Passengers on most American intercity trains eventually experience a long unexplained halt on a siding. The cause is ownership: outside the Northeast, passenger trains are guests on freight railroad property.
The host railroad owns the track
When the national passenger operator was created, it took over passenger service but not the infrastructure. The freight railroads kept their track and their dispatching.
Passenger trains therefore run under agreements with each host railroad, which controls signals, maintenance standards and the movement of everything on the line.
The Northeast Corridor is the significant exception, largely owned by the passenger operator itself, which is why performance there differs sharply from the rest of the network.
Dispatching decides who waits
A dispatcher sequences trains on a shared line, and much American mainline track is single track with periodic passing sidings rather than continuous double track.
Only one train can occupy a segment at a time, so meeting an opposing train requires one of them to stop in a siding until the other clears.
The choice of which one stops is made by an employee of the host railroad, weighing its own network fluidity alongside the passenger schedule.
Freight trains have grown harder to fit
Long trains are more efficient per ton for the railroad, and mainline trains have grown considerably in length over the past decades.
Many passing sidings were built for shorter consists. A train longer than the siding cannot pull in, so the passenger train must take the siding instead.
Restarting a heavy freight from a stop is slow and consumes fuel, which creates a practical bias toward keeping it moving even when a rule points elsewhere.
Maintenance windows add to the pressure. Crews take segments out of service for hours at a time, compressing every train on the route onto whatever track remains available.
Preference exists in law but is hard to enforce
Federal law gives intercity passenger trains preference over freight except in specified circumstances, a provision written into the original arrangement.
Enforcement has been the difficulty. Determining whether a specific delay violated preference requires reconstructing dispatching decisions from railroad records.
Congress later established on-time performance metrics and a process for investigating chronic delay, which shifted some disputes into a regulatory forum.
Schedules absorb the uncertainty
Timetables include padding so a train that lost time on a host railroad can still arrive as advertised. Padding is why trains sometimes wait at intermediate stations.
The padding is calculated from historical performance, which means chronic interference gets built into the published schedule rather than eliminated.
Capacity projects that add sidings or double-track segments address the root cause, but they are expensive and require agreement with the railroad that owns the land.