The 2022 Service Crisis is still too fresh in the minds of rail customers, inhibiting share recapture from trucks, and this will remain the case for some time. However, the fact is that methodical progress on rail service is being made.
Here’s a view of our State of the Rails chart we show every week, in this case truncated to just show 2020 through the present. As a reminder, it simply looks at weekly average velocity in every quarter and compares that to the weekly average since 1Q10 (post-recession), and does the same for terminal dwell. We then net them against each other, so 8% better velocity and 5% worse dwell nets to +3%, for example. Net 10% or better, we color green; –10% or worse is red, and everything in the middle is yellow.
You can see the recent high- and low-water marks, with lots of green during the COVID lockdowns in 2Q20 due to the absence of volume pressure, and then the subsequent crew crunch in late 2021, causing the service crisis the following year (black rectangle). The last railroad to inflect out of that was Union Pacific exactly two years ago, and since then it has been a race for the railroads to get their houses in order, one piece at a time.
What’s interesting, and the point of all this, is that if you look on the far right, the L3M column (last three months [13 weeks], rolling), CSX and BNSF are already in the green and Norfolk Southern (9.8%), CN (8.9%) and Union Pacific (7.9%) are all tantalizingly close. We’ve left out CPKC due to the distortions from the KCS acquisition.
How are they doing it? If we unpack velocity and dwell in the chart below you can see that the primary driver has been lower dwell times at yards and terminals. Union Pacific, for example, has struggled with velocity in 2Q24 and 3Q24 due to weather and the 3Q24 international intermodal surge, but the 10% loss in speed has been more than offset (in the context of the chart, not necessarily the real world) by an 18% improvement in dwell vs. its post-recession average.
BNSF, similarly, has made a big push improving its hump yards (from March) and flat switching yards (from July) to drive dwell 14% lower, while holding velocity roughly flat vs. its recent historical average.
It’s the same story at NS and CN, with dwell improvements of 8% and 11%, respectively, while velocity is within 2% of historic norms.
CSX, in contrast, is playing defense as its metrics soften from a high base, but note the multi-month outage on its Blue Ridge subdivision in North Carolina, due to Hurricane Helene damage, is acting as a drag.
Final Thoughts
As usual, we’ll take any win we can get, and the industry-wide productivity improvements at classification yards and intermodal terminals is certainly good news and a critical piece of the service puzzle. Velocity can be, and needs to be, better, but we’ll need to wait for the international intermodal surge to run its course at UP and BNSF, in particular. Also remember that the President-elect’s tariff threats, and importers’ efforts to beat them, borrows volumes from the future, so at some point in 2025 there will likely be a lull whereby all the rail intermodal networks can catch their breath and accelerate.
Finally, what matters most isn’t speed or dwell, it’s on-time performance, and we only have manifest on-time numbers that are released as part of the STB reciprocal switching reporting. The most recent numbers, for the week ending Nov. 22, aren’t great: UP 62%, BNSF 53%, CSX 41%, NS 61%, CN-U.S. 58%, CPKC-U.S. 53%, and it’s a stark reminder that, even once the fundamental operational building blocks are in place, the railroads still have a lot of work to do to narrow the gap to truck-like service.
