February 5, 2026
Bloomberg: $35/month. Financial Times: $42/month. The Economist: $17/month. Original analysis by Tatsu with 40+ footnotes: $8/month.
Share this preview with others.
Everyone assumes China can invade Taiwan.
Analysts war-game scenarios. Pundits debate timelines. The Pentagon issues warnings about the "Davidson window," suggesting 2027 as a year of maximum danger. Politicians call for urgent defense spending to prepare for the inevitable confrontation.
Run the numbers. The math doesn't work.
This is Part 4 of our series on the AI Dollar, examining what happens when China's "brittle peer" military (documented in Part 3) attempts to execute an amphibious invasion across 100 miles of contested water against a fortified defender with American submarine support. The short answer: it can't. The physics don't cooperate.
China can punish Taiwan through blockade or bombardment. It cannot conquer it.
Full investigation below. $8/month for novel, footnoted deep analysis.
The Lift Deficit
The fundamental arithmetic of amphibious invasion is the ratio between attacker lift capacity and defender force density. History suggests local superiority of at least 3:1 is required for success. Complex terrain demands more. Urban warfare demands much more.
Taiwan has roughly 88,000 active-duty ground troops, plus reserve and mobilization capacity in the millions. To achieve 3:1 superiority in the initial assault phase, the PLA would need to land somewhere between 300,000 and 400,000 troops.
How many can they actually move across the strait?
Ship Class | Active | Troop Capacity | Vehicle | Total Lift
| | | Capacity |
----------------+--------+----------------+---------------+---------------
Type 075 LHD | 4 | ~1,000 each | ~60 | ~4,000 troops
| | | amphibious |
| | | vehicles |
Type 071 LPD | 8 | ~600-800 each | ~20-30 | ~6,400 troops
| | | vehicles |
Type 072 LST | ~30-50 | ~200 each | ~10 tanks | ~10,000 troops
| | | (direct |
| | | beaching) |
Total dedicated | ~60 | -- | -- | ~21,000-25,000
The entire dedicated amphibious fleet of the PLA Navy can lift approximately 21,000 to 25,000 troops in a single coordinated wave. That's roughly 7-8% of the force required for successful invasion.
The transit time across the Taiwan Strait is 8-10 hours, plus loading and unloading that can exceed 12 hours. Round-trip turnaround for a second wave: 24-36 hours minimum. During which time the first wave, roughly two marine brigades, is fighting 88,000 Taiwanese defenders plus their fortifications, artillery, and anti-ship missiles.
This isn't a gap that clever tactics can bridge. It's a gap that makes invasion impossible with dedicated military shipping alone.
The Civilian Gamble
PLA planners know the math. Their solution: Military-Civil Fusion (MCF), integrating China's massive civilian merchant fleet into the invasion force. The Roll-on/Roll-off (RO-RO) ferries that normally transport cars across the Bohai Sea would become impromptu tank transports, carrying heavy armor to the beaches.
On paper, this works. China has thousands of RO-RO vessels. Mobilized, they could theoretically lift 300,000 troops with their equipment. Problem solved.
In contested water, it's suicide.
The soft-skin problem. Civilian vessels lack military design features that enable survival in combat:
* No compartmentalization. A single hull breach floods the entire vehicle deck, capsizing the ship.
* No damage control. No redundant firefighting or de-watering systems.
* No defensive weapons. Large, slow, radar-bright targets with zero self-defense capability.
Taiwan operates Hsiung Feng II and III anti-ship cruise missiles from mobile coastal launchers. A swarm of missiles against a convoy of civilian ferries is not a battle; it's a massacre. The loss of even a few large RO-ROs, each carrying thousands of troops or hundreds of tanks, would be mass-casualty events exceeding any threshold of political tolerance.
The crew psychology problem. Unlike naval personnel, civilian merchant mariners are not trained or psychologically conditioned to accept high casualty risks. A single mine strike on a ferry filled with conscripted civilian crew might not sink the ship; it might paralyze the entire civilian fleet as crews refuse to sail into waters they now know are lethal.
The Last Mile: Project 019
Even if RO-RO ferries could survive the crossing, they face an insurmountable problem at the destination: they can't offload without ports.
Standard RO-RO ferries require deep-water ports with specialized ramps to drive vehicles off. Taiwan will certainly destroy, mine, or scuttle its port facilities (Keelung, Kaohsiung, Taipei Port) at the onset of hostilities. The Chinese can't drive their tanks off ferries if there's nowhere to drive them.
The PLA's solution is the "Offshore Mobile Debarkation Platform," associated with Project 019. These are modular floating causeways that assemble offshore, allowing ferries to dock on a floating pier that transfers vehicles to lighterage (smaller craft) or directly to the beach.
This technology exists. It works. In calm water.
The Sea State problem. Technical specifications for these systems, similar to the US Navy's Improved Navy Lighterage System (INLS), show they're rated for operations in Sea State 3 or below. Sea State 3 means wave heights of roughly 0.5 to 1.25 meters, about 2 to 4 feet.
In Sea State 4 or higher, the relative motion between ferry ramps and floating causeways becomes too violent. Connectors shear. Ramps buckle. Vehicles fall into the sea.
The Gaza proof. We don't have to theorize about this. In 2024, the US military deployed a floating pier to Gaza for humanitarian aid delivery, a system essentially identical to Project 019. Moderate sea swells broke the pier apart. It was rendered inoperable by normal weather that would be unremarkable in the Taiwan Strait.
The PLA's entire heavy-armor logistics plan, the mechanism for getting tanks and artillery from civilian ferries to Taiwanese beaches, doesn't function if waves exceed four feet. As we'll see, that's most of the year.
The Weather Lock
The Taiwan Strait is not the English Channel. It's subject to two dominant weather phenomena that constrain operations to narrow windows:
Northeast Monsoon (late October through March): Generates sustained high winds and average wave heights of 2-3 meters (Sea State 4-5). The floating causeways don't work. Small landing craft can't operate safely. Helicopters face challenging conditions.
Typhoon Season (July through September): Brings unpredictable catastrophic storms that can scatter a fleet and destroy logistics infrastructure without warning. A typhoon hitting mid-invasion would be decisive in Taiwan's favor.
This leaves two transitional windows when sea states are statistically favorable: April and October. During these months, and only these months, the physics of over-the-shore logistics become possible.
But this creates a fatal strategic problem: surprise is impossible.
Taiwan and the United States know exactly when invasion is physically possible. Any mass mobilization of RO-RO ferries detected in January or July can be assessed as exercises or bluffing. Genuine invasion preparations observed in February or August would provide months of unambiguous warning before the April or October windows open.
There is no Pearl Harbor scenario. There is no bolt from the blue. Everyone can see it coming months in advance.
The Beach Problem
Taiwan's coastline is a natural fortress. Of roughly 1,200 km of coast, only about 10% is suitable for large-scale amphibious operations. Military planners identify approximately 14 usable beaches, the "Red Beaches."
The mudflat trap. Taiwan's west coast, facing China, features shallow waters and extensive mudflats that extend kilometers offshore at low tide. Amphibious ships must anchor far from shore, increasing transit time for vulnerable landing craft and exposing them to fire during long, slow runs to the beach.
The urban interface. Beaches near major cities (Linkou, Bali, approaches to Taipei) are immediately adjacent to dense urban terrain. Invaders don't land in open country where they can maneuver; they land directly into concrete jungles of skyscrapers, industrial parks, and elevated highways. Urban warfare negates China's advantage in mechanized mass and forces slow, grinding, casualty-intensive block-by-block fighting.
The rice paddy obstacle. Taiwan's coastal plains are dominated by rice paddies, flooded agricultural fields that are impassable for wheeled vehicles and treacherous for tracked armor. The Stimson Center's recent tactical analysis highlights how these paddies channel movement onto raised roads and causeways, creating natural chokepoints that Taiwanese artillery can pre-register. There is no "maneuver space." The terrain funnels attackers into kill zones.
The Fuel Clock
Modern mechanized warfare consumes material at staggering rates. A heavy armored brigade can burn tens of thousands of gallons of fuel per day. Without captured port facilities (which Taiwan will destroy), the PLA must establish Offshore Tanker Discharge Systems: flexible pipelines from tankers anchored offshore to fuel bladders on the beach.
This system is fragile. A single artillery shell, mortar round, or saboteur can rupture the line. Without fuel, tanks become pillboxes. Without fuel, the invasion force culminates, unable to advance or resupply.
Logistical analysis suggests that without a functional port captured within 72 hours, the PLA invasion force faces fuel starvation. The advance stops. The defenders counterattack.
The TSMC Paradox
One theory holds that China's goal is capturing Taiwan's semiconductor industry, specifically TSMC's fabs that produce 90% of the world's advanced chips. Take TSMC intact, the theory goes, and China solves its silicon deficit overnight.
This theory fails on inspection.
The ASML kill switch. TSMC's advanced fabs rely on ASML's EUV lithography machines, which require constant maintenance, software updates, and specialized spare parts from the Netherlands. ASML has the capability to remotely disable these machines. In any invasion scenario, that capability would certainly be exercised.
The chemical dependence. Advanced chip fabrication requires specialized photoresists and chemicals, primarily supplied by Japanese companies. In wartime, that supply chain is severed immediately.
The expertise exodus. The engineers who know how to operate these fabs at scale would flee or be evacuated. The machines without the people are expensive paperweights.
The inevitable damage. Any contested landing on Taiwan involves massive bombardment and urban warfare. TSMC's fabs are large, fragile, and clustered in the Hsinchu Science Park. The idea that they would survive invasion intact is fantasy.
China cannot "capture" TSMC. It can only destroy it. An invasion that "succeeds" leaves China holding the ruins of the industry it supposedly wanted.
The Underwater Threat
Even if every other problem were solved, the invasion fleet would face America's decisive military advantage: nuclear attack submarines.
Virginia-class and Los Angeles-class SSNs operating in the Taiwan Strait approaches would create what analysts call a "turkey shoot." Each submarine carries 24+ heavy torpedoes (Mk 48 ADCAP) and missiles. In the congested waters of invasion corridors, every weapon has a high probability of hitting something.
The civilian RO-RO ferries are the perfect targets: slow, loud, radar-bright, with no anti-submarine defenses. The Type 093B submarines meant to protect them are, as documented in Part 3, acoustically equivalent to 1980s designs, easily detected and killed by modern American boats.
The reload advantage. US submarines can reload from tenders operating out of Guam or forward positions in Japan and Australia. The kill zone in the strait can be sustained indefinitely.
The math of attrition. If American submarines sink 10% of the RO-RO fleet in the first 24 hours, the invasion timeline breaks. If they sink 20%, it's over. The combination of submarine attrition, mine strikes, and missile attacks would almost certainly exceed these thresholds.
The Warning Window
The mobilization required for an actual invasion cannot be concealed.
RO-RO requisition. Commandeering hundreds of civilian ferries would cause them to disappear from global shipping trackers. Chinese domestic supply chains would be visibly disrupted. This signal is unambiguous.
Medical preparation. Preparing for tens of thousands of combat casualties requires mobilizing blood banks, field hospitals, and reserve medical personnel. This activity is impossible to hide.
Troop movements. Concentrating the necessary ground forces in Fujian Province involves train movements, housing, and logistics that are visible to commercial satellite imagery and signals intelligence.
Analysts estimate the warning time for genuine invasion preparations at 30 to 90 days. This window allows Taiwan and the United States to:
* Disperse aircraft to hardened shelters and auxiliary airfields
* Mine port approaches and potential landing beaches
* Deploy submarines to ambush positions
* Move mobile missile launchers to concealed locations
* Evacuate or reinforce key facilities
The PLA's invasion plan requires speed and mass. The mobilization timeline guarantees Taiwan gets months of preparation.
The "Peace Disease"
Underlying all tactical and logistical challenges is a fundamental institutional weakness: the PLA has not fought a major war since 1979.
Amphibious invasion is the most complex military operation possible, requiring precise synchronization of naval lift, air cover, missile strikes, logistics, and ground combat. The choreography must work under fire, against a thinking enemy, in chaotic conditions.
The PLA trains for this. But training is not combat. The difference between exercises and war is the difference between sparring and a street fight. Fog of war, friction, the thousand small things that go wrong, these multiply in actual combat. Institutional experience matters.
Internal PLA assessments frequently cite problems with "jointness," the ability of different services to coordinate. Command structures are rigid. Communication protocols fail under stress. Attempting to integrate thousands of civilian vessels with military operations, in real-time, under fire, would add friction that invites catastrophe.
The one-child sensitivity. China's demographic structure creates a unique political vulnerability. The PLA is largely composed of soldiers from one-child families. The loss of a troop transport carrying 3,000 only-sons would have social and political reverberations that no Chinese leader wants to face. This creates pressure for caution precisely when aggressive action is required.
What China Can Do
None of this means Taiwan is safe. China has escalation options short of invasion:
Blockade. Interdicting Taiwan's seaborne trade and energy imports. Taiwan has limited strategic reserves. A prolonged blockade, especially of LNG shipments, could create economic crisis. But blockade is slow, visible, and invites international response and sanctions.
Bombardment. Missile and air strikes against Taiwanese infrastructure. Punitive, but doesn't achieve territorial control. Creates international outrage without strategic gain.
Quarantine. A partial blockade focused on specific goods or routes. More deniable, but also less effective.
Coercion. Continued gray-zone pressure, military exercises, air defense identification zone violations, designed to exhaust Taiwan's readiness and erode morale over years.
These options are real. But none of them puts PLA boots on Taiwanese soil or a Chinese flag over Taipei.
The Conclusion
A successful cross-strait invasion requires:
* Achieving 3:1 local superiority with 20,000-troop lift capacity
* Protecting soft-skinned civilian ferries from missiles, submarines, and mines
* Operating floating causeways in Sea State 3 or below
* Invading during April or October only (no surprise)
* Landing on 14 viable beaches into urban and rice-paddy terrain
* Establishing fuel supply within 72 hours without captured ports
* Somehow capturing TSMC intact despite bombardment
* Evading American submarine interdiction throughout
* Executing all this with an untested military and corruption-compromised equipment
The chain has too many links. The failure of any single link, a typhoon, a successful minefield, a breakdown of the floating pier system, leads not to a setback but to catastrophic failure of the entire invasion force.
Until the PLA builds a massive, hardened military amphibious fleet capable of operating independently of ports and weather (a project that would take another decade or more), the physics of the Taiwan Strait remain the island's most formidable defender.
China's leaders know this. That's why they rattle sabers but don't invade. The rhetoric serves domestic political purposes. The reality is the math doesn't work.
In Part 5, we'll return to the United States and examine the Trump administration's technology policies. Are they consciously building the AI Dollar infrastructure we described in Part 2? Or stumbling into it by accident?
Notes
Notes
[1] PLA amphibious lift capacity from Asia Times and CIMSEC.
[2] Taiwan active forces from Ministry of National Defense annual reports and IISS Military Balance.
[3] RO-RO vulnerabilities from Defense Priorities.
[4] Project 019 Sea State limitations from DTIC technical review of modular causeway systems.
[5] Gaza pier failure from multiple news sources, June-July 2024.
[6] Weather windows from RAND Taiwan analysis.
[7] Red Beaches from YouTube documentary analysis and defense publications.
[8] Rice paddy obstacles from Stimson Center.
[9] CSIS wargame methodology and results from "The First Battle of the Next War".
[10] Submarine reload logistics from USNI Proceedings.
[11] Mobilization warning timeline from CSIS economic indicators study.
