LACNIC's Electoral Commission and the Independence Test

Summary

• LACNIC's bylaws give the five commissioners authority to supervise and certify statutory elections, verify candidate documents, investigate incompatibilities, exclude or restrict candidacies, count votes, determine winners and call an Extraordinary Assembly after serious irregularities.
• Structural diversity rules limit commissioners from the same country and from the same employer or related organisation. Three-year staggered terms reduce the ability of one annual majority to replace the whole body at once.
• The Board still writes and approves major election regulations, competence standards, conflict rules and transparency guidelines. Public materials do not clearly identify a protected commission budget, independent legal counsel, compulsory evidence powers, a comprehensive reasons standard or an external appeal route.
• No evidence reviewed establishes that a named commissioner accepted improper direction. Independence is a systems question: whether a conscientious commissioner can obtain records, resist pressure, publish a reasoned decision and make it effective when the board, staff or an incumbent is affected.
• LACNIC should publish an independence charter covering appointment, recusal, resources, evidence custody, hearing rights, reasons, review, enforcement and annual assurance while preserving the commission's elected and regionally diverse character.

Formal authority is the beginning of independence

Article 20.3 supplies unusually broad language for a regional registry election body. The Commission does not merely observe a count. It verifies documents, resolves incompatibility challenges, may investigate on its own initiative, can remove or restrict candidates, performs the scrutiny, declares winners and may summon an Extraordinary Assembly when irregularities are serious. Those are adjudicative as well as administrative powers.

Five elected members share the same duties. Their three-year terms are staggered, and no more than one may come from the same country or be linked through employment or consultancy to the same organisation or related company. These rules disperse nationality, workplace interest and replacement timing. They create a stronger starting position than an ad hoc panel appointed after a dispute.

Yet independence is not identical to formal competence. The Board has special-majority authority to draft and approve the rules on incompatibility, candidate competence and suitability, electoral conflicts and the election process itself. It also defines transparency guidelines. The Commission applies a field whose major boundaries are set upstream by the body whose composition elections help determine.

That allocation can be legitimate. Boards commonly propose administrative rules while independent bodies apply them. The safeguard lies in constraints: rules should be prospective, published before nominations, consistent with bylaws and not amendable in response to a particular candidate or complaint. The Commission should be able to state that a board rule conflicts with the bylaws and refer the issue to the Assembly or an independent reviewer.

Public descriptions say commissioners work online through email and LACNIC's election system and follow the same ethics code as staff. They do not clearly disclose whether the Commission controls a budget, retains counsel without board approval, compels preservation of platform records, receives complete staff incident reports or publishes reasons for every material exclusion. Those missing safeguards do not prove dependence. They identify where independence cannot be verified from institutional design alone.

The five-part test

The first test is appointment. Commissioners are elected rather than selected solely by the Board, staggered terms limit abrupt capture, and nationality and organisational-affiliation limits improve pluralism. The remaining questions are whether incumbents control access to candidacy, whether campaign resources are equal and whether vacancies can be filled without creating a board-dependent majority.

The second is resources. A commission that must ask an interested executive for every legal opinion, technical audit or translation cannot exercise broad authority confidently. A protected annual allocation, authority to retain independent expertise and publication of aggregate spending would make the formal mandate usable without creating an unaccountable institution.

The third is evidence access. Certification requires more than the final tally. Commissioners need the frozen voter register, authorisation history, system logs, incident tickets, candidate declarations, conflict disclosures and a chain of custody for provisional results. Access should be direct or mirrored, not filtered through a party whose conduct is under review. Privacy can be preserved through restricted inspection and redacted reasons.

The fourth is decisional autonomy. Procedures should identify quorum, recusal, burden of proof, urgency measures, response rights, materiality and available remedies. A candidate should know whether an exclusion rests on ineligibility, conflict, misconduct or defective documents. A board or staff member should not be able to convert advice into an order outside the published route.

The fifth is effect and review. A commission is not independent if its certification can be ignored, but unreviewable finality creates a different risk. Decisions should bind operational staff immediately while allowing a narrow, time-limited appeal to an identified Assembly mechanism, arbitrator or court on jurisdiction, fairness and manifest error. Review should not become a way to rerun political merits.

LACNIC meets meaningful parts of this test in its bylaws. The undeclared parts deserve an independence charter rather than assumptions about personal virtue.

Membership is a constitutional sorting decision

The word “member” sounds descriptive. It appears to identify the organisations that use a service, fund an association or belong to a community. At LACNIC, it does more. Membership is a constitutional sorting decision that determines who may enter the Assembly with a vote, who may nominate candidates, who may help compel extraordinary deliberation and who may speak without deciding.

That distinction is easy to miss because technical coordination creates a broad circle of dependence. Internet service providers, universities, public agencies, community networks, content companies, hosting firms and enterprises all rely in different ways on accurate number registration. Some receive address space directly. Some receive it through national registries or providers. Some hold only an autonomous system number. Others operate in the region while managing address space issued through a different historical route. Their networks can be equally exposed to registry continuity even when their corporate standing is not equal.

LACNIC’s current bylaws make the political conversion explicit. Article 5 says members hold power within the institution and identifies the Assembly as its governing body. Articles 6 and 7 then divide members into categories and attach different rights to them. Article 19 adds a second distinction by assigning Active A members a graduated number of votes according to address space.

This is not an accusation that classification is illegitimate. Every association needs boundaries. A regional registry must distinguish customers, members, entities, staff, directors and the wider public. Equal technical dependence does not automatically establish an equal claim to corporate control. The analytical question is narrower: what characteristics does the system convert into political power, and are those characteristics still a defensible proxy for the interests that governance decisions affect?

The answer matters because category design can shape outcomes before any ballot opens. A debate about board quality, fees or constitutional reform may appear open to “members,” yet the decisive constituency has already been selected by resource type, service path, payment status and address scale. Governance begins at that earlier boundary.

Three categories conceal several different gates

The bylaws identify Active members and Adhering members, with Active membership divided between Active A and named founding organisations. The top-level list is short. Its consequences are not.

Active A membership covers organisations that receive IP address space directly from LACNIC, organisations that receive it indirectly through a national registry under the relevant agreements, and organisations that received from ARIN space later distributed to the LACNIC region and apply for admission. LACNIC’s membership guidance translates that rule into current operational language: organisations receiving IPv4 or IPv6 directly from LACNIC or indirectly through NIC Brasil and NIC Mexico become Active A members.

The founding category is historical and named. The bylaws list six organisations associated with LACNIC’s formation: AHCIET, CABASE, CGI-Br, eCOMLAC, ENRED and NIC-Mx. Their status is not derived from a current address threshold. It follows institutional origin.

Adhering membership contains several routes. It can include regionally based organisations involved in Internet development or composed of service providers, organisations managing addresses outside the space distributed to the region while geographically located there, persons or institutions recognised by the Assembly for advancing LACNIC’s goals, and persons or legal entities making significant financial contributions.

Those routes mix function, geography, recognition and patronage. A regional Internet organisation and a significant financial supporter may enter the same broad category for different reasons. That is not necessarily defective, because the category is designed to create affiliation beyond address-based membership. But the common consequence is political: the adhering member does not receive the full franchise.

A further gate sits outside the category list. The current guidance states that organisations receiving only an autonomous system number do not become members. An ASN is technically consequential. It identifies an autonomous system in interdomain routing. Yet receipt of that resource alone does not cross LACNIC’s corporate threshold.

The resulting map is therefore not simply member versus non-member. It is address-qualified Active A member, historically privileged founding member, recognised but non-voting adhering member, or resource recipient without membership through the ASN-only route. Each status carries a different relationship to institutional power.

The first inequality is admission, not vote weight

Discussion of voting inequality often begins with the one-to-eleven scale. That is the visible numerical distinction, but it is not the first one. Before votes can be weighted, an organisation must enter a voting category.

Consider two small network operators. One receives a modest IPv4 or IPv6 allocation through LACNIC and becomes Active A. The other needs an ASN but obtains address space through an upstream provider. Both may run autonomous networks, maintain routing policy, respond to security incidents and depend on accurate registry information. Under the current membership guidance, only the first enters membership by virtue of its number-resource relationship.

The distinction has an administrative logic. Address allocation produces an ongoing registration relationship, fee schedule and stewardship obligation that differs from assigning an ASN. LACNIC can identify a stable class of organisations whose address holdings create continuing service and policy exposure. The line is therefore more grounded than a rule based on website traffic, employee count or public prominence.

Still, the line converts technical architecture into political eligibility. A network’s decision to use provider-assigned address space may reflect cost, scarcity, scale or business history rather than lower dependence on regional coordination. An ASN-only operator can be deeply engaged in routing and still lack the corporate rights attached to Active A membership.

Another operator may manage address space not distributed to the LACNIC region while operating geographically within it. The bylaws place such organisations within a possible adhering route. They can be recognised as part of the institutional community, but they do not receive the Active A vote merely because their operations are regional.

Admission therefore answers more than “who is a customer?” It answers whose resource history LACNIC treats as sufficient for governance. That rule should be stated plainly whenever the institution describes itself as member-led. The member community is not identical to the routing community, the policy community, the customer population or the population affected by regional Internet continuity.

A voice without a vote is participation, but not control

Adhering members have meaningful rights. They may use member services other than those relating to Internet resources, present initiatives to the Board and participate in the General Assembly with speaking privileges. This is not symbolic exclusion from every institutional venue. A well-prepared intervention can shape debate, frame evidence and influence voting members.

But speaking is not voting. Article 7 reserves the full combination of voting, speaking and nomination rights to Active A and founding members. The distinction becomes decisive when consensus does not emerge and the Assembly moves to a formal count.

Institutional language can blur this boundary. An Assembly may be described as open to members, and an adhering organisation may truthfully say it participates. A public report may list community engagement without separating those who could speak from those whose votes determined the result. The appearance of a broad membership body can therefore exceed the size of the actual electorate.

That difference should not be treated as an insult to deliberation. Governance needs more than ballots. Technical communities often resolve policy through discussion, and expertise does not correlate perfectly with formal voting power. An adhering member may persuade many Active A members. A non-member may make a decisive contribution in an open policy forum.

The constitutional question remains. When the Assembly approves financial statements, changes fees, amends bylaws or elects statutory bodies, persuasion is contingent while a vote is an enforceable unit of decision. A speaker depends on others to carry the conclusion. A voter can join the decisive denominator directly.

This is why “voice” and “power” should not be used interchangeably. LACNIC has built a channel for organisations that support its purposes without meeting the address-based admission rule. That is valuable. The same system has chosen not to make those organisations co-equal governors. An honest account of institutional legitimacy should acknowledge both facts.

The second inequality begins after admission

Active A membership opens the voting gate but does not equalise those who pass through it. Article 19 assigns between one and eleven votes according to address holdings. Founding members receive one vote. The scale uses specified IPv4 and IPv6 thresholds, increasing voting weight as the amount of managed address space rises.

The rule converts stewardship scale into corporate power. An organisation managing a small block enters at the lower end. Larger holdings move it through successive tiers. The largest qualifying category receives eleven votes. The system is therefore neither one organisation, one vote nor direct proportionality to every individual address. It is a stepped weighting model.

A stepped model moderates extremes. If votes tracked every address arithmetically, historic holders of very large IPv4 blocks could dominate by orders of magnitude. By capping weight at eleven and using tiers, the bylaws recognise scale without reproducing the full numerical disparity of address holdings.

Yet the cap does not remove inequality. Eleven votes can outweigh eleven one-vote organisations if all participate. A coalition of larger members may assemble a decisive share with fewer organisations. Conversely, a sufficiently large number of smaller members can defeat them. The practical balance depends on the distribution of members across tiers and on turnout.

That denominator is not visible in the reviewed public material. The bylaws state the scale, and current electoral guidance repeats that organisations receive one to eleven votes. But a reader cannot calculate concentration without knowing how many eligible organisations sit in each tier and how many cast their votes.

This missing information prevents two opposite claims. Critics cannot responsibly say that large address holders control LACNIC without concentration and outcome data. Defenders cannot responsibly say that the cap guarantees pluralism without the same evidence. The correct conclusion is that the constitutional design permits unequal weight and that its actual effect should be measured.

Why address holdings can be a defensible proxy

Weighted voting has a principled case. Regional registry members do not all have the same service relationship. An organisation managing more address space may serve more customers, maintain more registrations, face larger transfer and routing-security exposure, pay higher fees or carry wider operational responsibilities. Decisions about service continuity, fees and registration practice can impose different burdens across that range.

Address holdings also provide an observable criterion. LACNIC already maintains the relevant records. Unlike revenue, political status or self-reported user numbers, holdings can be categorised consistently within the registry’s own function. A rule tied to the service is less vulnerable to lobbying over subjective importance.

The one-to-eleven cap further suggests an effort to balance scale and association equality. The largest member does not receive thousands of times the vote of the smallest. Each legal organisation remains a recognised unit, while a bounded multiplier reflects operational size.

There is also a financing argument. Members whose fees contribute more to the institution may reasonably seek assurance that major financial decisions cannot be made entirely by a large number of minimally exposed entities. The 2024 Assembly’s fee decision affected address categories and future adjustments. A weighting mechanism can be understood as an attempt to match financial stake and political responsibility.

These arguments deserve more than dismissal as plutocracy. A one-member-one-vote system has its own distortions. A corporate group could fragment operations among many legal entities. Very small or dormant members could outvote organisations serving much of the region. Equal votes can conceal unequal exposure just as weighted votes can entrench incumbency.

The stronger critique is therefore not that weighting exists. It is that the proxy requires periodic proof. Address holdings may have correlated with scale when the rule was designed, but IPv4 exhaustion, transfers, IPv6 deployment, provider concentration and new network forms can alter that relationship. A legitimate proxy should be auditable and revisable rather than treated as naturally correct forever.

Resource history can become inherited political capital

Address holdings are not produced only by present need. They also reflect timing. Organisations that entered the Internet earlier often obtained larger IPv4 holdings under conditions different from those facing later entrants. Transfers and acquisitions can move holdings, but scarcity means the historical distribution remains consequential.

When voting weight follows address holdings, part of that history becomes political capital. An incumbent’s larger allocation can produce more votes even if a newer operator serves a comparable number of users through address-sharing techniques, provider arrangements or intensive IPv6 use. The political system may reward the form in which scale appears in the registry rather than scale itself.

IPv6 thresholds reduce the problem by recognising a newer resource family. The bylaws do not rely only on IPv4. But IPv4 and IPv6 blocks are not simple measures of customer population, market share, public value or network complexity. Allocation size can reflect policy, architecture and operational model.

This does not make the scale arbitrary. The registry governs number resources, so number holdings are closely related to its mandate. The issue is institutional inheritance. A current ballot can carry traces of allocation decisions made years earlier under different market and scarcity conditions.

The founding-member category is an even clearer form of historical recognition. Each founding member receives one vote because of its role in establishing the institution, not because of current address volume. Historical contribution is a legitimate value, particularly for an association whose formation required regional coalition building. But permanent constitutional recognition should be transparent about continuity, succession and inactivity. If a named organisation changes form, merges, ceases activity or no longer represents its original constituency, the bylaws need a clear answer about the vote.

Institutional memory can stabilise a young organisation. Over decades, it can also freeze an earlier map of regional authority. The answer is not to erase history, but to review how historical categories remain connected to living constituencies.

National-registry mediation does not remove Active A status

One of the design’s more inclusive features is its treatment of organisations in Brazil and Mexico. Receipt of address space indirectly through national registries operating under agreements with LACNIC can qualify an organisation for Active A membership. The regional franchise is therefore not limited to direct contractual service from the Montevideo office.

That choice recognises institutional diversity. The region did not develop one uniform service channel. National arrangements existed and continue to perform functions within the regional system. Excluding all indirectly served organisations would have turned administrative route into a severe political penalty for networks in major countries.

Including them also avoids a misleading equation between direct payment and governance legitimacy. An organisation can be part of the regional constituency even when an intermediary handles part of the service relationship. The relevant connection is receipt of qualifying regional address space through an accepted institutional chain.

The design nevertheless raises questions that deserve public measurement. Does the membership contact and voting process work equally well for directly and indirectly served organisations? Are notices delivered through the same channels and on the same timetable? Do national-registry customers understand their LACNIC rights? Are category and vote-tier calculations equally visible? Can they correct the voter register with comparable ease?

LACNIC’s 2026 statutory elections guidance says membership contacts registered with LACNIC, IAR MX or Registro.br perform electoral functions as applicable. That is evidence of an integrated design. It does not provide participation rates by route.

A formal right can be equal while its exercise is uneven. Language, communication path, platform familiarity and the perceived distance between a national service relationship and a regional association may affect turnout. Publishing route-specific participation in aggregated form would show whether indirect inclusion works in practice rather than only in the bylaws.

The ASN-only boundary exposes the difference between routing and membership

An autonomous system number is central to interdomain routing, but it is not address space. This technical distinction helps explain LACNIC’s membership boundary. The registry can assign an identifier to an autonomous system without giving that organisation an independent block of addresses.

RFC 7020 describes the Internet Numbers Registry System and carefully separates registry administration from operational routing. Registries maintain uniqueness and registration. They do not command every router. An ASN identifies an autonomous system, while route origination and acceptance depend on network operators.

Corporate membership is another distinct layer. The current LACNIC page states that an ASN-only recipient does not become a member. That rule should not be misunderstood as saying the operator is technically irrelevant. It says the particular registry relationship is limited public evidence for corporate admission.

The boundary reveals the limits of “stakeholder” language. An ASN-only operator is a stakeholder in number coordination and routing. It may participate in policy discussion, attend community events and depend on service continuity. Yet it does not possess the Active A rights attached to address receipt.

There are defensible reasons. ASN assignment may involve a lower ongoing service burden. Some ASN holders use upstream address space and may be represented indirectly through provider members. Automatic membership for every identifier holder could expand the electorate in ways unrelated to financial or registration exposure.

There are also risks. Provider representation is not the same as customer representation. An upstream and its customer can disagree on fees, competition, transfers or policy. Small networks, Internet exchange entities and specialised operators may have interests not captured by address-holding organisations.

The proper response is not necessarily automatic voting membership. LACNIC could retain the boundary while publishing how many ASN-only organisations exist, how they participate, whether they can petition institutional bodies and how their concerns reach voting members. A boundary is more legitimate when the excluded constituency has a visible channel and the rationale is periodically tested.

Organisational form matters even when technical dependence is similar

The adhering category demonstrates that institutional form can outweigh operational similarity. A company, association or institution may be deeply involved in Internet development, may consist of providers, or may operate regionally with address space from outside the LACNIC pool. It can join the association, present initiatives and speak, but it does not receive the full vote.

Another company receiving qualifying regional address space enters Active A and can vote. The technical difference may be clear in the registry. The political difference is broader: nomination, extraordinary-assembly powers and weighted votes follow.

Legal form can also shape how one underlying economic group appears. A telecommunications group may contain multiple subsidiaries, licences and resource-holding companies. A public university system may centralise addresses in one entity while campuses operate separately. A community-network federation may coordinate many small networks through one association. The unit counted by the bylaws is the admitted member organisation, not every network, customer or constituent behind it.

This creates familiar representation questions. Multiple legal entities in one group can potentially create more organisational entries, although address-based vote tiers and admission rules may constrain the benefit. A federation with many constituents may have only one vote if it qualifies, while a corporate group may have several member entities. Without beneficial-control and affiliation data, outsiders cannot assess whether organisation counts equal independent interests.

Privacy and commercial sensitivity limit publication. LACNIC should not expose confidential ownership files merely to satisfy curiosity. It can still report aggregate concentration, common-control rules, treatment of mergers and standards for deciding when related entities vote separately.

The core principle is that legal personality is an administrative unit, not a natural measure of public constituency. It is useful because associations need named holders of rights and duties. Its political effect should be acknowledged rather than hidden behind the neutral language of an organisation list.

The Assembly exercises more than ceremonial authority

Category rules would matter less if the Assembly merely received presentations. The bylaws give it substantive functions, including consideration of institutional reports, financial matters and constitutional questions. Members also elect the bodies that manage, supervise and administer electoral processes.

The minutes of the 2024 Ordinary Member Assembly show this authority in operation. The annual report, balance sheet, inventory, income and expense statement, external auditor’s report and Fiscal Commission report were approved with 654 votes for, none against, 11 abstentions and 47 votes not cast. A proposal to adjust membership fees for inflation received 404 votes for, 208 against, 41 abstentions and 60 not cast.

These are weighted votes, not necessarily counts of organisations. That distinction is essential. The fee proposal’s 404-to-208 result does not mean exactly 404 organisations supported it and 208 opposed it. Each eligible member could carry a different number of votes. The published totals show decision weight, not the number or diversity of institutional supporters.

The minutes therefore prove both institutional activity and a transparency limit. A reader can see the outcome and the volume of voting weight. The reader cannot see how many organisations voted, how votes were distributed across tiers, whether small and large members diverged, or whether a few high-tier organisations supplied a substantial share.

Publishing both weighted and unweighted totals would greatly improve interpretation. The legally decisive result would remain the weighted count. An additional count of voting organisations, grouped at a safe level, would show whether the decision had broad organisational support or a narrower coalition with greater assigned weight.

This is especially useful for fee decisions, where contribution and vote weight may correlate. A proposal can be valid under the bylaws while still raising a political question about who paid, who voted and who bore the change. Better denominators do not invalidate the decision; they let members understand it.

Good standing is a separate switch on political rights

Category and vote tier are not the only determinants of participation. Current election guidance requires associated organisations to be up to date with annual membership payments for nomination and supporting functions, and it states that organisations in good standing at the close of the voter register can vote.

The bylaws provide that members behind on fees or established contributions are automatically suspended and may lose status if non-payment continues beyond applicable deadlines. Payment therefore acts as a switch on governance rights.

This is normal for membership associations. A member that does not meet shared financial duties cannot necessarily demand the same standing as one that does. The rule protects the institution from organisations that retain political power while refusing the obligations that sustain it.

Yet fee-linked suspension can have distributive effects. Economic distress may silence precisely the smaller operators most affected by fee changes. Currency volatility, cross-border payment barriers and administrative errors can turn a financial requirement into political exclusion. If a fee proposal is contested, the members least able to pay may disappear from the electorate that judges it.

The answer is not to detach all rights from payment. It is to publish safeguards. LACNIC should disclose the voter-register cutoff, notice process, correction window, number of organisations excluded for non-payment, reinstatement process and whether a payment dispute differs from undisputed default.

The bylaws also provide procedural protection for penalties. The affected party may present a defense, and an appeal to the next Member Assembly within thirty days has suspensive effect. That is a meaningful check against immediate loss of rights through disciplinary action.

Financial suspension deserves comparable clarity because it can occur automatically. Automatic rules are efficient, but they can make administrative error politically consequential. A rapid correction channel should operate before the voter register closes, and aggregate data should show whether exclusions are rare or systemic.

Nomination rights shape the ballot before voters see it

Voting is only the final stage of electoral power. Active A and founding members can nominate candidates to statutory bodies. The 2026 guidance adds practical requirements: a nomination comes through the organisation’s membership contact, an organisation may present one candidate, and each nomination needs support from two additional associated organisations in good standing.

These rules create a pre-ballot coalition threshold. A person can meet substantive eligibility standards yet never reach the slate without three organisational commitments: one nominator and two supporters. The threshold can filter frivolous candidacies and demonstrate minimum community confidence.

It also makes category boundaries relevant before voting begins. Adhering members may have expertise and speaking rights, but the bylaws do not give them the same nomination franchise. ASN-only organisations do not enter through membership. Their preferred candidates need support from organisations inside the active electorate.

The nomination stage can therefore reproduce established networks. Large or well-connected members can secure endorsements more easily. Smaller members may need time and cross-border relationships. A candidate independent of major institutions may face a coordination burden before voters can assess the merits.

LACNIC’s public procedures seek to improve legitimacy through eligibility checks, conflict rules and an open-source voting system. Those are significant protections. The institution should complement them with nomination-denominator reporting: number of expressions of interest, completed nominations, rejected nominations, reasons for rejection, supporting organisations by broad category and candidates who withdrew.

Such reporting need not reveal private deliberations. It would show whether the nomination gate regularly narrows a diverse field or merely verifies credible support. It would also help distinguish low candidate supply from restrictive screening.

Political power includes the capacity to put an issue or person before the electorate. A category audit that counts only ballots misses this earlier authority.

Extraordinary meetings reveal agenda-setting power

The bylaws allow Active A and founding members to request an Extraordinary Member Assembly under Article 14. The mechanism is not merely a right to complain to the Board. If the relevant support threshold is met and the Board refuses, the request can proceed through the Fiscal Commission under stated conditions.

This is a valuable accountability design. It prevents the Board from holding an absolute monopoly over the Assembly’s agenda. Members can escalate an issue and use another statutory body when management declines.

The power is category-bound. Adhering members may present initiatives to the Board and speak at an Assembly, but they do not hold the same route to compel extraordinary consideration. An ASN-only operator has no membership-based claim to use it. The difference between a suggestion and a procedurally enforceable agenda right is substantial.

Threshold design also interacts with vote weighting in an interesting way. The support needed to demand a meeting is described in relation to voting members, while decisions at the resulting meeting use assigned votes. An organisation’s significance can therefore be measured once as a member for triggering deliberation and again by weighted votes for deciding the issue, depending on the precise provision.

This separation can be healthy. A number of organisations must show concern before the institution incurs the cost of an extraordinary meeting, while the established voting scale governs the final resolution. It can also be confusing if members do not know whether a threshold counts organisations, votes or both.

Every constitutional threshold should state its denominator in plain language. LACNIC should publish worked examples and historical use: requests made, support obtained, Board responses, Fiscal Commission referrals, time to meeting and outcome. A right that has never been tested may be robust, or it may contain hidden procedural friction. Evidence distinguishes the two.

Recognition in 2002 did not settle every later constituency question

LACNIC’s recognition announcement records final recognition as the fourth Regional Internet Registry on 31 October 2002 after provisional approval earlier that year. Recognition confirmed that the region had built an institution capable of assuming registry functions through a transition from ARIN.

The ICP-2 criteria used for new RIR recognition emphasise broad support, bottom-up governance, neutrality, documented procedures, competence and financial stability. LACNIC’s membership design helped demonstrate regional organisation and sustainable institutional form.

Recognition should not be read as permanent approval of every internal rule. It establishes the regional registry’s status in the wider system. It does not answer whether a voting scale remains representative after two decades of market, resource and technology change.

This distinction protects both stability and accountability. If every disagreement about member categories threatened RIR recognition, internal reform would become dangerously escalatory. Corporate rules should ordinarily be reviewed through the institution’s own constitutional mechanisms. At the same time, recognition history should not be used to close debate by implying that initial community support validates all later distributions of power.

LACNIC’s own history demonstrates adaptation. The bylaws record amendments to membership and voting provisions across multiple Assemblies. The design is not frozen in 2002. Members have revised categories, procedures and institutional bodies as circumstances changed.

That adaptive capacity is an argument for a fresh category audit. The question is not whether LACNIC was legitimately established. It is whether the current conversion from resource relation to political standing remains understandable, evidence-based and open to correction.

Policy participation does not substitute for corporate accountability

Regional Internet governance uses more than one constituency. Open policy development invites technical and community participation beyond the corporate electorate. A person or organisation can influence number policy without holding a statutory vote. This openness is an essential counterweight to narrow membership rules.

But policy participation and corporate governance decide different things. An open policy forum may debate criteria for allocation or transfer. The Assembly approves accounts, fees, bylaws and statutory bodies. A community entity cannot assume that success in one venue confers authority in the other.

Institutions sometimes invoke openness in one process to answer criticism of exclusion in another. That response is incomplete. A non-member’s ability to comment on technical policy does not let it vote on the Board that oversees finances and executive leadership. An adhering member’s Assembly speech does not let it determine the resolution.

The reverse is also true. An Active A member’s weighted statutory vote should not let it dominate an open policy process merely because it holds corporate power. Policy legitimacy often rests on rough consensus, technical evidence and broad participation rather than the Assembly’s vote scale.

Clear separation prevents authority from leaking between venues. LACNIC should publish a simple map of which body decides which question, who can participate, who can vote, how appeals work and how one process refers matters to another.

This map would help the wider public understand that “bottom-up” is not one mechanism. It is a family of mechanisms with different constituencies. The legitimacy of the whole institution depends on each being fit for its purpose and on no narrow constituency claiming to represent everyone in every domain.

The annual report can show activity but not political concentration

The LACNIC Annual Report 2024 provides institutional context: services, regional initiatives, community engagement and the organisation’s continuing work. Annual reports are valuable because they connect governance to the operational mission that members fund.

They are not a substitute for a franchise audit. Activity counts can show how many people attended events, how many resources were administered or how many projects advanced. Those figures do not reveal how statutory power is distributed.

A high level of community participation can coexist with concentrated voting weight. A broad training programme can coexist with a narrow nomination network. Strong service performance can coexist with unclear category decisions. Conversely, unequal votes do not prove poor service or captured governance.

The institution should therefore report operational and political denominators separately. A governance annex could show, for each year, the number of Active A, founding and adhering members; Active A distribution across vote tiers; organisations eligible at voter-register close; organisations that voted; weighted votes cast; nominations; endorsements; category changes; suspensions; appeals; and common-control treatment.

Time series matter more than a single snapshot. They would show whether the electorate is broadening, shrinking or concentrating. A sudden shift might reflect address transfers, mergers, changes in payment, new IPv6 members or revised rules. Members could investigate before the change becomes entrenched.

Publication should protect confidential data. Aggregated tiers and country groupings can avoid identifying small classes where necessary. The purpose is not to expose individual voting choices. It is to reveal the architecture of political capacity.

Governance reporting is itself an accountability service. The institution already knows the formal categories. Turning them into understandable denominators would let members debate evidence rather than impressions.

A practical rights matrix

The category system can be summarised without collapsing its details:

Position	Entry basis	Assembly voice	Statutory vote	Candidate nomination	Vote weight
Active A	Qualifying regional IP address receipt through specified direct, national-registry or legacy route	Yes	Yes, subject to standing and procedure	Yes	One to eleven according to address holdings
Active Founding	Named historical founding organisation	Yes	Yes, subject to standing and procedure	Yes	One
Adhering	Regional contribution, non-regional address management in the region, Assembly recognition or significant financial support	Yes	No	No equivalent full franchise under Article 7	None
ASN-only recipient	Receipt of an ASN without qualifying IP address membership basis	Community channels may be available, but not by membership right	No membership vote	No membership nomination right	None

The matrix is necessarily simplified. Good standing, membership contacts, election rules, disciplinary decisions and appeals affect exercise. Open policy processes create other forms of participation. A legal entity may qualify through facts not visible to outsiders.

Even so, the matrix reveals the thesis. Technical dependence is not the unit of political equality. Address-based membership is the primary gate, history creates a separate founding route, affiliation creates a speaking category, and address scale weights the admitted Active A vote.

This structure may be defensible if each line has a clear purpose. It becomes harder to defend when the denominators are missing, classifications are opaque or the proxy no longer tracks the interests it was meant to represent.

The missing denominator is the central evidence gap

How many Active A members have one vote? How many have eleven? What share of total eligible weight sits in the top three tiers? How many adhering members speak regularly? How many ASN-only organisations participate through other channels? How many organisations were removed from a voter register for payment or contact issues?

The reviewed official pages do not provide a complete current answer. This absence is not evidence of capture. It is an inability to test competing claims.

Suppose 90 percent of Active A organisations hold one or two votes but rarely participate, while a smaller group of high-tier members votes consistently. Formal breadth would coexist with practical concentration. Suppose instead that small and medium members cast most of the eligible weight and high-tier members are diverse and divided. The same bylaws would produce a more plural result.

Turnout should be reported in two ways: organisations and weighted votes. A 60 percent turnout by vote weight may conceal participation by only a small share of organisations. A high organisational turnout may still represent a lower share of weight. Both measures are valid and answer different questions.

Category transitions matter too. When an organisation receives more space, merges, transfers resources or changes service route, does its vote tier change automatically? At what record date? Can it challenge the calculation before an election? Historical transition data would show how fluid political weight is.

LACNIC’s election page describes a voter register and correction process. That procedural foundation can support stronger transparency. Aggregate register statistics should be published when the provisional register opens, after corrections close and with final results.

The denominator is not clerical detail. It is the evidence needed to understand the phrase “members decided.”

Reform should test proxies rather than promise abstract equality

A simplistic reform would replace weighted voting with one organisation, one vote. That change would be easy to explain but could create new distortions. It could reward corporate fragmentation, discount operational scale and sever financial exposure from decision weight.

Another simplistic reform would extend voting membership to every ASN holder, policy entity or affected organisation. That could produce an electorate too loosely connected to corporate duties, increase verification costs and invite strategic enrolment around a contested vote.

A better process begins with objectives. Is the franchise meant to represent service exposure, financial contribution, regional networks, address stewardship, community legitimacy or some combination? The bylaws currently combine these values through different categories, but the relative purpose is not always explicit.

Once the objective is stated, LACNIC can test proxies. Address holdings can be compared with fees, customer scale, organisational independence and participation. The institution can model alternative caps, tier boundaries and one-vote floors. It can examine whether national-registry customers and direct members participate similarly.

Reform should also consider chambers or dual tests. A major constitutional change might require both a weighted majority and support from a majority of voting organisations. That would preserve scale recognition while preventing a small number of heavily weighted members from acting alone. Conversely, it would prevent a large number of minimal-stake entities from disregarding most service exposure.

Any dual rule adds complexity and potential deadlock. It should be reserved for high-consequence decisions, not routine administration. The point is to design around stated risks rather than declare one arithmetic rule universally democratic.

The process itself must include adherents, ASN-only operators, non-member community entities and smaller members. Voting members may retain formal authority to amend the bylaws, but those outside the current franchise should be heard when the franchise is the subject.

Minimum reforms for category legitimacy

First, publish a yearly membership census by top-level category and Active A vote tier. Include eligible organisations, not only total assigned vote weight. Preserve confidentiality where a small cell could identify a member.

Second, report every statutory ballot with both weighted results and the number of organisations voting each way, plus abstentions and non-participation. The legal result remains weighted; the unweighted view supplies political context.

Third, publish category decisions in aggregate. Show applications, approvals, refusals, transitions, suspensions, terminations, corrections and appeals, with reasons grouped by rule. This would reveal whether boundaries are stable and consistently administered.

Fourth, explain the ASN-only rule as a governance choice, not a footnote. State what channels these operators can use, how many are affected and whether the Board periodically reviews their representation.

Fifth, disclose the treatment of affiliated organisations and changes in legal form. Members need confidence that mergers, subsidiaries and restructuring do not produce arbitrary multiplication or loss of votes.

Sixth, audit indirect participation through NIC Brasil and NIC Mexico. Compare delivery of notices, voter registration, turnout and correction experience with direct members.

Seventh, test the one-to-eleven scale every three to five years. Publish models showing how current holdings distribute power and whether tier boundaries still reflect meaningful differences.

Eighth, require an equality-impact statement for bylaw amendments affecting admission, standing, nomination or vote weight. The statement should identify which categories gain or lose rights and why.

Ninth, provide a rapid, independent review route for voter-register and category disputes before ballots close. A later appeal may not repair exclusion from a completed election.

Tenth, publish a plain-language constitution map linking category, right, executor, deadline and review. Members should not need to reconstruct political standing from several documents.

None of these reforms predetermines equal voting. They make the existing choice testable and correctable.

What the evidence does not prove

The public record does not prove that large members dominate LACNIC. The one-to-eleven scale permits unequal influence, but actual control depends on the distribution of members, turnout, coalitions, issue preferences and constitutional thresholds.

It does not prove that adhering members are powerless. Speaking, initiative and community participation can affect outcomes. Their lack of a vote means they cannot decide directly, not that they have no influence.

It does not prove that ASN-only organisations require corporate membership. Their service relationship differs from address holders, and alternative representation may be sufficient if it works.

It does not prove that address weighting is unlawful or inconsistent with association governance. This analysis does not interpret Uruguayan law, and the English bylaws are a translation whose Spanish original prevails where discrepancies exist.

It does not establish current category counts or concentration. Those are precisely the missing denominators. Nor does it infer the status of any named founding organisation beyond what the bylaws list.

Finally, it does not equate registry governance with routing control. LACNIC administers number records and services within a coordinated system. Networks make operational routing decisions. Political rights inside the association matter because they shape the institution, not because an Assembly ballot directly configures the region’s routers.

These limits strengthen the argument for disclosure. Claims of capture and claims of broad equality are both premature without evidence. A transparent institution should make neither necessary.

Organisational form is political architecture

LACNIC’s membership categories perform real work. They connect a large and diverse technical region to a legal association in Uruguay. They recognise founders, include nationally mediated address recipients, create affiliation for wider contributors and attach voting weight to service scale. The result has supported a functioning regional registry since 2002.

The same categories distribute power. An address recipient can vote; an ASN-only recipient does not become a member through that relationship. An adhering organisation can speak but not decide. A small Active A member receives one vote; a large one can receive eleven. A founder receives a vote through historical status.

None of these distinctions follows inevitably from technical dependence. They are constitutional choices about which relationships count and how much. That is why organisational form carries weight. It moves an operator from the wider community into a speaking category, from a speaking category into the electorate, and from one unit of voting power toward eleven.

The central legitimacy test is not whether every affected party receives an identical vote. Regional registry governance must remain workable, resistant to strategic enrolment and connected to those who carry duties. The test is whether each inequality has a stated purpose, current evidence, a correction route and enough public data for the region to judge its effect.

LACNIC already publishes the rules and records Assembly decisions. The next step is to publish the political denominators those rules create. Category counts, vote-tier distribution, weighted and organisational turnout, classification outcomes and affiliation safeguards would turn an abstract member-led claim into an auditable account of power.

Technical coordination often appears neutral because addresses and autonomous system numbers are expressed as numbers. Governance begins when institutions decide which number relationship creates a member, which member may vote and how many votes that member carries. Those decisions should be treated with the same care as any other architecture on which the region depends.