UP-NS merger and CPKC precedent show only a handful of corridors and hub gateways go from two rail options to one

Summary

Short answer: There is one confirmed, completed case and one large pending case. (1) After the 2023 Canadian Pacific–Kansas City Southern merger created CPKC, the Laredo, TX gateway to Mexico went from two competing US-side connections (UP and KCS, now both CPKC) to one owner controlling the historically dominant crossing — a loss of practical choice real enough that BNSF and J.B. Hunt diverted roughly 170,000 containers/year to the smaller Eagle Pass, TX crossing (via Ferromex) specifically to keep a second option alive. (2) In the pending, not-yet-approved Union Pacific–Norfolk Southern merger (STB docket, filed Dec. 2025, rejected as incomplete Jan. 2026, refiled/amended through April 2026, still under review as of Sept. 2026), the applicants themselves acknowledge a small number of '2-to-1' and '3-to-2' shipper situations — concentrated around the St. Louis gateway (where UP and NS jointly own the Terminal Railroad Association of St. Louis, and must now divest that stake), the Houston/Gulf Coast petrochemical corridor served through the Port Terminal Railroad Association, and 'a handful of customer locations in the Midwest' where UP and NS currently compete head-to-head. Beyond that, the exact facility names are confidential in the docket; opponents (BNSF, CPKC, state AGs, CN) argue the applicants are undercounting how many shippers would actually lose a second carrier.

1. The one completed, verified case: CPKC and the Laredo, TX gateway

Before April 2023, Union Pacific and Kansas City Southern were separate, competing railroads that both interchanged with Mexican carrier Kansas City Southern de México (KCSM) at the Laredo, TX border crossing — historically the single largest US-Mexico rail gateway by volume. UP estimated that more than half of the 300,000+ cars/containers it interchanged with KCSM in 2019 moved to or from points served exclusively by KCSM.

When CP and KCS merged into CPKC, that gateway effectively became a one-railroad chokepoint for shippers who had used UP–KCS Laredo interchange to reach single-served Mexican destinations. The market response confirms the loss was real, not just theoretical: BNSF and J.B. Hunt subsequently shifted roughly 170,000 containers a year away from CPKC's Laredo route to the smaller Eagle Pass, TX crossing, connecting instead with Mexican carrier Ferromex, purely to preserve a second, non-CPKC option for that traffic.

2. The pending case: UP-NS merger, where the docket itself names the pinch points

The Union Pacific–Norfolk Southern merger (targeted close: H1 2027) is an 'end-to-end' combination with almost no route overlap nationally — which is why the applicants can truthfully say most corridors are unaffected. But three specific locations recur across the applicants' own filings and their critics' briefs as the places where a shipper's rail options genuinely shrink:

• St. Louis / the Terminal Railroad Association of St. Louis (TRRA). UP already owns 42.84% of this Class III bridge carrier across the Mississippi at St. Louis, and NS owns 14.29%. Combined, that gives the merged company a controlling-adjacent stake in a facility that other railroads rely on to reach St. Louis-area shippers — enough of a concern that UP and NS were forced to commit, in their amended April 2026 application, to divest or relinquish control of TRRA as a condition of closing.
• Houston/Gulf Coast petrochemical corridor via the Port Terminal Railroad Association (PTRA). PTRA currently provides shared switching access to Houston-area petrochemical plants and port facilities for UP, BNSF, and CPKC. Analysts flag this corridor because a combined UP-NS, controlling more of the inbound/outbound traffic, changes the bargaining dynamic for plants that today can effectively choose among carriers reaching PTRA.
• 'A handful' of Midwest joint-served customer locations that UP CEO Jim Vena has publicly acknowledged are served today by both UP and NS — the closest thing to genuine head-to-head overlap in an otherwise end-to-end merger.

On the regulatory numbers: UP and NS's own filings state they identified no corridor where they are the only two rail options serving it, but do acknowledge a small number of '3-to-2' facilities (reported as roughly four) and, in different filings, either 'five shippers' or 'three facilities' that would move from two Class I options to one. For all of these, the applicants commit to preserving access to a non-UP/NS Class I carrier via trackage rights, haulage, or similar arrangement, where legally possible. Critics — CPKC, CN, BNSF, and a bipartisan group of state attorneys general — argue this undercounts the true number of affected facilities, since the underlying shipper-level traffic data is confidential and the STB has directed UP/NS to file a fuller accounting.

3. What the public record does NOT let anyone name

Individual shipper facility names (specific plants, elevators, refineries) that would go 2-to-1 or 3-to-2 are treated as confidential/proprietary in the STB docket — competitors and the STB itself can see facility-level detail, but it is not published for the public. So the honest, checkable answer is: two verified hub-level corridors (Laredo/CPKC, already realized; St. Louis/TRRA and Houston/PTRA, flagged in the pending UP-NS case) plus an unspecified but small number of individual shipper facilities the applicants themselves concede exist and have committed to remedy — not a complete named list. Any source claiming to name the individual affected plants beyond these institutional chokepoints is going beyond what's in the public docket as of September 2026.

Separately, none of this is about facilities that were always single-served (about 95% of US grain elevators, for example, have only ever had one rail carrier) — that is a pre-existing captive-shipper condition, not something 'recent consolidation' newly created. The question asked about corridors newly reduced to one option by recent mergers, which is the narrower and more specific claim addressed above.