I. Why the First Region Has to Be Done Twice
Ten papers ago this series began with North America because that is where the concept began. Gottmann named Megalopolis on the Northeastern seaboard, the Regional Plan Association’s America 2050 project produced the eleven-region map that most readers carry in their heads, and the analytical vocabulary of commuting fields and interlocking metropolitan areas was built to describe that continent.
The cost of starting there has gone unpaid until now. Every paper after the first was written as a comparison — how does primacy differ from coalescence, how does water differ from land, how does a legally permeable border differ from a controlled one — and the term on the right side of each comparison was always North America. It became the reference standard, and a reference standard is never itself measured. The first paper’s typology of twelve binding agents and ten morphologies was derived from North American cases; when a category fit North America it was because North America had generated it.
The framework now stands at roughly twenty binding agents in five families and sixteen morphologies, and thirteen of those binding agents were discovered somewhere else. This paper asks the question the series has been deferring: what does the continent look like when the whole apparatus is turned back on it?
The answer is that the first paper was wrong in ways that are systematic rather than random. It was not wrong about what it saw. It was wrong about what it did not think to look for, and the omissions cluster in three of the five families — resource-binding, movement-binding, and authority-binding — which is to say in exactly the families that the Americas did not teach the framework. The first two papers described a world bound by jobs and goods. The rest of the series established that this was a description of two continents rather than a definition. It now turns out that it was not even an adequate description of one of them.
II. Corrections to the Span
The original inventory listed roughly eighteen formations across three countries. Four categories of correction are needed.
Omissions of territory. Puerto Rico was absent entirely. This was a defensible choice under the first paper’s framework, since a single metropolitan labor market covering an island of three million does not look megaregional by any North American measure. It is not defensible under the Caribbean paper, which established that a metropolitan area coextensive with a jurisdiction acquires megaregional political salience without megaregional size, and which classified San Juan alongside Kingston and Santo Domingo. Puerto Rico belongs in both inventories, and its presence in the North American one supplies the continent’s clearest contraction case.
Omissions of formation type. The original inventory contained no purely extractive camp systems, because Type K — rostered-commute extraction — did not exist until the Oceania paper found it in the Pilbara. Northern Alberta’s oil sands operation, the Bakken in North Dakota, and the Permian camp complex are large, structurally distinct, and were classified in the first paper either as fringes of other regions or not at all.
Omissions of binding system. The Colorado River basin was not in the original inventory in any form, because it is not a settlement formation. Under the MENA paper’s Type O it is a binding agent operating across four named North American megaregions simultaneously, and it is arguably the most consequential single structure on the continent’s western half.
Omissions of internal boundary. Mexico’s northern border free zone — a strip along the frontier with reduced value-added tax, reduced income tax, and a separately set minimum wage — is an internal fiscal seam inside a single sovereign state. The East Asian paper introduced M16 for exactly this configuration in the Greater Bay Area and explicitly flagged that weaker members of the class probably existed elsewhere. The Mexican free zone is one, and it sits directly underneath four formations the first paper did classify.
III. Binding Mechanisms Reconsidered
The additions
Type O — hydraulic — is the largest omission in the original paper. Four formations were classified there as Type E, amenity-migration: Florida, the Arizona Sun Corridor, the Front Range, and, partially, Southern California. Type E was and remains correct about why people arrived. It says nothing about whether they can stay, and in the western three cases the answer is set by a single water system that is simultaneously the precondition of settlement and its ceiling — the MENA paper’s definition, written for Iran and the Gulf, transferred without modification.
The current state of the Colorado is a precise illustration of what Type O binding does to governance. Reclamation released its Final Environmental Impact Statement for post-2026 operations at the end of July 2026, replacing the 2007 Interim Guidelines and the 2019 Drought Contingency Plans as they expire. The framework runs ten years through 2036, allows Lower Basin shortages of up to 3.0 million acre-feet in dry years, and was issued by the federal government because the seven states and Mexico had not reached a consensus deal in time for the October 1 deadline. The river supplies municipal water to roughly forty million people and sustains 5.5 million acres of farmland.
Under the framework’s amended governance proposition — statutory creation predicts survival, ownership predicts effectiveness, and the two are independent — this is nearly a laboratory case. The Colorado River Compact has statutory permanence and no allocation mechanism that can compel anyone; it has therefore survived a century and settled nothing. The Bureau of Reclamation owns the dams. When negotiation failed, the party that owned the asset wrote the rules. The series has stated this proposition nine times using foreign cases; the sharpest instance was on the first continent all along.
Type O also has an inundation mode, added in the Southeast Asian paper, and coastal Louisiana is a North American case of it.
Type I — shared-hazard — was introduced for Caribbean cyclone exposure and applies to Florida and the Gulf Coast with greater force. What makes it a binding agent rather than a shared risk is the insurance market: a single reinsurance pricing environment, a state-backed insurer of last resort, and a construction-cost structure set by wind and flood standards, all operating at the scale of the formation rather than the municipality. The first paper listed insurance as a fragility of Type E regions. It is better understood as a separate binding agent with its own instruments, and California’s wildfire and residual-market situation is the second North American instance.
Type K — rostered-commute extraction — was present in Alberta and the Dakotas before it was found in Western Australia. The correction matters beyond bookkeeping, because the European paper drew a policy conclusion from Type K’s absence there: Europe’s extraction regions were settled, so when the industry ended they contained millions of people and became a permanent policy problem, while the Pilbara is engineered so that this cannot happen. North America is the only region in the series that contains both models at scale. Appalachia and the Rust Belt are the settled version and are living with the bill. Fort McMurray and the Permian camps are the rostered version. The continent is therefore the natural site for the comparison the European paper called for and could not run.
Type H — remittance-diaspora — binds the North American interior to central and western Mexico at an intensity the first paper did not register. The US–Mexico corridor is the largest remittance corridor in the world. Inflows reached $61.8 billion in 2025, a 4.6 percent decline that was the largest since 2009, amounting to about 3.4 percent of Mexican GDP. A one percent tax on cash, money-order, and cashier’s-check remittances took effect at the start of 2026, and the receiving population is concentrated in rural and semi-urban areas, with BBVA estimating that remittance income keeps roughly 1.1 million Mexicans out of multidimensional poverty.
The Caribbean paper’s finding was that circular, multi-generational migration produces populations functionally inside two labor markets and fully counted in neither. That describes Zacatecas, Michoacán, and Guanajuato as accurately as it describes Jamaica, and the Bajío — classified in the first paper purely as Type C production-sharing — is a Type C and Type H formation simultaneously.
Type R — circular internal migration — appears in North America in the seasonal agricultural visa circuits, which are the voluntary-circular mode rather than the administratively-constrained mode the East Asian paper distinguished. The remedy the series has repeatedly recommended for the voluntary mode is entitlement portability, and it applies here.
Type F — administrative-primate — was underweighted. The first paper treated Type F as essentially a South American condition with Mexico City as a partial case. Mexico City is not a partial case. It is the hemisphere’s second-clearest instance after Lima, and Guadalajara is a secondary one.
Type G — maritime-networked — covers the Yucatán–Caribbean corridor’s cruise economy, which the first paper folded into amenity-migration.
Type J — regulatory-arbitrage — has a North American form in the corporate-domicile and trust-jurisdiction industries and, more consequentially, in the state-level tax and regulatory differentials that shape interregional migration. Whether household-level arbitrage rises to the level of a binding agent is arguable; firm-level arbitrage in Delaware, Nevada, and South Dakota clearly does.
The reclassification that matters most
The first paper classified the four dense border crossings as Type C production-sharing with M5 seam morphology, and called their cohesion policy-contingent. The full framework says something sharper.
North America contains the world’s largest instance of a suppressed Type A. Every precondition the European paper identified for a cross-border commuter field is present at San Diego–Tijuana and at Paso del Norte: direct adjacency, a steep wage gradient, dense pre-existing social ties, and enormous demonstrated crossing demand. The one missing element is the legal instrument. USMCA and its predecessor liberalized the transaction family comprehensively and the labor family essentially not at all, which makes North America the exact mirror of Europe, where the labor regime is complete and the infrastructure lags.
Under the seam taxonomy’s seven configurations, the US–Mexico crossings are goods-only with asymmetric personal mobility — not simply “seams.” And the Southeast Asian evidence establishes that the European solution is not the only one available: Johor–Singapore sustains more than 300,000 daily crossings across a wage gradient far steeper than the US–Mexico differential, with no free movement whatsoever, using a work-pass class plus dedicated crossing capacity. That is a reproducible template, and the first paper did not know it existed.
Two further seam corrections. Cascadia contains a seam the original paper did not treat at all — Vancouver–Seattle, asymmetric-open under trusted-traveler programs — because Cascadia was filed as Type B and the border was left out of its description.
And Detroit–Windsor has just changed category. The Gordie Howe International Bridge opened to traffic on July 27, 2026, the first new vehicle link between the two cities in nearly a century; until then the Ambassador Bridge, opened in 1929, had been the only international bridge connecting them. The crossing carries six bridge lanes, sixteen toll lanes, and sixty Canadian and US inspection lanes. Detroit–Windsor was never M12, severed — the legal regime always permitted crossing. It was capacity-constrained, which is the M13 condition the Sub-Saharan paper isolated at Kinshasa–Brazzaville, in a milder form. It is now a repaired seam, and it joins Öresund as the series’ second observable natural experiment on whether adding a fixed link converts adjacency into integration. Öresund took roughly a decade to produce a functioning commuter field. The Windsor case has a variable Öresund lacked: the joint opening celebration was cancelled after tariff escalation, and the two governments held separate events. A repaired seam under a deteriorating trade regime is a configuration the series has not yet seen, and it is worth watching deliberately.
IV. The Typology Applied Whole
| Formation | Paper I | Revised | Morphology | Consolidation |
|---|---|---|---|---|
| Northeast | A | A / T (emerging) | M1 | Consolidated; growth now migration-dependent |
| Great Lakes | A / C | A / C / D-legacy | M1 | Consolidated; interior contracting |
| Southern California | A | A / O / I | M2 | Consolidated; net loss 2025 |
| Northern California | A | A / J | M2 / M3 | Consolidating |
| Central Mexico | A | A / F / O | M2 | Consolidated; aquifer-constrained |
| Guadalajara | — | F / C | M2 | Consolidating |
| Windsor–Quebec | B | B / F | M3 | Consolidated; jurisdictionally covered |
| Texas Triangle | B | B | M4 | Consolidating rapidly |
| Piedmont Atlantic | B | B | M4 | Consolidating |
| Cascadia | B | B | M3 / M5 | Consolidating |
| Calgary–Edmonton | B | B / D / K | M3 | Emergent |
| Bajío | C | C / H | M4 | Consolidating |
| Monterrey–NE Mexico | C | C | M4 / M5 | Consolidating |
| Gulf Coast | D | D / K / I / O (inundation) | M3 / M4 | Consolidated, low-density |
| Florida | E | E / I | M4 | Consolidated; insurance-constrained |
| Sun Corridor | E | E / O | M3 | Consolidating |
| Front Range | E | E / O | M3 | Emergent |
| Yucatán–Caribbean | E | E / G | M3 | Emergent |
| US–Mexico seams (three) | C / M5 | C with suppressed A | M5 (goods-only/asymmetric) | Policy-contingent |
| Detroit–Windsor | C / M5 | C / A-partial | M5 repaired (2026) | Newly altered |
| Oil sands / Bakken / Permian | — | K | M9-adjacent | Cyclical |
| Appalachia / Delta | — | D-legacy | M4 | Contracting |
| Puerto Rico | — | F / H | M8-adjacent | Contracting |
| Northern Mexico free zone | — | J / C | M16 | Stable |
| Colorado Basin | — | O | (cross-cutting) | Federally administered from 2026 |
Four observations follow.
First, North America is the world’s Type B heartland, and the series now shows why. Business-travel and headquarters binding across several hundred kilometers of empty country is common here and rare almost everywhere else. Europe binds comparable distances through Type A because the legal regime permits daily crossing. East Asia binds them through Type C and infrastructure. Africa binds them through Type Q. The preconditions for Type B — dense cheap domestic aviation, one language, one currency, one legal system, and corporate structures organized at continental scale — coincide here and in few other places. The Australian paper found a partial analogue and nothing else in the series matched it. Type B is less a general category than a North American and Australian specialty, and the first paper had no way to know that.
Second, the resource family now touches nine formations that were classified without it. This is not a marginal adjustment. It changes which instruments are relevant: high-speed rail is a Type A and Type B instrument, and proposing it for the Front Range or the Sun Corridor addresses neither of those regions’ binding constraints. Their constraint is an allocation regime on a river, and the relevant instruments are pricing, forbearance agreements, and the ten-year federal framework now in force.
Third, the consolidation qualifier needs the values the series added later. The first paper used consolidated, consolidating, emergent, and policy-contingent. It had no value for contracting — added at Silesia and Northeast China — and North America has at least four contracting formations, none marked as such in the original.
Fourth, M14 — the saturated plain — is absent, and its absence explains the concept’s birthplace. The megaregion depends on interstitial emptiness: cities as islands in a materially less-settled surface. North America has that everywhere. Rural Bihar does not. The concept was invented here because here it describes something true, and the eighth paper’s difficulty was the mirror image of this continent’s convenience.
V. Governance: The Gap Reassessed
The first paper’s verdict was that every North American megaregion is a jurisdictional orphan — too large for a metropolitan planning organization, too oddly shaped for a state or province, and in several cases internationally divided. That verdict stands in outline and is wrong in three particulars.
The continent has one formation that satisfies the series’ central governance condition, and the first paper undersold it. The finding confirmed across five continents is that megaregional governance succeeds where one general-purpose jurisdiction covers the formation. The Greater Golden Horseshoe is substantially covered by Ontario, and Ontario’s growth plan is a binding instrument to which municipal official plans must conform. That is a general-purpose jurisdiction holding a compulsory competence over the whole formation — the arrangement the series spent ten papers looking for. It is the North American case that matters most, and it was listed in the first paper as one line in a corridor.
The ownership proposition explains the North American failures precisely. Metropolitan planning organizations own a revenue stream — federal transportation formula funds — and therefore survive and function; their defect is geographic, not institutional. The megaregional coalitions and corridor advocacy bodies own nothing, and have mostly dissolved. The bi-state port authorities own assets and revenue and are the most durable multi-jurisdiction bodies on the continent. And Mexico’s Comisión Ambiental de la Megalópolis holds a compulsory competence over air quality across six federal entities, which is why it persists where planning-only bodies have not. Every prediction the proposition makes is confirmed here, which the first paper could not have shown because the proposition had not yet been stated.
The relief-city verdict has a North American dataset the series never counted. Ten papers accumulated six state-built relief cities across five regions — Brasília, Abuja, New Administrative Capital, Naypyidaw, Nusantara, Xiong’an — with one consistent result: administrative functions relocate, economic functions do not. North America ran this experiment first and with roughly thirty replications, in the nineteenth-century American practice of siting state capitals away from the largest city. Albany and New York, Sacramento and Los Angeles, Springfield and Chicago, Harrisburg and Philadelphia, Olympia and Seattle, Jefferson City and St. Louis. In every case the administrative function moved and stayed moved; in no case did the economic center follow. The verdict does not rest on six modern cases. It rests on dozens, with two centuries of follow-up, and the confirming evidence was on the first continent the series examined.
On jurisdictional redraw, the tenth paper concluded from Japan’s municipal mergers and Vietnam’s 2025 provincial consolidation that redraw is achievable through sustained financial incentive rather than requiring single-party authority. North America’s record is mixed and instructive: US city-county consolidations have mostly failed at referendum with a few durable exceptions, while Canada’s provincially imposed amalgamations succeeded and were partially reversed. The lesson consistent with the Japanese case is that incentive outperforms compulsion, and North America has never tried sustained incentive at the relevant scale.
VI. Objections
The paper reclassifies rather than discovers. Every North American fact recorded here was available in 2001. What has changed is the vocabulary. That is a real limitation, and it means the paper’s value depends entirely on whether the categories are doing work — whether calling the Sun Corridor Type E and O produces different recommendations than calling it Type E. It does, but the reader is entitled to check.
Multiple assignment weakens the typology. Nearly every revised row carries two or three binding agents where the original carried one. A classification in which most cases are multiply classified is closer to a checklist than a taxonomy. The defense is that the five families remain mutually distinct, and that most formations are bound in two families rather than in two ways within one — which is a real structural claim rather than analytical hedging.
The framework’s own bookkeeping does not reconcile. The series has been reporting nineteen binding agents while the lettering appears to run A through T. One of these is wrong, and the discrepancy is the kind that accumulates unnoticed in a project maintained across many sittings. It should be reconciled against the source papers before the framework is used further.
Type J may not be a binding agent in this region. Regulatory arbitrage between US states is a real driver of firm and household location, but calling it a binding agent risks classifying every price differential as cohesion. The Caribbean form — offshore finance producing a formation with no other reason to exist — is cleaner, and the North American form may be a flow rather than a bond.
VII. The Finding That Changes the Series’ Forecast
The first paper’s growth section assumed continued growth and asked only about its distribution. That assumption no longer holds on either side of the northern border.
US population growth slowed to 0.5 percent in the year to July 2025, reaching 341.8 million — the slowest since the early pandemic — as net international migration fell from 2.7 million to 1.3 million, a decline of nearly 54 percent, with Census projecting roughly 321,000 by July 2026. Natural increase was 519,000, well below the 1.6 to 1.9 million range of the 2000s, and California recorded a net loss of 9,500 people in 2025 after gaining 232,000 the year before.Canada’s population fell by 55,025 in the first quarter of 2026, to 41.4 million, on lower permanent admissions, a decline in non-permanent residents, and negative natural growth.
Permanent resident admissions were down 20.2 percent year over year, and the non-permanent resident population fell by nearly 118,000 in the quarter. Under the 2026–2028 plan, new temporary resident arrivals fall from 673,650 in 2025 to 385,000 in 2026, a 43 percent reduction, with the aim of bringing the temporary population below five percent of the total by the end of 2027.
This is Type T — absorptive concentration — arriving in North America, and it arrives with a difference the East Asian paper could not have anticipated. In Japan and Korea the condition is demographic: the country cannot reproduce itself, and the megaregion grows by draining provinces that are emptying anyway. In North America the same condition is being produced by policy, quickly, and in principle reversibly. Mexican fertility is now below replacement; Canadian and US natural increase are near or below zero; and the variable holding the whole system up is a migration regime that both northern governments have deliberately reduced within eighteen months.
The consequence for this series is direct. Ten papers treated megaregional growth as evidence of vitality and contraction as evidence of trouble, and the tenth paper established that both readings fail under Type T. North America is now entering the condition where they fail. On present trajectory, the Texas Triangle and the Piedmont will keep growing by taking population from the Northeast and the Great Lakes rather than by adding it, which makes megaregional planning a zero-sum interregional exercise for the first time on this continent — and makes the European literature on managed contraction, which the fifth paper found nowhere else, the relevant literature for parts of the American interior within a decade.
VIII. Conclusion
The revision produces one methodological result and one substantive one.
The methodological result is that the region a framework is built on is the region it describes worst. Every category North America generated fit North America by construction, and the categories North America did not generate were invisible here even where the cases were large, old, and obvious. Alberta’s camp system predates the Pilbara’s. The Colorado’s allocation crisis predates Iran’s. The remittance corridor to Michoacán is larger than any in the Caribbean. The state-capital experiment predates Brasília by a century and a half. None of these were seen until a smaller, stranger, more constrained region forced the concept into existence. Whatever else the series demonstrates, it demonstrates that the periphery is where the vocabulary for the center gets made.
The substantive result is that North America is less exceptional than the first paper implied and more instructive than any other single region. It contains both extraction models, both seam repair mechanisms, the world’s largest suppressed commuter field, the sharpest instance of the ownership proposition, the longest-running relief-city dataset, and now the first case of Type T conditions produced deliberately rather than demographically. It is not the standard against which the others should be read. It is the region where the most of the framework can be observed operating at once, which is a different and more useful thing to be.
