We recently noted we’re already getting tired of writing about Union Pacific+Norfolk Southern; so, invariably, here’s a piece on UP+NS …
How Powerful is Single-line Service?
The ultimate success of the UP+NS merger application before the Surface Transportation Board (STB) will depend on satisfying the public interest test in the merger rules, and a key piece of that is making a strong case for taking trucks off the highway. It’s an easy argument to make for a merger that eliminates interchanges and transitions interline services to single-line services.
UP and NS have ball-parked the benefits at around $1.75 billion in EBITDA synergies, which grosses up to $3.48 billion in revenues using the LTM (Last Twelve Months) consolidated EBITDA margin of 50.2%,. Dividing this by the consolidated revenue per load of $2,251 results in about 1.55 million additional loads (10% above current). In terms of time frame, these market share gains are expected within three years of the deal closing (in early 2027), so by the summer of 2030 UPNS would expect to be hauling about 1.55 million more loads than they otherwise would have done as independent entities.
(Editor’s Note: LTM should really be PTM, Past Twelve Months. Think about it: If it really was the Last Twelve Months, it would mean the company is out of business. Right? — William C. Vantuono, Nitpicker-in-Chief)
If the merger goes through, we’ll obviously be rooting for them to succeed in this regard, but it’s still an aggressive target in our view, and whether they achieve it or not will depend on the power of single-line service.
Are the market share benefits of single-line service more modest or dramatic?
The benefits are also in two pieces: 1) A one-time gain in share after the interchanges are removed. 2) A superior ongoing growth profile of single-line vs. the prior interchange services. There’s no way to test for part 1, but we can look at historical data to get a sense for whether single-line traffic does indeed have better growth dynamics than interchange traffic over time.
Carload Traffic
We’ll start with Union Pacific’s carload business, and the chart shows total freight tonnage hauled between 2006 and 2024. We’re using tonnage rather than loads to remove distortions from different equipment of varying capacities. We’ve taken total tonnage on the railroad and then removed intermodal, coal and grain to get a sense for what UP’s core carload traffic has been doing. The blue columns show tonnage for single-line service, while the yellow columns show tonnage on interchange service, either originated on UP and delivered to another carrier, or received from another carrier and delivered to destination by UP.
In terms of growth during the 18-year period, single-line tonnage grew from 144 to 173 million tons, which is a CAGR of 1%, vs. just 0.3% for interline service. While 1% is unremarkable, the superior growth profile vs. interchange is what you’d expect.
If we run the same analysis for Norfolk Southern, the picture isn’t so tidy, with single-line underperforming interchange over the period. Single-line tonnage fell from 115 to 86 million tons, a –1.6% CAGR, due in part to three meltdowns in 2014, 2018 and 2022, and the switch to PSR in 2018, with its associated lane rationalization and hump yard closures (UP also adopted PSR in 2018 and had one less meltdown). In contrast, interline tonnage was 2.1 million tons higher over the same period.
Intermodal
Shifting to intermodal, and the plot thickens further. The data we’re using comes from the Freight Commodity Statistics submissions by the railroads to the STB, which in Union Pacific’s case paints a more dire picture than the intermodal volume data reported weekly and accompanying quarterly financial results, given there’s a difference in how intermodal is defined between the two sets of data.
Total intermodal tonnage starts at 27 million in 2006, takes a hit during the financial crisis and fails to bounce back, and then tails off again following the PSR adoption in 2018 and associated lane rationalization. There’s a modest bounce back to 17 million tons in 2024, but we’re still looking at a dismal tonnage CAGR of –2.5% over the period; due in part to average weight per load falling from 12.1 to 10.6 tons. Interline fared even worse, shedding an additional five million tons at a CAGR of –3.3%.
Where did those 10 million tons of single-line intermodal freight go? It wasn’t to BNSF, which gained 3.6 million tons over the same period. Shifting trade patterns may have removed some, but it’s hard not to conclude that millions of tons of intermodal freight that used to be on Union Pacific has been put back on the highways—in other words, the opposite of what they’re now pledging to do. Even if we (generously) use the intermodal volume load counts reported weekly, the annual CAGR between 2006 and 2024 is just 0.1%. Union Pacific is the one Class I railroad that hasn’t been able to grow intermodal (due in part to prioritizing pricing and margins, which isn’t irrational).
Did Norfolk Southern fare any better? Yes it did, with single-line service generating an additional 6 million tons of intermodal freight at a CAGR of 1.4% since 2006, vs. flat interline growth. If this merger ultimately goes through, it sounds like UP management needs to put the NS intermodal guys in charge.
Now, let’s drill down into a couple of smaller, but key, traffic types…
Automotive
Automotive is another service-sensitive commodity and uniquely interline-heavy, given production is spread across all three USMCA countries (at least for now), and you can see that in Union Pacific’s chart, where interline tonnage is more than double single-line. It’s also a weak picture that mimics the intermodal chart, with tonnage fading after the 2018 PSR adoption and into the pandemic that results in negative tonnage growth over the period. Single-line tonnage shrank at a CAGR of –2.3% between 2006 and 2024, while interline tonnage softened at –1.2%. Again, shifting trade patterns may well be to blame, but in terms of the raw numbers, single-line underperformed interline.
Norfolk Southern did worse, due largely to a high starting point in 2006 and a huge hit during the recession. Again, we have dismal tonnage growth of –6.4% for single-line and –2.4% for interline tonnage, with interline tonnage now running higher than single-line.
Chemicals
Let’s look at chemicals, because it has a different profile again. It’s a bit of a natural monopoly for the railroads because it’s typically long-distance bulk, and shippers prefer to put it on the railroads than highways because they’re safer. Union Pacific has a good track record with chemicals, with single-line tonnage growing from 16.5 to almost 30 million tons over the past 18 years, at a CAGR of 3.3%. Interline tonnage gained 3 million tons at a lower CAGR of 0.6%. We’re back to normalcy, with single-line outperforming. The Norfolk Southern chemicals chart is a bit of a nothing-burger. Single-line tonnage has barely changed: 0.1% CAGR; while interline wasn’t much better: 0.2% CAGR.
Coal and Grain
We’ll skip charts here because these aren’t service-sensitive commodities, and this section of the report is getting too long, but for the record: Between 2006-2024, Union Pacific single-line coal tonnage fell from 164 to 57 million (–5.7% CAGR) and interline tonnage from 98 to 19 million (–8.7% CAGR).
Norfolk Southern single-line coal tonnage fell from 138 to 60 million (-4.5% CAGR) and interline tonnage from 46 to 15 million (-6.0% CAGR). Union Pacific single-line grain tonnage fell from 27 to 23 million (-0.9% CAGR) and interline tonnage grew from 13 to 17 million (+1.7% CAGR). Norfolk Southern single-line grain tonnage fell from 17.5 to 15 million (-0.9% CAGR) and interline tonnage grew from 3.7 to 5.1 million (+1.8% CAGR).
Some Conclusions
Each of these individual commodity analyses is flawed in the sense that shifting trade patterns can muddy the waters and distort the outcomes, so you need to do a few of them across different traffic types to get a sense for whether single-line service does indeed produce a better growth outcome over time compared to interline service. Readers can look at the charts and arrive at their own conclusions, but it’s not a slam dunk, in our view, with interline service outperforming single-line in half of the charts we show (ignoring the big natural monopolies of coal and grain).
Logic suggests that single-line service is materially better (how can it not be?), and that first Union Pacific carload chart should be most representative of the advantage. However, our takeaway from this exercise it that single-line service is not an automatic win button. You’re still asking customers for a larger share of wallet vs. trucking vendors they like and trust, and there will be a lot of reluctance there, and in some cases (around the watershed, for example) you’re reaching out to customers that either haven’t used rail in the past or haven’t used rail on certain lanes. That’s an even tougher sell.
