City. Canadian and Mexican moves are excluded, simply because we don’t have good truck data for those markets.

While the results are necessarily imprecise, they are nevertheless informative. Truck freight follows a logical, expected curve, with more freight at shorter hauls and steadily smaller percentages moving longer distances. But the intermodal numbers don’t follow any pattern. Volume centers on the 751- to 1,000 mile single-line eastern haul (or SLEH), and the 2,001- to 2,500-mile single-line western haul (or SLWH) range. Mid-range hauls of 1,001 to 2,000 miles are underrepresented (the Donut Hole) as are super-long-Haul (SLH) moves of over 2,500 miles.

But there is a problem with this analysis, in that a substantial amount of intermodal volume is “cross-towned” between Eastern and Western railroads, i.e. moved between terminals on rubber tires. From a data perspective, this has the effect of taking one SLH interline move and cleaving it into two shorter, single-line moves, inflating the numbers for the key SLEH and SLWH lengths of haul, and depressing the SLH numbers. Total intermodal activity is also inflated because one through move is reported as two.

No hard data exists that defines how big the cross-town practice is, but after conversations with knowledgeable individuals, I have what I feel to be a good estimate. After adjusting the numbers to back out these effects, the shares of intermodal activity moving in the SLEH and SLWH ranges are lower and the share moving SLH is raised. Figure 6 translates these adjusted numbers into intermodal market share by mileage range. The current situation is on the left side of the graph.

The highest intermodal share by far is in the SLWH range at 37.6%. This is followed by the SLH range at 17.0%, and then the SLEH range, with 9.5% intermodal share. Market share is lower still in the Donut Hole.

Based on intermodal economics, this isn’t how things are supposed to work. One would expect that the longer the haul, the more low-cost rail miles are involved, and the better intermodal should look versus truck. Hence, intermodal share should increase with length of haul. Yet we have share dropping once we move beyond 2,500 miles. I think that’s because there are very few U.S. single-line intermodal moves with lengths of haul of more than 2,500 miles.

The IANA data allows us to assign share responsibility to international or domestic. This is important because the railroads have little, if any, ability to affect international volumes. These are generally from the port to inland locations and probably fall very naturally into the single-line buckets.

For the purposes of this growth potential analysis, then, I focus only on the domestic intermodal volumes. This is the market where intermodal has true freedom of action and ability to influence results. What if we wipe away the interchange barriers and assume the intermodal economics can be fully unleashed? Two scenarios are examined.

Scenario One is a conservative approach (chart center). Domestic intermodal share in the Donut Hole is raised from the current 3.0% to 5.2% to match the share already achieved in the shorter, theoretically more competitive and challenging SLEH range. This generates an additional 739,000 annual loads. Second, share in the SLH range is raised from the current 14.7% to match the 18.5% domestic share already achieved in the shorter SLWH range. This generates another 296,000 annual intermodal loads. Making these two changes collectively would bring about one million more intermodal loads to the network, an increase of 7.8% on 2024 U.S. intermodal volumes or 5.7% on the total North American intermodal activity. Nice, but enough to warrant a merger?

A more aggressive Scenario Two follows the same basic approach but share in the Donut Hole is boosted to 11.8%, midway between the shares currently achieved in the SLEH and SLWH ranges. This makes good sense in terms of how intermodal economics work. Under this scenario, in 2024, an additional 3.0 million more loads would have moved in the Donut Hole range. Added to the SLH gains, U.S. intermodal volume would increase by 3.3 million, or 24.7%, translating into a gain of 18.2% in total North American intermodal revenue moves.

But is the aggressive Scenario Two possible? It requires domestic intermodal share in the Donut Hole range to increase four-fold. A heavy lift is an understatement! The freight in this mileage range is more dispersed, with no high-volume Chicago-L.A.-type lanes. Much of the opportunity lies in the watershed markets — say, 300 miles on either side of the Mississippi River. Tapping these markets will involve new terminals, new services and more complexity. These new lanes won’t be able to support point-to-point PSR-sized trains, at least, not without some form of mixing centers along the lines of how the CSX North Baltimore facility once operated.

In my opinion, the answer for how much intermodal volume might be gained by transcontinental mergers probably lies somewhere between Scenarios One and Two. Of course, keep in mind that this analysis assumes two transcontinental railroads. It is not realistic to expect that only one railroad will single-handedly capture the gains.

Further, is a full merger the only way to accomplish this? A coast-to-coast railroad merger could achieve these benefits, but it is a brute force approach laden with risk — regulatory, operational and market risks, to name a few. It’s by no means the only way. NS’ Triple Crown subsidiary achieved success in the mid-range lengths of haul by engineering seamless services that transcended normal interchanges. Two Class Is could construct a coast-to-coast intermodal network without merging the underlying organizations. Or a third party could put one together.

Regardless of the structure, where exactly things land will depend on how much the newly merged railroads are willing to bend on their relentless drive to simplify the intermodal network and its operating characteristics. The more flexibility they show in this regard, and the more complexity they are willing to accept, the more freight will be converted from truck back to rail.

Larry Gross is a 45-year veteran of the intermodal sector. Known as “The Intermodalist,” he serves today as an intermodal analyst and is the author of the monthly analytical report “Intermodal in Depth.” In 2024, he received IANA’s Silver Kingpin award for lifetime contributions to the intermodal sector.