The regional transit lobby is obsessed with a multi-billion-dollar lie.
Every few months, another glossy report drops, claiming New York’s Penn Station is at its breaking point. The proposed cure is always the same: spend $13 billion to demolish a chunk of midtown Manhattan, dig a massive cavern south of the existing station, and add more tracks.
It is a monumentally stupid plan.
Penn Station does not need more tracks. It does not need a massive real estate land grab under 31st Street. It has 21 tracks. For comparison, London’s St. Pancras International handles its massive volume with 15 tracks. Tokyo's busiest regional rail platforms handle trains every two minutes on just a pair of tracks.
The crisis at Penn Station is not a space problem. It is an operational failure disguised as an infrastructure deficit.
If you want to run more trains through North America’s busiest transit hub, you do not need dynamite and billions of dollars in eminent domain payouts. You need to dismantle the bureaucratic fiefdoms that treat Penn Station like three separate regional border crossings rather than a single transit hub.
The Myth of the Hard Capacity Limit
The transit industrial complex loves to point at the two existing rail tunnels under the Hudson River and the four under the East River and declare that the station is physically maxed out.
They are wrong.
Let’s look at the basic physics of train movement. The capacity of a rail terminal is governed by a simple equation:
$$C = \frac{T}{D + S}$$
Where:
- $C$ is the maximum hourly capacity of a platform track.
- $T$ is time (60 minutes).
- $D$ is the dwell time (how long a train sits at the platform).
- $S$ is the minimum safe separation time between trains dictated by the signaling system (headway).
Right now, New York's transit agencies—Amtrak, the Long Island Rail Road (LIRR), and NJ Transit—operate with dwell times ($D$) that would get a transit planner in Seoul or Munich fired on their first day.
It is common for an LIRR or NJ Transit train to sit at a Penn Station platform for 15, 20, or even 30 minutes. The train arrives, slowly unloads passengers through narrow, poorly placed stairwells, waits for a cleaning crew to sweep the aisles, waits for a crew change, and then slowly reverses out the way it came.
When a track is occupied for 20 minutes, that track can only handle three trains per hour.
If you reduce that dwell time to three minutes—which is standard across high-efficiency European and Asian regional networks—that same single track can suddenly handle 12 to 15 trains per hour. Multiply that across Penn Station's 21 tracks, and the theoretical capacity of the existing footprint skywells past anything the region actually needs.
We do not need more concrete. We need faster operations.
The Through-Running Fantasy and Its Real Bottleneck
Advocates love to scream about "through-running." They argue that if we just ran NJ Transit trains straight through to Long Island, and LIRR trains straight through to New Jersey, Penn Station would magically transform into a high-speed through-station.
It is a beautiful theory on paper. In practice, it ignores physical and institutional reality.
I have spent years looking at the track profiles, traction power systems, and labor agreements of these three agencies. Through-running is not a simple switch you flip.
1. The Power System Nightmare
NJ Transit runs on overhead AC catenary power (varying between 12.5kV and 25kV). The LIRR runs on 750V DC third rail. Amtrak uses a mix of overhead catenary systems. To run trains through from New Jersey to Long Island, you need highly specialized, multi-system rolling stock.
While dual-system trains exist, they are significantly heavier, more expensive to maintain, and prone to breakdowns. NJ Transit’s multi-level fleet is built to different clearance envelopes than the LIRR’s fleet. You cannot just run an NJ Transit train down an LIRR branch line without scraping the platform edges or hitting low-hanging tunnel ceilings.
2. The Throat Bottleneck
Even if you solved the rolling stock issue, Penn Station’s track layout—the "throat" where tracks merge into the tunnels—is a tangled bowl of spaghetti designed in the 1900s.
To get a train from the Hudson River tunnels to the northernmost LIRR tracks, it has to cross diagonally across multiple other active tracks. This creates "conflicting movements." When one train moves from south to north, it blocks three other trains from moving at all.
True through-running requires a complete, flat-junction untangling of the station throat. Without rebuilding the throat, through-running just moves the bottleneck from the platform to the switches outside the tunnels.
The Three-Headed Bureaucracy
The real barrier to running more trains is not engineering. It is politics.
Penn Station is occupied by three agencies that behave like warring sovereign nations:
- Amtrak owns the station and treats the commuter railroads like annoying tenants.
- The LIRR (operated by the MTA) operates under its own distinct work rules, scheduling protocols, and union contracts.
- NJ Transit operates under a completely different set of civil service rules, dispatching priorities, and equipment standards.
Step onto the concourse. You will see three different ticketing systems, three different sets of police officers, and three different passenger information displays.
This siloing extends down to the tracks. NJ Transit crews cannot operate LIRR trains. LIRR dispatchers do not coordinate seamlessly with Amtrak dispatchers. If an NJ Transit train experiences a mechanical delay on Track 5, Amtrak cannot easily route an LIRR train onto that track because of rigid, archaic track allocation agreements that date back decades.
This is institutional cowardice. It is easier for politicians to approve a $13 billion capital project—which hands out lucrative contracts to construction firms and looks great at a groundbreaking ceremony—than it is to force three public agencies to sit in a room, tear up their labor agreements, and integrate their operations.
How to Double Capacity Without Pouring New Concrete
If we want to fix Penn Station, we must stop thinking like real estate developers and start thinking like industrial engineers. Here is the blueprint to double the station’s capacity using the tracks we already have.
Step 1: Enforce a 4-Minute Dwell Time Limit
No commuter train should sit at a Penn Station platform for more than four minutes.
To achieve this, we must change how trains are operated.
- Step-on, Step-off Crew Changes: Crews must change at the ends of the lines, not at the central hub. When a train arrives at Penn, the engineer who brought it in must step off, and a new engineer must already be standing on the platform ready to step on and drive.
- No Mid-Station Cleaning: Trains should be cleaned at the suburban yards, not while occupying precious platform space in the heart of Manhattan.
Step 2: Fix the Vertical Circulation
The bottleneck inside Penn Station is not the tracks; it is the stairs.
Penn’s platforms are narrow strips of concrete fed by incredibly narrow, slow-moving escalators and staircases. When a train carrying 1,200 people doors-open, it takes five to seven minutes just for the passengers to clear the platform. The train cannot leave because people are still spilling out of the doors.
We must widen the staircases and add high-capacity exits directly to the street level. By stripping away redundant mechanical rooms and unused storage areas underneath the current station footprint, we can double the width of the platform access points. If passengers can clear the platform in 90 seconds, the train can depart in two minutes.
Step 3: Implement Moving-Block Signaling (CBTC)
Amtrak still controls the Penn Station interlocking area with outdated signaling principles. We are still running trains using fixed-block limits that assume braking profiles from the era of steam locomotives.
By installing modern Communication-Based Train Control (CBTC) or European Train Control System (ETCS) Level 2 across the entire Penn Station complex, we can drastically reduce the safe spacing ($S$) required between trains. Trains can follow each other into and out of platforms with mere yards of separation, safely controlled by onboard computers rather than wayside colored lights.
Step 4: Mandate a Unified Dispatching Command
We must strip Amtrak of its unilateral control over Penn Station dispatching.
A single, independent joint-dispatching entity must be created. This unit would have the absolute authority to route any train, from any agency, onto any available track. If an LIRR track is clear, an NJ Transit train gets routed into it. No exceptions, no turf wars, no legacy agreements.
The Hard Truth of Infrastructure
Building "Penn South" is a lazy, expensive cop-out.
It allows transit executives to dodge the difficult work of reforming their operations, renegotiating union work rules, and modernizing their technology. It asks taxpayers to write a $13 billion check to cover up organizational incompetence.
We do not have a space crisis. We have a management crisis. Until we run the trains we have with the efficiency the rest of the developed world takes for granted, we should not be allowed to lay a single new foot of track.
