Summary

  • LACNIC geopolitical-fragmentation analysis asks how cross-border connectivity, route dependence, banking restrictions, data-localisation rhetoric and cloud concentration pressure registry neutrality.
  • When regional politics becomes noisy, portable number resources become more valuable because operators need continuity across customers, upstreams, cables and jurisdictions.
  • A credible ledger stays thin, evidence-bound and portable; a thick political allocation system helps produce the fragmentation it claims to prevent.

A route before a doctrine

A small regional operator in the Caribbean does not meet geopolitics first as a speech, a treaty or a slogan. It meets it as a path. A customer in one island wants a cloud workload in North Virginia to remain reachable from users in Central America. A bank in another jurisdiction wants secure connectivity into a Miami data centre. A content platform wants a local cache announced from a facility in Panama, backed by transit in Brazil and peering in the United States. The operator may hold addresses registered in the LACNIC region, buy transit from providers whose compliance departments sit elsewhere, use equipment financed in dollars, pay upstreams through banks exposed to another legal system, and depend on route filters, RPKI validation, reverse DNS, contact records and abuse desks that must be legible to networks far beyond its domestic regulator.

Nothing in that scene looks theatrical. It is ordinary Internet business. Yet it is exactly where fragmentation risk begins. A route announcement is accepted because many parties quietly trust the surrounding record. A payment clears because a bank is willing to process the relationship. A prefix remains usable because upstreams have no reason to doubt the holder, the origin, the authorization or the continuity of the registration. A cloud provider accepts a dependency because the network identity is portable enough to survive commercial change. The operator is not asking for ideological recognition. It is asking strangers, upstreams, banks and platforms to keep treating its identifiers as valid.

The risk is therefore more subtle than a minister announcing digital sovereignty or a sanctions authority naming a prohibited counterparty. Those questions matter, but they are not this question. National sovereignty asks who should have lawful public authority. Sanctions ask when a legal system may restrict dealings with a target. Geopolitical fragmentation risk asks what happens when the common technical and administrative layers become contaminated by the rivalries, payment restrictions, data-location claims and political expectations that surround them. It is less about a state drawing a line on a map than about the cost of persuading other actors that a route, a holder and a record still belong to the common Internet.

LACNIC matters because it sits in a region where that question cannot be reduced to one state, one bloc or one language. Latin America and the Caribbean contain export economies, island states, dollarized markets, fragile currencies, large telecom incumbents, regional fibre projects, US-facing cloud dependency, European regulatory influence, Chinese infrastructure finance, domestic security politics and many small networks whose practical resilience depends on cross-border continuity. The region is not peripheral to Internet fragmentation. It is one of the places where dependency is most visible because so much connectivity runs through other people's chokepoints and because many operators must turn regional routes into offshore commercial relationships.

The address ledger is one of the few layers that can either soften those dependencies or harden them. If it remains thin, evidence-bound and portable, it gives operators a stable identity across volatile politics. If it thickens into political allocation, social approval, regional loyalty testing or quasi-sovereign discretion, it becomes another chokepoint. A regional address ledger survives geopolitical fragmentation only by staying narrow. When it becomes a political allocator, it helps produce the fragmentation it claims to prevent.

That is the economic danger for LACNIC. Its value is highest when it is least ambitious. It is most useful when it records durable facts that operators, counterparties, courts and networks can verify without absorbing a regional political narrative. It becomes dangerous when it begins to imagine that regional service territory is a mandate to define regional destiny. Latin America and the Caribbean need connectivity that can cross governments, currencies, cables and clouds. They do not need an address book that learns to behave like a frontier post.

The ledger is valuable because it is boring

The most valuable function of a regional Internet registry is also the least glamorous. It preserves uniqueness. It records which network holds which number resources. It makes contactability, transfer history, authorization, reverse DNS, routing-adjacent security assertions and dispute metadata discoverable. It allows other networks to form a reasonable view of whether a route announcement, holder claim or transfer request is consistent with the record they expect to see. That work is not trivial. It is essential precisely because it is limited.

Uniqueness is a narrow technical condition, not a theory of political representation. The same number resource cannot be treated as belonging to incompatible holders inside the same compatibility space without damaging trust. If a registry prevents duplicate claims, preserves historical state, records valid changes and publishes stable data, it is doing work that the Internet genuinely needs. If it preserves RPKI and reverse DNS continuity during disputes, it is protecting the running network. If it records conflict without destroying the last verified operational state, it is reducing harm. None of that requires the registry to decide whether an operator's business model is morally attractive, regionally loyal or geopolitically convenient.

That narrowness is not weakness. It is the source of legitimacy. A ledger that sticks to objective evidence can be trusted by parties that disagree about everything else. A Peruvian access provider, a Mexican content network, a Caribbean government, a Brazilian bank, a US transit provider and a European cloud platform may disagree about law, politics, language, privacy, tax, payment exposure and strategic alignment. They can still rely on a record that says who controls a block, what the last valid transfer was, which autonomous system is authorized to originate it and whether a dispute is pending. The ledger creates a working minimum beneath disagreement.

The moment the ledger thickens, that minimum starts to dissolve. A registry can become thick in many ways. It can turn need assessment into continuing commercial supervision. It can convert regional service territory into political entitlement. It can treat leasing, customer geography or financing as though they were threats to uniqueness. It can use administrative standing as leverage in conflicts that belong in courts or contracts. It can treat a failure to satisfy a soft community expectation as a reason to impair hard operational status. Each move may be defended as stewardship. Economically, each move turns a neutral record into a permission layer.

Permission layers create discounts. They make assets harder to finance, transfers harder to value and continuity harder to insure. A number resource is not valuable only because it is scarce. It is valuable because it can be used predictably in revenue-producing networks. If a holder cannot be confident that the record will remain evidence-bound, the asset loses some of its capital character. If a buyer cannot know whether a transfer will be recognized without discretionary friction, the price falls. If a lender cannot rely on the portability of the registration, collateral value falls. If a small operator cannot plan around administrative risk, it pays more for growth or avoids growth altogether.

The effect is sharper in Latin America and the Caribbean because capital is already expensive in many markets. Currency volatility, shallow credit, imported equipment, dollar financing, uneven competition and regulatory unpredictability raise the cost of network expansion. In such an environment, the registry layer should not add another political premium. Its job is to lower uncertainty around network identity, not to attach a regional political option to it.

This is why the humble image of a ledger is more than a metaphor. A ledger is useful when it is auditable, consistent and separable from the personality of its operator. The records should be able to survive board disputes, institutional failure, litigation, payment disruption, political pressure and transition to a successor operator. The ledger is the continuity object. The registry corporation is a service arrangement around it. Confusing the two is the first step toward hostage architecture.

LACNIC's strongest case, therefore, is not that the region must preserve an incumbent institution at all costs. It is that the region must preserve a trusted, neutral, auditable, portable number-resource record under conditions where geopolitical noise is likely to increase. That case becomes weaker, not stronger, if the institution claims more discretion over the economic and political lives of the resources it records. Boring recordkeeping is not a lack of ambition. It is the discipline that allows many ambitions to coexist on one network.

Fragmentation enters through economics before topology

The word fragmentation often invites images of competing roots, national firewalls or incompatible technical standards. Those images are useful only up to a point. The more common path is less dramatic. Fragmentation enters first through economics: who can pay whom, which bank will process a transaction, which cloud region is acceptable, which cable route is trusted, which data must remain local, which vendor can be used, which counterparty creates compliance risk, which legal order can force disclosure, and which record remains credible when politics changes.

Great-power rivalry makes those questions routine. The United States, China and Europe do not need to split the global Internet formally in order to change the incentives around it. Export controls reshape equipment choices. Banking restrictions reshape counterparties. Cloud regulation reshapes where workloads sit. Industrial policy reshapes data-centre location. Security rhetoric reshapes procurement. Strategic distrust reshapes cable routes, satellite partnerships, content-delivery decisions and peering strategy. The network remains technically interoperable, but the cost of using it as one global market rises. Fragmentation begins when compatibility is still present but trust has become expensive.

Regional blocs add another layer. Latin America and the Caribbean are not one bloc in the way policy speeches sometimes imply. The region contains trade alignments that face north, south, across the Atlantic and across the Pacific. Some economies depend heavily on the United States as a commercial and connectivity hub. Others cultivate Chinese finance or European legal models. Some states are dollarized or dollar-dependent. Others manage volatile local currencies. Some Caribbean islands depend on external transit and external disaster recovery because scale makes domestic redundancy uneconomic. Regional blocs may talk about autonomy, but the operators inside them still buy transit, cloud, equipment, insurance and finance through overlapping spheres. A common LACNIC record sits across this diversity. Its virtue is not that it makes the region politically uniform. Its virtue is that it can help networks remain interoperable without requiring uniform politics.

Data-localisation rhetoric complicates the picture because it translates political anxiety into technical-sounding demands. A government may want citizen data hosted at home, public-sector workloads held by domestic providers, or sensitive records insulated from foreign subpoenas. Some of those goals may be legitimate within a legal system. But localisation rarely contains dependency. A domestic data centre may still rely on foreign equipment, foreign cloud software, foreign content-delivery networks, foreign capital, foreign submarine cables, foreign DNS resolution, foreign security updates and foreign payment rails. Moving a workload across a border does not move the whole dependency chain with it.

Number resources expose the same illusion. An address registered in a regional ledger is not a piece of national territory. It is a globally meaningful identifier whose usefulness depends on recognition by networks outside any one state. A government can regulate operators within its jurisdiction. It can impose lawful duties. It can require records, enforce consumer law or protect critical services. But it cannot make a route globally accepted by assertion alone. Acceptance depends on a surrounding mesh of trust, routing practice, security objects, counterparty decisions and operational convention. The more politics pushes on that mesh, the more valuable a neutral ledger becomes.

Fragmentation risk should therefore be measured not only by whether packets still cross borders, but by the transaction costs attached to making them do so. Do operators need additional legal opinions before announcing a prefix? Do banks refuse routine registry or transit payments? Do cloud providers treat certain origin networks as reputational risks? Do upstreams require redundant proof because registry data are seen as politicized? Do courts in one country distrust records maintained in another? Do resource holders begin to prefer private attestations, national lists or club-based recognition because the common record no longer feels neutral? The practical measure is not a dramatic break in topology. It is the growing paperwork, delay and discount that appear before the break.

Those are early signs of fragmentation. They do not require a formal break. They require only a decline in confidence that common layers are common. In a region already dependent on cross-border capacity, expensive capital and external platforms, a small increase in that confidence cost can have large effects. The harm appears as delayed transfers, higher compliance spend, slower deployments, lower asset values, reduced redundancy and fewer small networks willing to expand.

LACNIC cannot control great-power rivalry, global banking, cable geopolitics or cloud concentration. It can control whether the number-resource record remains narrow enough to be trusted across them. That is a modest role. It is also the role that matters most.

Latin America's dependency map is regional, offshore and cloudy

Latin America and the Caribbean are often discussed as if the relevant question were how to build regional digital autonomy. The operational map is more complicated. A large share of regional connectivity relies on submarine systems that tie countries to a small number of landing points, metropolitan exchange hubs and offshore facilities. Miami has long been a practical junction for traffic, commerce and hosting relationships involving the Caribbean and parts of Latin America. São Paulo, Buenos Aires, Santiago, Bogotá, Panama City and other hubs matter in their own right, but the region's path diversity remains uneven. Geography, scale and capital expenditure ensure that many operators depend on facilities and counterparties outside their domestic political control. The route dependency is therefore regional and offshore at the same time.

Cable chokepoints are not merely physical. They are financial and contractual. A new submarine system requires landing rights, permits, vendor finance, anchor customers, maintenance arrangements, insurance, electricity, data-centre ecosystems and cross-border capacity sales. Island states face harsher economics because redundancy is expensive and demand is smaller. A terrestrial network can sometimes route around a failed path through neighbouring countries. An island cannot pretend that ocean geography is a policy preference. It buys resilience from cables, satellites, caches, upstreams and contracts. In that world, stable network identity is not a luxury. It is part of business continuity.

Cloud concentration deepens the dependency. The major cloud platforms, content networks and security providers do not distribute strategic control evenly across the hemisphere. They place capacity where power, cooling, regulation, customer density, tax treatment, fibre and skilled labour justify it. A bank in the region may want local resilience but still run core dependencies through a global cloud. A media company may distribute content through caches controlled by a handful of platforms. A public service may host locally for legal reasons while depending on external identity, monitoring, software updates and DDoS protection. The address layer has to remain intelligible across all of those relationships.

The economics of IPv4 make the issue more acute. Many networks in the region did not accumulate large historical pools at the scale of early US institutions. Smaller operators face the same scarcity market as everyone else, but with higher capital costs and less leverage. If address transfers, leases or service continuity become trapped in regional politics, the penalty falls hardest on the firms least able to absorb delay. A large incumbent can hire counsel, maintain spare capacity, negotiate with multiple upstreams and wait through a dispute. A growing ISP serving a secondary city cannot easily do that. For it, a delay in recognized resources is a delay in revenue, customers and creditworthiness.

Currency and banking restrictions convert those technical dependencies into balance-sheet risk. Many regional operators earn in local currency and pay for transit, equipment, cloud services, security tools or loans in dollars. When exchange controls tighten, inflation accelerates or correspondent banks reduce exposure, ordinary network operations become harder. A registry fee, transfer payment or leasing arrangement can become entangled with payment frictions that have nothing to do with uniqueness. If the registry's administrative stance amplifies those frictions, it becomes part of the region's cost-of-capital problem. The ledger should be the part of the stack that reduces financing uncertainty, not the part that adds another foreign-exchange or compliance premium.

The temptation in such circumstances is to call for more regional control. That impulse is understandable but often misdirected. More control over the ledger does not create more cables, more cloud regions, cheaper dollars, more redundancy or stronger courts. It can even do the opposite if it makes outside counterparties doubt the neutrality of the record. Latin America and the Caribbean do not become less dependent because a regional institution uses louder regional language. They become more resilient when operators can carry their network identity across suppliers, jurisdictions and political cycles without asking a registry to bless their business logic.

This is where the economics of portability meet the politics of dependency. A portable number-resource claim gives an operator options. It can change upstreams, restructure, finance, sell, merge, lease, move workloads or defend itself in court without turning each change into a struggle for continued recognition. The address resource is not a substitute for cables or capital. It is the stable identity around which cables and capital can be reorganized. When that identity is locked inside a discretionary regional gate, every other dependency becomes harder to manage.

LACNIC's region therefore needs the registry to be less like an industrial-planning ministry and more like a resilient reference layer. It should help the network economy adapt to dependency, not pretend dependency can be abolished by registry policy. In a noisy world, the most useful regional infrastructure may be the part that refuses to make noise of its own, because it gives operators a stable point from which to negotiate everything else.

A regional institution is not a regional sovereign

LACNIC's location in Montevideo gives it a legal home, not a political personality for the hemisphere. Its service region gives it a coordination perimeter, not a territorial title. That distinction is particularly important in Latin America and the Caribbean because the region contains sovereign states with different constitutions, different regulatory traditions, different alignments and different tolerances for foreign influence. No registry can convert that diversity into one regional will. Nor should it try.

The error begins when administrative geography is treated as political geography. A regional registry needs a defined service area so that records can be administered coherently and duplication can be avoided. That does not make the addresses regional property. It does not make the registry a steward of regional sovereignty. It does not make policy meeting participation a substitute for legal authorization from the operators, lenders, customers and states that bear the consequences of registry action. A service map is not a mandate.

The distinction matters because fragmentation risk feeds on symbolic overreach. When a registry presents itself as the guardian of a region, states ask why a private body under one country's law should speak for them. Operators ask why a member process should discipline their assets. Outside counterparties ask whether the record is a neutral technical reference or an instrument of a geopolitical camp. Rival powers ask whether the ledger has become another surface for influence. The more the registry claims, the more actors have reason to contest it.

That does not mean states should seize the registry layer. State capture would create its own fragmentation risks. A registry that becomes an instrument of one government, one coalition or one ideological project cannot remain a common record for operators who must interoperate across borders. Public law has a real role in matters of fraud, contracts, consumer protection, security, sanctions, insolvency and lawful process. But the registry's own function should not become public authority by imitation. It should record facts, preserve continuity and route disputes to proper forums without turning itself into the forum, the claimant and the executioner.

The same point applies to community language. A regional Internet community can produce useful expertise. It can identify problems, debate policy, educate operators and expose risks. But participation is not the same as authorization. A mailing-list speaker, conference attendee, technical contact or consultant does not automatically represent every company, creditor, customer or government affected by a registry decision. The more consequential the decision, the less persuasive informal participation becomes as a mandate. That is not a criticism of participation. It is a boundary around it.

For a thin registry, the boundary is manageable because fewer decisions require a mandate. If LACNIC is preserving uniqueness, recording transfers, maintaining publication services and protecting evidence, the legitimacy burden is comparatively light. The work is testable. The records can be audited. Mistakes can be corrected. Operators can verify much of the substance through network practice and documentary proof. But if LACNIC claims discretion over commercial geography, political acceptability, market morality or regional loyalty, the legitimacy burden grows beyond what the institution can carry.

This is why neutrality is not passive. It is an active refusal to let the institution's service region become a political identity. It requires saying no to governments that want the ledger to enforce their strategic preferences, no to insiders who want community status to become coercive authority, no to market participants who want administrative rules to burden rivals, and no to institutional self-expansion dressed as stability. The registry's duty is not to have no values. Its duty is to keep the common layer narrow enough that many lawful values can coexist around it.

For LACNIC, that discipline may be harder than it appears. Regional institutions often gain prestige by speaking in larger terms: development, inclusion, sovereignty, resilience, security, trust. Those words are not false. The problem is that they can become containers for discretion. A registry can support development by lowering friction and preserving portability. It can support inclusion by reducing arbitrary barriers to resource use. It can support resilience by making records replicable and failover credible. It does not support any of those goals by transforming a neutral ledger into a regional political instrument.

The region needs LACNIC to be legitimate precisely because it is limited. A regional institution that knows what it is not can survive disagreement. One that tries to become the region's digital voice invites every unresolved regional disagreement into its own records.

When geopolitical pressure reaches records

Geopolitical pressure rarely arrives at the registry desk with a label that says fragmentation. It arrives as a request, a payment problem, a court filing, a compliance concern, a security inquiry, a national-interest letter, a bank refusal, a data-localisation demand or an argument that a particular holder should not enjoy ordinary registry services. Each item may look administratively manageable. Together they test whether the ledger is a record of evidence or a tool of pressure.

Banking is the quietest vector. A registry may be incorporated in one country, bank in another, use correspondent relationships in a third and serve members across dozens more. If a payment cannot be processed because of currency controls, banking risk or external compliance rules, the narrow question is how to preserve the holder's operational status while the payment channel is resolved. The dangerous answer is to treat banking friction as proof of technical invalidity. A bank can decide not to process a payment. A registry can comply with the law that binds it. But the inability of one payment rail to carry money should not become a universal statement about whether a number resource remains part of the Internet's valid record.

Legal restrictions create similar problems. Sanctions are one form, but not the only one. Capital controls, anti-money-laundering de-risking, foreign-exchange approvals, procurement bans and beneficial-ownership inquiries can all interrupt ordinary operations. The registry should not pretend those rules do not exist. Nor should it universalize them beyond their legal scope. A Dutch, Uruguayan, US or Caribbean compliance constraint may bind the actor subject to it. It should not silently rewrite the technical validity of the resource for every other network. The common ledger should record relevant constraints when they affect service, preserve evidence and isolate disputes. It should not convert local legal exposure into global invalidity by administrative habit.

State pressure on records is more direct. A government may ask for holder data, dispute a transfer, seek to preserve critical connectivity, request help with abuse, or demand action against an entity it considers hostile. Some requests will be lawful and grounded in proper authority. Others may be political attempts to obtain leverage through the registry layer. The registry's answer should be procedural humility. It should ask what evidence supports the request, what legal authority applies, what the narrow effect should be, how unrelated operations will be protected and where an independent decision should be made. It should not improvise coercion through the address book.

Data-localisation rhetoric can also reach records through the claim that resources issued in a region should serve regional users, stay with regional businesses or reflect regional policy priorities. That sounds attractive in development language but becomes dangerous when applied to number-resource mobility. A prefix may support customers in multiple countries, cloud workloads outside the region, disaster recovery in foreign facilities or a content-delivery arrangement that shifts traffic according to demand. Customer geography is an operating decision unless it breaks a true uniqueness, security or fraud invariant. Treating it as a registry loyalty test turns the ledger into an economic border and punishes the very cross-border resilience that regional operators need.

Route-announcement dependency is where these pressures become visible. An upstream does not need to understand every political dispute. It needs enough confidence that accepting a route will not create avoidable risk. RPKI objects, registry contact data, route filters, transfer records and public disputes all influence that confidence. If the registry record is seen as objective, the upstream can treat politics as external noise. If the record is seen as politicized, the upstream may demand extra proof, refuse ambiguous routes, require costly indemnities or prefer better-known networks. The result is not liberation from foreign influence. It is more dependence on the largest networks and platforms, because only they can afford the new complexity.

The same pattern affects smaller states and operators. A large economy may tolerate extra friction because its market is too important to ignore. A small island network or a regional enterprise provider may not have that leverage. If common records become politically noisy, small networks will be filtered first, delayed first and priced with the highest risk premium. Fragmentation is not evenly distributed. It burdens the least powerful before it reaches the largest.

LACNIC's institutional discipline should be judged by how it handles precisely these small pressures. The question is not whether it can publish broad commitments to stability. The question is whether it can keep a payment issue from becoming a validity issue, a political request from becoming a registry punishment, a local law from becoming global truth, and a record dispute from becoming customer harm. That is the operational meaning of neutrality.

The false comfort of thicker governance

When geopolitical pressure rises, institutions often respond by thickening themselves. They add review, policy, committees, eligibility tests, public-interest language, risk categories and discretionary gates. The move feels responsible because it creates visible activity. It reassures insiders that someone is managing the problem. In a registry layer, however, thickening can make the underlying risk worse. It increases the number of decisions that can be politicized and the number of actors who have an incentive to capture them.

The reason is simple. A thin registry has few levers. It can maintain records, verify evidence, publish data, preserve uniqueness, support security objects, record disputes and execute valid changes. A thick registry has many levers. It can delay, classify, review, freeze, reinterpret, condition, moralize, approve or disapprove. Each additional lever becomes a target for governments, incumbents, competitors, activists, banks, litigants and internal factions. The registry may believe it is building resilience. It is also building more handles for pressure.

In Latin America and the Caribbean, the pressure would not come from one direction. A government worried about foreign cloud dominance may want records to encourage domestic hosting. A large incumbent may prefer transfer friction that slows smaller rivals. A bank may want registry processes to mirror its compliance assumptions. A security ministry may want faster administrative action against suspicious networks. A development constituency may want resource policy to favour particular local narratives. A foreign counterparty may want evidence that the registry will not become a political instrument. The thicker the registry, the more all of them have reason to fight over it.

The economic result is rationing by another name. Scarce address resources become less like portable network assets and more like licenses whose value depends on continued administrative favour. Need tests, usage reviews and regionally framed restrictions may sound egalitarian, but in unequal markets they often favour actors with the best paperwork, largest compliance departments and closest institutional relationships. Small and poor networks do not win from discretion merely because discretion is described as protective. They win from predictable access, clear transfers, portable rights and records that cannot be easily weaponized.

Thick governance also changes the time horizon of investment. A network expansion project depends on confidence that resources, routes and customer commitments will remain usable for years. If registry policy is expected to shift with regional politics or global rivalry, investors price that uncertainty. They may still invest, but at a higher expected return, which means higher prices for users or fewer projects. They may prefer larger incumbents because incumbents can absorb policy shocks. They may prefer foreign cloud or satellite substitutes because those platforms internalize some regulatory complexity. A registry that claims to protect regional development may thereby make regional competition harder.

The false comfort is especially tempting because fragmentation is frightening. If the world feels more divided, a regional body may be tempted to become a shield. But a shield against what? It cannot shield the region from cable geography, currency risk, cloud concentration, export controls, regional-bloc bargaining or great-power pressure. It can shield operators only from uncertainty in its own layer. The way to do that is not to accumulate more discretionary power. It is to reduce the number of questions that depend on discretion.

That is why the institutional-economics answer differs from the political answer. Politics often wants visible control. Economics often wants credible limits. A registry that credibly cannot do much beyond evidence-bound recordkeeping may be more valuable than one that promises to manage every risk. Limits create expectations. Expectations support contracts. Contracts support financing. Financing supports deployment. Deployment supports connectivity. The chain begins with restraint.

Thick governance also corrodes trust among outsiders. A US transit provider, European bank, Chinese equipment financier or Caribbean court does not need to share LACNIC's internal culture. It needs to know whether the record can be relied upon without interpreting regional politics. The more the registry's decisions appear tied to social legitimacy, policy fashion or geopolitical alignment, the more outside parties will discount the record. They may then substitute their own private risk systems. That is one practical path to fragmentation: the public record remains in place, but serious actors stop treating it as sufficient.

LACNIC's best defence against that outcome is not to claim broader authority. It is to narrow the authority it exercises. A registry cannot prevent fragmentation by becoming a miniature geopolitical system. It prevents fragmentation by denying geopolitics unnecessary entry into the ledger.

Alternate roots are a warning, not the model

The familiar analogy for Internet fragmentation is the alternate root. One naming system points one way, another points somewhere else, and users no longer share the same resolution space. The image is powerful because it is easy to understand. It is also an imperfect model for number-resource fragmentation. Addresses and routes can fragment more quietly. They can remain formally inside a shared system while trust, acceptance and portability diverge beneath the surface.

In the DNS, an alternate root is a visible challenge to a naming hierarchy. In the number-resource layer, the equivalent may not be a single rival registry announced with banners and policy papers. It may be a series of private decisions. One group of networks accepts a particular transfer history. Another treats it as disputed. One set of relying parties accepts a security assertion. Another filters it because the surrounding record is politically contested. A bank accepts the holder as ordinary. A cloud platform escalates it. A court preserves a last-known operational state. A registry tries to alter it. A national authority publishes its own view for domestic operators. No one needs to declare a separate Internet. Compatibility erodes by degrees.

That makes number-resource fragmentation more dangerous in some respects. It may not produce an immediate public break that forces reform. It may instead produce a risk premium attached to certain regions, holders, transfers or origin networks. Routes from smaller or politically exposed operators may require more explanation. Leasing relationships may be discounted. Transfers may take longer. RPKI practices may diverge. Private allowlists and reputational databases may begin to matter more than the common registry. Over time, the shared ledger becomes less central because actors with enough market power build their own parallel confidence systems. Fragmentation then arrives as concentration: the common record weakens, and the largest intermediaries become the substitute.

The alternate-root analogy is still useful as a warning because it shows the consequence of turning a common reference layer into a contested authority layer. A common layer works when most participants would rather share it than fight over it. They share it because it is neutral enough, cheap enough, predictable enough and limited enough. Once it becomes a prize, losing factions have a reason to create alternatives. If the address ledger becomes a vehicle for political allocation, the incentive to route around it rises.

Latin America and the Caribbean should care about that incentive because the region has many operators that cannot easily impose their own reality. A hyperscale platform can manage private trust at global scale. A major carrier can negotiate exceptions. A government can direct domestic operators. A small ISP, regional hosting company or Caribbean enterprise network depends much more on the common layer being accepted as common. If that layer weakens, the largest actors gain relative power. Fragmentation often presents itself as sovereignty, but its market effect is frequently concentration.

This is why the solution is not to build a louder regional root for numbers, nor to imagine that LACNIC can become the region's shield through thicker recognition rules. The solution is to make the shared number-resource state more portable, more auditable and less dependent on any one institution's discretion. A record that can be replicated, verified and transferred without political interpretation is harder to fragment. A record that depends on insider status is easier to contest.

That is the deeper contrast with alternate roots. The goal is not to prevent any future divergence by force. Forced unity is brittle. The goal is to make the common record so thin and useful that divergence is usually not worth the cost. Where divergence happens, it should be visible as compatibility choice, not hidden as administrative punishment. Where law requires local refusal, that refusal should remain local rather than being laundered into global invalidity.

LACNIC's fragmentation risk is therefore not that a rival Latin American number root appears tomorrow. The risk is that trust in the existing ledger becomes conditional on politics, and powerful actors quietly reduce their dependence on it. The registry could still exist, meetings could still occur, records could still be published, and yet the practical authority of the common record would have shrunk. That is how fragmentation often looks in its early economic form: not collapse, but discount.

Portability is economic insurance

Portability is often treated as an administrative reform. It is better understood as insurance against geopolitical and institutional failure. A portable number-resource claim allows a holder to preserve network identity when a registry, payment channel, legal environment, institutional board or political climate becomes unsafe. It does not eliminate disputes. It prevents the institution hosting the dispute from becoming the only place where the holder's existence can be recognized.

The insurance value is obvious in a fragmented world. If a Caribbean operator's bank can no longer process a payment through one channel, the operator should not risk losing operational standing while it proves that the payment problem is external. If a state pressures a registry over a politically sensitive customer, the holder should have a way to preserve the last verified state while the dispute is handled elsewhere. If a registry's own governance becomes unstable, the region's networks should not discover that their continuity depends on the survival of a particular corporate shell. If a great-power dispute contaminates certain counterparties, operators should be able to move relationships without renumbering customers.

Portability should not be confused with shopping for the most permissive administrator. The point is not to let holders escape evidence, fraud review or genuine conflict rules. A portable system still needs proof of control, authenticated change history, conflict metadata, protection against duplicate claims, security continuity and transparent treatment of disputes. The difference is that these requirements should attach to the resource state, not to the permanent discretion of one incumbent. The holder should be able to demonstrate validity through records that other qualified actors can verify.

The practical components are straightforward in concept even if hard in implementation. The authoritative state should be replicated and auditable. The chain of changes should be preserved. RDAP, WHOIS, reverse DNS and routing-adjacent security services should have tested continuity paths. RPKI succession should be designed before crisis, not improvised during one. Disputes should be recorded without destroying unrelated operations. Transfers should be judged by evidence, not by political taste. A holder's ability to maintain recognition should not depend on discretionary permission from the very institution it may need to leave.

For LACNIC, portability would strengthen rather than weaken regional trust. Institutions often fear exit because they treat locked-in members as proof of legitimacy. The opposite is true in voluntary coordination systems. Exit rights make continued participation more meaningful. If operators can leave but stay, the registry has to compete on accuracy, neutrality, service quality and institutional restraint. If operators cannot leave without losing network identity, stability becomes confused with captivity.

Portability would also reduce the temptation for states to nationalize the ledger. Governments become more anxious when critical infrastructure depends on a foreign or regional private body with no clean failover. If number-resource continuity is portable, auditable and separable from one institution, the public risk is lower. States can focus on lawful public functions instead of trying to seize a registry lever because no other lever appears available. A thin portable ledger is therefore not anti-sovereign. It is a way to reduce the conditions that make sovereignty panic predictable while preserving the global recognition that national control alone cannot provide.

The economic benefits are equally important. Portable resources are easier to finance because the recognition risk is lower. They are easier to transfer because the buyer can see the history and does not depend on political favour. They are easier to lease because the lessee can evaluate continuity. They support competition because smaller operators are not trapped by the procedural posture of one registry. They support regional resilience because networks can reorganize around cable failures, cloud changes, mergers, insolvency or banking shocks without treating every operational move as a registry crisis.

Portability also disciplines the registry's own imagination. If an institution knows holders can move valid state elsewhere, it has less incentive to thicken its mandate. It cannot easily turn policy discretion into monopoly leverage. It must justify itself as a service operator, not as the embodiment of regional continuity. That is healthy. Critical ledgers should not depend on institutional immortality. They should be able to survive the failure, capture or replacement of the entity operating them.

The design principle is broader than LACNIC. A common layer should contain only what must be common: uniqueness, proof of control, accurate records, security assertions, transfer history, dispute metadata, publication continuity, auditability and exit paths. Business model, customer geography, financing, political preference and ordinary commercial strategy should stay outside unless they directly threaten those invariants. Portability is the mechanism that keeps the boundary credible. Without it, every promise of neutrality is weakened by lock-in.

In a quiet geopolitical climate, portability may look like an efficiency measure. In a noisy one, it becomes the difference between a registry serving the network and the network serving the registry.

Neutrality becomes more valuable when politics gets louder

Neutrality is often misunderstood as indecision. In the registry layer it is closer to discipline. It means the institution refuses to transform ordinary political, commercial or social pressure into number-resource validity. It keeps evidence separate from preference, recordkeeping separate from enforcement, service continuity separate from institutional prestige and regional identity separate from resource title. It is not silence about risk. It is precision about which risks belong in the ledger.

That precision becomes more valuable as geopolitics gets louder. In a calm world, many institutions can appear neutral because no one seriously tests them. In a contested world, neutrality has to be operational. Can the registry preserve a record when a government dislikes the holder? Can it maintain publication services when banks are awkward? Can it record a dispute without poisoning a route? Can it follow the law that binds it without pretending that law is a universal technical truth? Can it resist pressure from large incumbents that want administrative friction imposed on rivals? Can it avoid turning development language into allocation politics?

The answer matters to every actor in the region. Operators need neutrality because their routes and customers cross political boundaries. Governments need neutrality because a politicized registry makes national infrastructure more vulnerable to external contest. Investors need neutrality because assets with discretionary recognition carry a higher discount. Cloud and content platforms need neutrality because they cannot build efficient regional service if the underlying network identifiers become political claims. Users need neutrality because they experience fragmentation as slower service, higher prices, fewer providers and less redundancy, not as a governance theory.

Neutrality also protects the registry. A thin evidence-bound institution is less attractive to capture because there is less discretionary value to capture. A thick institution invites capture because it can decide winners and losers. If LACNIC wants to remain trusted across rival powers, blocs and domestic politics, it should make itself a poor instrument for political ambition. The best registry for a polarized region is one that disappoints anyone hoping to use it as a lever.

This does not mean LACNIC should ignore abuse, fraud or security. It means those issues must be handled through the narrowest appropriate mechanism. Fraudulent transfers should be corrected through evidence. Duplicate claims should be blocked. False authorization should be rejected. Security objects should be maintained according to clear rules. Abuse handling should preserve contactability and cooperation without becoming punishment through resource status. Where coercion is needed, courts and lawful regulators should carry it. The registry should not borrow the language of safety to accumulate enforcement power.

The same discipline applies to regional development. Latin America and the Caribbean do need more infrastructure, more competition, more local hosting, more affordable capacity and more resilience against external shocks. But an address registry does not create those outcomes by political allocation. It helps create them by lowering friction around the scarce inputs operators need. Clear title-like expectations, predictable transfers, transparent records and portability make it easier for small networks to raise capital and serve customers. A registry that suppresses liquidity in the name of fairness may end by protecting incumbents and raising barriers for the operators it claims to help.

Neutrality is also a form of geopolitical realism. No institution in Montevideo can make the United States, China, Europe, Brazil, Mexico, Argentina, Caribbean governments, banks, cloud platforms and cable owners share one strategic worldview. The attempt would be absurd. What LACNIC can do is maintain a ledger that remains useful despite that divergence. The ambition should not be to solve politics. It should be to keep politics from unnecessarily damaging the minimum common record.

There is a quiet moral advantage in that restraint. People often assume the most public-spirited institution is the one that speaks most broadly about the public interest. In infrastructure, the opposite is often true. The public interest is served when a narrow function is performed so reliably that others can build around it without asking who controls it. Electricity grids, payment settlement, land records, shipping documents and address registries all become dangerous when their operators confuse control of a record with command over the underlying society.

LACNIC's neutrality will be judged not by declarations but by edge cases. The politically awkward holder. The unpopular transfer. The cross-border lease. The payment interruption. The state request with weak evidence. The cloud customer outside the region. The operator that changes business model. The court order that requires patience. If the registry can keep those cases bounded by evidence and continuity, it becomes more valuable precisely because the surrounding world is less calm.

LACNIC's regional choice

LACNIC does not face a choice between regional responsibility and market indifference. That is a false pairing. The serious choice is between two kinds of regional responsibility. One treats the registry as a political instrument for shaping outcomes inside a contested region. The other treats it as a thin continuity layer that lets operators, states, courts, markets and users manage their own conflicts without breaking the common record.

The first path will be tempting. It offers status. It lets the institution speak the language of sovereignty, development, security, inclusion and strategic autonomy. It creates more committees, more discretion and more occasions for public relevance. It may satisfy insiders who believe the registry should be more than a bookkeeper. It may appeal to governments that want a visible regional response to great-power rivalry or cloud dependence. But its economics are poor. Every extra political function adds uncertainty to resource value. Every discretionary gate increases the return to lobbying. Every attempt to turn regional identity into allocation logic makes outside networks less certain that the record is neutral. Every thickened rule gives fragmentation another place to enter.

The second path is less glamorous but more durable. It begins from the fact that the region's operators are already cross-border actors even when they are small. Their traffic, customers, suppliers, banks, clouds, cables and security dependencies rarely stop at national borders. They need number resources that behave like portable infrastructure, not like revocable permissions in a regional political project. They need records that courts can understand, banks can trust, upstreams can verify and counterparties can price. They need the registry to be accurate enough to matter and restrained enough not to become the main risk.

For LACNIC, that implies a clear institutional posture. Keep the ledger thin. Make evidence the basis of changes. Publish and preserve state. Separate recordkeeping from punishment. Treat disputes as reasons for metadata and preservation, not for destructive action. Build failover before crisis. Design RPKI and reverse DNS succession as continuity obligations. Support transfer clarity. Avoid customer-geography ideology. Do not convert payment friction into validity judgment. Do not confuse regional meetings with authorization over absent principals. Do not treat the incumbent institution as identical to registry continuity. Above all, make exit and portability normal enough that continued reliance on LACNIC reflects trust, not lock-in.

The alternative is predictable. If the registry thickens, operators will adapt. Large networks will build private assurances. Governments will press harder for domestic control. Courts will be asked to intervene more often. Banks and platforms will add their own risk filters. Small networks will pay more for uncertainty. Transfers will slow. Leasing will be discounted. RPKI and routing trust will carry more politics. The system may still call itself regional coordination, but its economic substance will have shifted toward fragmentation.

The most important fact about fragmentation is that it can be produced by those who claim to resist it. A registry that fears geopolitical disorder may respond by asserting more control. More control invites more contest. More contest weakens trust. Weaker trust encourages private workarounds, state intervention and market concentration. The result is a less common Internet, achieved in the name of protecting the common Internet.

LACNIC's advantage is that it can choose a narrower strength. It can be valuable not because it speaks for the region, but because it does not need to. It can be trusted not because it settles political questions, but because it refuses to smuggle them into the ledger. It can support regional resilience not by allocating destiny, but by preserving the portable network identity around which operators can adapt.

The operator in the opening scene does not need LACNIC to explain geopolitics. It needs a record that survives geopolitics. It needs a route that can be accepted by strangers. It needs a number-resource claim that remains intelligible across banks, clouds, cables, courts and borders. That is a modest demand, and in the present world a demanding one. The registry that can meet it will be more useful than the registry that tries to become important. The future of the regional ledger depends on that humility.

Sources and further reading

These references provide the article's public doctrine and background context. They are used for institutional-economic framing, not for adopting any registry or official-sector narrative.