| Field | Value |
|---|---|
| Author | RIR Watchdog |
| Published | 2026-07-01 |
| Primary category | arin |
| Categories | governance; rir-watchdog; arin |
| SEO title | ARIN and the economics of consensus capture |
| SEO description | A research analysis of how ARIN's open consensus process can still be captured by procedural insiders in the North American IPv4 transfer-market era. |
| Focus keyword | ARIN consensus capture |
| Primary domain | Governance |
| Content type | Research |
| Topic | consensus capture |
| Subject | ARIN |
| Region | North America |
| Time horizon | 12-24 months |
| Impact | HIGH |
| Confidence | B / 0.88 |
The market hidden inside the meeting
ARIN is usually described in the language of Internet self-governance: open participation, public discussion, community policy, elected representatives, transparent minutes and technical stewardship. Those claims are not empty. The American Registry for Internet Numbers publishes policy text, board materials, meeting records, fee schedules, transfer requirements and regional service information. Anyone can subscribe to the Public Policy Mailing List. Public Policy and Members Meetings are free to attend, with remote participation and archived transcripts.
The harder question is whether the open door creates effective contestability. A process can be public and still be captured if the price of serious participation is high, if the people with the strongest financial exposure attend most persistently, if procedural gatekeepers decide which problems are admissible, and if support is measured among the active minority rather than the affected majority. That is the central risk in ARIN's post-exhaustion environment. It is not that ARIN is formally closed. It is that openness may be doing more work in the institution's claim to legitimacy than it can bear.
The issue matters because ARIN sits at a peculiar junction. It is a private non-profit corporation, a regional Internet-number registry, a participant in global identifier coordination, and the registrar for scarce IPv4 resources in North America and parts of the Caribbean and North Atlantic. It does not own the Internet. It does not route packets. It cannot force operators to accept a route. Yet it maintains the registry records that turn a block of numbers into a recognized operational resource. Those records support reverse DNS, routing-security practice, transfer recognition, contractual standing and market confidence. After IPv4 exhaustion, that function became more economically important, not less.
When a large free pool still existed, number policy disputes were mainly about conservation, documented need, aggregation, routability and fair distribution. Scarcity was present, but it was moderated by a continuing supply of new addresses. After exhaustion, the same policy machinery increasingly decides the terms under which a fixed stock may move between historic holders, new entrants, consolidators, brokers and cross-regional buyers. A registry that once rationed access to a public pool now validates movements in a semi-private market.
Consensus capture in such a setting rarely requires conspiracy. It is more often a product of accumulated advantage. Repeat participants know when a proposal must contain specific Number Resource Policy Manual language. They know when an issue is treated as number policy and when it is diverted to a consultation or suggestion channel. They know the culture of the mailing list, the rhythm of public meetings, the expectations of staff and legal review, and the threshold needed to defeat, delay or narrow a change. They also know that most organizations affected by number policy do not have staff available to follow a months-long textual debate, appear at meetings, answer procedural objections and defend a proposal through successive revisions.
This produces a distinctive political economy. A small access provider may care deeply about the cost and availability of IPv4 addresses, but the provider may not have a policy specialist who can track ARIN proceedings over several meeting cycles. A cloud platform, broker, large network, legacy holder or long-serving community participant is more likely to have that capacity. The gap is not merely one of interest. It is a gap in procedural capital. Like financial capital, procedural capital compounds: the more often an actor participates, the better it understands which arguments count, which forms of evidence are persuasive and which institutional paths are usable.
The result is an institution that can be open and still vulnerable to capture. Outsiders are not told they may not participate. They merely fail to attend at the right moment, speak in the expected register, provide the right form of text, sustain attention across successive stages, gather support inside a narrow window, or understand where a market concern belongs in a process divided between policy, services, fees and corporate oversight. In infrastructure governance, these distinctions are decisive. Formal access is cheap. Effective voice is expensive.
The thesis of this article is that ARIN's consensus system is exposed to procedural capture because it governs scarce economic recognition through a process built around active community participation. It claims stewardship principles, yet it now operates in a world where transfer rules, waitlists, fee schedules, legacy-resource certainty and member elections shape incentives as much as engineering norms do. It is accountable to members and the wider Internet community, but voting status, meeting participation and policy competence require organizational attention that is unevenly distributed. Its decisive legitimacy risk is regional: whether networks, entrants, resource holders and smaller jurisdictions continue to believe that the registry is neutral enough to use, not merely documented enough to inspect.
Public ARIN materials provide enough evidence to examine the risk without accepting the institution's own narrative as the conclusion. The Policy Development Process says policy changes must be open and transparent, that the public policy list is archived, and that meetings include polling of in-person and remote participants. It also says support need not be unanimous and may be demonstrated by a subset of the community if support substantially exceeds opposition within the active part of the community. Membership materials say that membership is not required to discuss policy, but that only General Members in good standing vote in elections and that General Members that do not vote in any of the previous three annual elections revert to Service Member status. Transfer rules require recipients to meet operational-need criteria and, in many cases, to sign registration service agreements. The 2026 fee schedule sets annual charges by aggregate holdings and imposes processing fees on transfers. None of those facts proves capture. Together they show how capture can occur while every step remains visible.
Scarcity changed the institution
IPv4 exhaustion changed the economic meaning of ARIN's rules. The Internet's original IPv4 address space is finite. Once unallocated IPv4 space was exhausted at the global level and then across regional registries, a question that had once looked mainly administrative became an allocation problem with asset-market effects. ARIN's public materials describe a service region covering Canada, the United States and many Caribbean and North Atlantic islands. Its Number Resource Policy Manual still uses the classic vocabulary of Internet-number stewardship: registration, conservation, routability and stewardship. These principles were coherent when address assignment operated as rationing from a shared pool. They are less complete if treated as self-executing answers after the pool has become a market.
The pre-exhaustion bargain was legible. Applicants with operational need could obtain resources by satisfying policy requirements and paying registry fees. The registry was not meant to confer private property; it was meant to record who held which resources, prevent duplicate use, support routing and maintain operational contact data. That bargain was imperfect, but its distributive logic was understandable. The price of a new allocation was not the market price of the addresses. It was the cost of documentation, compliance and ordinary registry service. A new network could view the registry as a gateway to needed inputs rather than as a certifier of scarcity rents.
Post-exhaustion, the old vocabulary operates in a different economy. New entrants cannot rely on ordinary allocations from a deep free pool. ARIN's waitlist policy says that future IPv4 assignments or allocations outside certain reserved pools come from the waitlist, that the maximum size for which an organization may qualify at one time is a /22, and that organizations holding more than a /20 equivalent of IPv4 space are not eligible to apply. It also says that waitlist-distributed space is not eligible for transfer, except in merger and acquisition situations, for 60 months. In practice, the waitlist is a fairness mechanism for small quantities, not a substitute for a functioning address supply. It matters symbolically and operationally, but the economic center of gravity has moved to transfers.
Transfers are the bridge between registry policy and market value. ARIN's transfer rules distinguish in-region specified-recipient transfers under section 8.3 from inter-regional transfers under section 8.4, where reciprocal and compatible needs-based rules are required. For in-region transfers, the source and recipient submit requests, the source must be the current registered holder and free of dispute over the resource, the minimum transfer size is a /24, and the recipient must meet transfer requirements. Inter-regional transfers involving ARIN exclude IPv6 and require compatibility with the other regional registry's rules. Recipient organizations coming into the ARIN region must demonstrate need for up to a 24-month supply. These requirements are not administrative trivia. They define who can turn money into registry-recognized IPv4 capacity, how quickly, with what documentation burden and under what uncertainty.
Policy design therefore distributes economic advantage even when written in neutral technical language. If transfer rules are strict, incumbent holders may enjoy scarcity premiums because supply is harder to mobilize and new entrants face higher compliance costs. If transfer rules are loose, address blocks may move more readily to those with capital, potentially weakening conservation and the historic claim that number resources are not ordinary property. If inter-regional transfers remain constrained by reciprocal needs-based policies, regional markets are partly segmented. If those constraints are relaxed, North American legacy space can be exposed more directly to global demand. Each position has plausible arguments. The point is that each position also changes bargaining power.
Scarcity also changes the value of delay. In an abundant environment, procedural delay may be irritating but survivable. In a post-exhaustion market, delay can protect incumbents. A firm already holding addresses can continue to operate while a rival waits for pre-approval, gathers documentation or searches for a seller. A broker with deep knowledge of transfer requirements can monetize complexity. A large buyer can absorb legal review, escrow arrangements and registry interaction as part of acquisition cost. A small network may find the same complexity prohibitive. The registry does not need to intend such outcomes. Once scarcity exists, every procedural cost becomes a form of price discrimination.
For that reason, ARIN's consensus system should be analyzed as an institution in a market, not merely as a technical club. The relevant question is not whether ARIN says participation is open. The question is whether the structure of voice, voting, review, petition and adoption lets affected outsiders change rules when insiders benefit from continuity. Institutional economics treats rules as allocations of bargaining power. In ARIN's case, the power lies in policy-text control, interpretation of community support, staff and legal assessment, board adoption, member elections and the accumulated knowledge required to move through all of those stages. IPv4 exhaustion made that power more valuable.
PDP design and the active minority
The most important feature of ARIN's Policy Development Process is not simply that it is open. It is the way openness is transformed into usable authority. A proposal must present a clear problem statement and suggested changes to specific Number Resource Policy Manual text. The process does not allow a policy proposal to define the exact processes by which ARIN staff will implement policy, nor to establish services or fees. Those issues belong to consultation, suggestions, corporate decision-making or operational practice. This division is administratively rational. It also fragments reform.
Many post-exhaustion problems do not fit neatly into a single box. IPv4 transfer friction may arise from policy text, staff review practice, legal documentation, processing fees, member expectations, waitlist rules, fraud risk and the professional role of facilitators. A reformer who says that the transfer market is unfair will be asked to translate the complaint into manual text, avoid service design, avoid fee-setting, respect legal constraints and show active community support. Insiders know how to divide the problem and where to file each part. Outsiders experience the division as a maze.
The requirement for specific policy text favors people who can draft in a legalistic and operational style. A participant may understand the economic harm of a rule without knowing how to write acceptable language for the manual. Policy Shepherds and staff can help improve proposals, and that support has real value. Yet assistance can also narrow a proposal before it reaches the widest discussion. A challenge to the economic settlement may be recast as a technical adjustment. A broad contest over market access may become a small clarification. The process rewards proposals that can be made tidy.
The Advisory Council's role is central. It helps implement the process, facilitates communication, deliberates over proposals, works through shepherds and recommends policy text to the Board. At an early stage the council is not supposed to judge a proposal on its merits; it evaluates criteria such as clarity, suggested text and scope. But clarity and scope are not neutral concepts. A proposal that challenges the economic effects of transfer requirements can be said to concern fees, services, legal risk or business practice rather than number policy. A proposal that seeks data on market concentration can be said to lack manual text. A proposal aimed at participation inequality can be diverted toward meeting, membership or consultation channels. Each decision may be defensible in isolation. A pattern of such decisions can protect the settlement already embedded in the rules.
The consensus standard deepens the problem. ARIN's process recognizes that support need not be unanimous and that it may be shown by a subset of the community if support substantially exceeds opposition among active participants. The rule is practical; otherwise one determined objector could block useful change. But it also converts the active minority into the population from which legitimacy is inferred. The active community is not a neutral sample of the affected region. It is the group that can afford to follow the process, speak in the accepted style and return repeatedly.
Consensus in this setting is not the same as consent. It is evidence of visible convergence among participants. That can be sufficient for many technical questions. It is weaker for questions with large distributional effects. Transfer policy, legacy treatment, fee design, waitlist constraints and accountability rules affect organizations that may never appear on a mailing list. A small rural provider, a Caribbean operator, a public-interest network, a new hosting company or an enterprise buying addresses for the first time may experience the consequences of consensus without ever joining the active minority that formed it.
Later stages add further filters. Last Call, staff and legal review, board consideration and implementation review are necessary safeguards. A registry needs policies to be lawful, clear and implementable. But those stages also make successful change non-self-executing. Even after community support appears, an institutional actor can remand, narrow, delay or reject a change in the name of compliance, risk or feasibility. Legal caution is not illegitimate. Nor is staff expertise. The capture risk lies in asymmetry: change must justify itself repeatedly, while the status quo remains the operating baseline.
Capture often appears as a preference for incrementalism. Large changes need more work. Economic critiques are outside scope. Distributional language is too political. Participation reform belongs elsewhere. Legacy questions are legally delicate. Transfer-market data are private. The result is not a ban on dissent. It is the conversion of dissent into homework. The burden of proof falls on challengers, while existing rules enjoy a presumption of operational maturity.
The answer is not to weaken procedure. A registry without procedure would be arbitrary. The answer is to recognize procedure as a site of power. Every requirement for format, scope, timing, support, review and legal assessment is a barrier that may be justified and still unequal in its effect. A healthy consensus system audits those barriers against actual participation. It asks who uses the process, who gives up, whose proposals are remanded, whose concerns are diverted and whether economic losers are visible. A captured system points to open archives and says the opportunity existed.
Mailing-list dynamics and the cost of seriousness
ARIN's mailing lists are both a strength and a weakness. The Public Policy Mailing List is open to the general public and is an intrinsic part of the policy process. Its archives are public. Other lists cover announcements, consultations, technical matters, issued-resource matters and General Member discussions. This record is more accessible than the documentary trail of many infrastructure institutions. Yet a public archive does not automatically broaden participation. Mailing lists reward a specific type of actor: literate in the history of past debates, comfortable with adversarial technical prose, available to reply quickly and willing to endure repetition.
The cost of being taken seriously on such a list is high. A participant must know the relevant manual sections, the history of previous proposals, staff interpretations, the distinction between policy and service issues and the objections that familiar participants are likely to raise. They must write clearly enough to avoid being dismissed as confused, but not so sharply that they violate community norms. They must respond to technical and procedural challenges, often from people who have debated similar questions for years. They must continue after the first round of silence, correction or fatigue. They must also accept that a lack of broad response may be read as a lack of support, rather than as a sign that the affected majority is not listening.
This favors insiders without requiring censorship. The problem is attrition. A new entrant may post a concern about IPv4 transfer costs and discover that the thread shifts quickly to whether the concern belongs in policy, whether the proposed text is too broad, whether the market premise is supported, whether staff practice already addresses the issue, whether legal risk makes the reform impractical and whether routing scalability is implicated. Each point may be reasonable. Together they form a maze. The experienced participant sees the maze as normal. The newcomer sees an institution that is hard to use.
Mailing-list culture also amplifies those who treat number governance as part of their professional identity. There is nothing inherently wrong with professional participation. Internet number policy is complex, and casual rule-making can do real harm. The danger is that professionalism becomes a gatekeeping standard. If only participants fluent in registry idiom can shape outcomes, the active community becomes self-reproducing. It develops a shared sense of what is practical, what is naive, what has already been settled and what kind of evidence counts. Consensus then reflects not only preference but culture.
Meetings add a second layer. Public Policy and Members Meetings are held twice a year, registration is free, remote participation is supported, and ARIN archives presentations, transcripts and videos. Those practices have value. Yet free registration is not free participation. Travel, time zones, workload, preparation and social capital all matter. A small operator may not pay a fee to register but still cannot afford to send staff to a multi-day event, prepare comments, monitor sessions and sustain follow-up. Remote access lowers the cost, but it does not erase the attention burden or the advantage of people who are physically present and socially known.
Repeated meetings also create informal networks that the public record only partly captures. People who attend year after year know which concerns will persuade the Advisory Council, which objections are decisive, which speakers carry influence and which compromises may survive staff and legal review. They build trust. Trust is useful for technical coordination. It can also harden into insider status. An argument made by a familiar participant may receive charitable interpretation. The same argument from a newcomer may be treated as underdeveloped. This is ordinary human behavior, but in a scarce-resource market it becomes institutionally consequential.
The result is a paradox of visibility. Everything is visible, but the social cost of becoming effective is hidden. ARIN can point to public archives, open subscription and free meetings. Critics can point to the persistence of a narrow active core. Both observations can be true. The institutional question is whether the benefits of specialized deliberation outweigh the costs of underrepresentation. In a purely technical standards forum, the answer may often be yes. In a registry governing access to scarce IPv4 resources, the answer is less comfortable.
One sign of capture is when low turnout is treated as consent. If few people object to a transfer-policy refinement, the active list may read that as support. But low turnout may reflect resignation, ignorance, opportunity cost or the belief that outcomes are predetermined. The affected majority may not know that a proposal matters until after it becomes policy. ARIN cannot wait for every affected party to notice every issue. It should, however, avoid converting passivity into endorsement. A consensus system that does not distinguish silence from support is vulnerable to capture by those who speak continuously.
The transfer market as the new constitution
The North American IPv4 transfer market is not a side issue in ARIN governance. It is the constitutional fact of the post-exhaustion registry. It decides whether new networks can buy growth at a tolerable cost, whether unused or underused holdings can be redeployed, whether legacy holders are given certainty or pressured toward public stewardship, and whether ARIN remains a neutral registrar or becomes a gatekeeper of private transactions. The market is also where capture is easiest to miss. Debates sound technical: block size, needs period, documentation, waiting period, reserved pools, inter-regional compatibility. Underneath the detail sits an asset-allocation question.
ARIN's transfer guidance makes the market visible while preserving the institution's regulatory posture. It says negotiations and financial terms are matters for the parties, but all transfers must comply with current policy. It describes a Qualified Facilitator Program for organizations seeking help finding buyers or sellers. It states that a source in an 8.3 transfer must not have received a transfer or allocation of IPv4 resources from ARIN within the prior 12 months, that reserved-pool space is not eligible, and that a source transferring IPv4 resources to another party cannot apply to the waitlist for 36 months. Recipients must meet transfer requirements, and waitlist status may be affected by transfer activity. The result is a semi-regulated market: private price discovery paired with public recognition.
The registry's power lies in making private bargains usable. A buyer can negotiate a price, but without recognized transfer the block is riskier and less valuable. A seller can claim surplus, but a serious buyer wants registry records, reverse DNS control, routing-security alignment and a clean chain of authority. ARIN does not set the purchase price, but it helps define the asset's quality. That gives it power to discourage speculation, reduce fraud and maintain data integrity. It also gives repeat participants an incentive to shape recognition rules in ways that favor their business models.
Needs-based transfer requirements illustrate the tradeoff. In principle, they preserve the idea that number resources exist for operational networks rather than pure financial storage. ARIN's policy says transfers are made for use on an operational network and that organizations seeking larger initial or additional blocks must document use of at least half the requested IPv4 block within 24 months. That is coherent stewardship. But needs review is costly. It requires documentation, planning and interaction with ARIN. A large operator can prepare a demand forecast. A broker can help a client translate commercial plans into acceptable evidence. A smaller firm may not know how to present its case or may hesitate to expose expansion plans. A rule designed to prevent hoarding can become a barrier to less sophisticated entrants.
Waiting periods create a similar ambiguity. They can prevent immediate flipping and protect the waitlist from arbitrage. They also strengthen organizations that already hold inventory or can buy once and wait. A 36-month restriction on waitlist access after selling resources may discourage some holders from moving space to market if they think they may later need registry-issued resources. A 60-month transfer bar on waitlist distributions protects the integrity of the waitlist but confirms that waitlist space and purchased space are different economic instruments. The rule may be rational. It still has distributional effects.
In a captured consensus environment, those effects are often not debated in plain economic language. The argument is not presented as a claim that large holders should enjoy greater pricing power, that brokers should earn more from complexity or that entrants should carry higher compliance costs. It appears as concern for efficient utilization, registry accuracy, routability, fraud prevention or legal risk. Those concerns are real. The problem is that they can displace a second question: who bears the cost of satisfying them, and who benefits when others cannot? A mature registry asks both questions. A captured one asks the first in public and leaves the second to the market.
Transfer policy also affects competition beyond the registry. IPv4 scarcity can raise entry costs for hosting firms, access networks, managed-service providers, content platforms and enterprises modernizing infrastructure. Incumbents with large holdings can expand without buying at the same margin, lease or sell surplus, or use inventory as a strategic buffer. Rules that appear neutral can change liquidity, documentation cost and certainty, reinforcing concentration while the registry insists that it only maintains records.
Legacy resources and the price of historical luck
Legacy IPv4 resources are the most delicate part of ARIN's political economy because they combine legal caution, historic accident and contemporary scarcity. Early allocations occurred before today's regional-registry framework matured. Some organizations received blocks that would be implausibly large under modern conservation standards. Decades later, those blocks can carry substantial market value. The issue is not whether every legacy holder is abusing a windfall. Many use their space operationally, and stable registration serves the Internet. The issue is whether a consensus system can fairly govern resources whose value was created by collective scarcity but whose control often reflects historical luck.
ARIN has strong reasons to provide certainty. If holders distrust the registry, they may resist signing agreements, updating contacts or participating in transfers. Poor data quality harms everyone. Routing-security practices are harder to coordinate when records are stale. Fraud risk rises when abandoned organizations, outdated contacts or unclear corporate succession obscure who may authorize a transfer. A registry that threatens legacy holders too aggressively could push activity into gray markets or litigation, weakening the stewardship it is meant to protect.
Yet certainty is not free. It can become a shield for private gain. The public interest in accurate records can be converted into an argument that legacy holders must receive favorable treatment indefinitely. ARIN's 2026 fee schedule provides a small but revealing example: organizations with active Legacy Registration Services Agreements entered before 1 January 2024 receive a cap on total annual fees for legacy resources, limited to $250 annually for 2026 and rising by $25 per year, while legacy resources brought under agreement after that date fall under the regular Registration Services Plan. The cap may encourage formalization, but it also differentiates historic holders from ordinary customers who pay according to aggregate holdings.
The fee cap is not the main driver of IPv4 economics. It is a sign of the settlement. A very large legacy holder under a capped agreement may face registry fees that bear little resemblance to the market value or scarcity of the resource. That may be a reasonable price for better records and lower legal conflict. But if so, the bargain should be explicit: certainty and favorable treatment in exchange for accurate data, accountable transfers and continuing contribution to the registry system. Without such accounting, certainty can look less like stewardship and more like a subsidy to historical luck.
Transferability is the larger issue. A legacy holder with recognized standing can sell, reorganize or transfer resources subject to policy. A buyer can obtain registry recognition subject to policy. Each step depends on the registry's willingness to treat the resource as a transferable registration right rather than a simple administrative entry. ARIN continues to insist that number resources are not ordinary property. Markets nevertheless price them. This tension is manageable only if the registry is credible as a neutral steward. If affected parties conclude that rules are shaped mainly by those already holding valuable space, neutrality becomes less believable.
Different actors have different incentives around legacy certainty. Large holders may favor clean standing, stable fees and predictable transfers. Brokers may favor enough complexity to justify intermediation but not enough uncertainty to stop deals. Large buyers may favor a regulated market that filters dubious sellers while allowing well-advised transactions. Small networks may favor lower barriers, more transparent market data and stronger pressure to bring unused space into productive use, but they often lack the time and legal support to advance those positions. The resulting consensus can look balanced because it contains several sophisticated interests. It may still exclude the less organized demand side.
There is also an epistemic problem. Legacy-resource debates quickly become legalistic. Which organization signed what agreement? Which successor is legitimate? What documentation proves control? When does a corporate restructuring become a transfer? How should disputes be handled? These questions are necessary. But legal and procedural complexity can obscure the economic question: how much of the scarcity premium should accrue to historic holders, and how much should policy recapture for broader access, accuracy or public benefit? If that question is never asked directly, policy will answer it indirectly in favor of those best equipped to navigate complexity.
The legitimacy risk is cumulative. One accommodation may be pragmatic. A series of accommodations, combined with high transfer prices and a culture that treats redistribution concerns as naive, can make the registry appear to guard incumbents. Technical correctness will not be enough if that perception takes hold. Regional registries survive by recognition. Operators use ARIN records because they believe the records are useful, stable and legitimate. If legacy certainty begins to look like a private settlement among insiders, the registry's public standing weakens even while the records continue to function.
Member power and procedural capital
ARIN's membership structure matters because policy participation and corporate authority are not the same thing. Policy discussions are open to a broad Internet community, but elections are conducted through General Members in good standing. ARIN's membership materials identify Service Members, General Members and Trustee Members. They say membership is not required to obtain direct Internet number resources, participate in policy discussions, submit suggestions or join public consultations. That openness matters. The same materials say that only General Members in good standing may vote in ARIN elections through a designated Voting Contact, and that General Members must vote to maintain status; after the 2023 annual election and each election thereafter, a General Member that did not cast a ballot in any of the previous three elections reverts to Service Member status.
This creates a paradox. ARIN wants broad participation, but its formal electorate is self-selecting and attention-sensitive. Organizations that do not designate a voting contact, miss repeated elections or lack staff attention lose voting power. They can still speak, but they no longer elect the people who sit on the Board of Trustees and Advisory Council. In most corporate settings this might look ordinary. In a registry that claims community legitimacy over scarce public infrastructure inputs, it is more consequential. The policy arena may be open to all, but continuing institutional authority is shaped by organizations sufficiently organized to maintain voting status.
The small-electorate problem is about composition as much as numbers. General Members who vote consistently are more likely to understand the value of ARIN governance. That may include public-spirited operators, experienced engineers and civic-minded institutions. It may also include large holders, brokers, policy veterans, vendors, cloud firms, access providers and others with financial exposure to transfer rules, fees and registry services. The repeated-voting rule may improve election quality by removing dormant members from the roll. It may also narrow the electorate toward those already invested in the institution's existing culture.
This matters because elected and embedded actors interpret the process. The Advisory Council is central to the PDP. Policy Shepherds, selected by the council chair, guide proposals and can be changed by the chair. The Board reviews recommended policies and may adopt, reject, remand or seek clarification. The broader community can speak, but the council decides whether text advances, how feedback is integrated and whether support is sufficient. The Board then supplies corporate authority. A persistent electorate therefore shapes not only who appears in title but how the process understands seriousness.
The petition mechanism appears to mitigate this concentration, but its design also reveals the cost of opposition. ARIN's process allows a member of the Internet community to initiate a petition asking the Board to overrule certain Advisory Council decisions believed not to have followed the process. Support must come from registered points of contact for ARIN member organizations. The support window is seven calendar days, and a petition succeeds only if at least 15 valid support forms from 15 different member organizations are received. Fifteen may sound low. In a niche policy community, across one week, requiring valid forms from member organizations is not trivial. A petitioner must already know which organizations care, who their registered contacts are, how to reach them quickly and how to persuade them to act inside the deadline. That favors people with networks.
Capture through member power need not involve ballot manipulation or exclusion. It can operate through attendance, familiarity and time. Actors most capable of sustaining institutional routines gradually define what seriousness looks like. They know the difference between a policy proposal and a suggestion. They know when the Board can act, when the Advisory Council can revise, when staff and legal review will matter and when a petition is possible. Their advantage is not secret information. It is procedural capital.
The legitimacy risk is not that every elected person represents a narrow interest. Many long-serving participants act from genuine stewardship commitments. The risk is institutional: a system can be populated by conscientious individuals and still produce biased outcomes if it overweights those with capacity to participate. ARIN's challenge over the next 12-24 months is not simply to keep elections running. It is to show that the electorate, the candidate pool and the deliberative center are not drifting toward a guild of policy specialists whose decisions mainly validate incumbent expectations.
Accountability without a single sovereign
ARIN's accountability problem is subtle because it has many accountability mechanisms but no single public sovereign. It is a Virginia nonstock corporation with bylaws, a board, officers, members, an Advisory Council, published corporate documents and public policy materials. It participates in the wider regional-registry system. It provides operational registry services. It is accountable to customers through service expectations, to members through elections, to the Internet community through policy participation and to law through corporate obligations. That multiplicity is often presented as strength. It can also blur responsibility.
The bylaws locate corporate authority in the Board of Trustees, subject to legal and corporate limits. The Board retains oversight authority over nomination, appointment, election and removal processes. The President is a voting board member. The Advisory Council acts in an advisory capacity on Internet-number policy matters and has up to 15 elected persons, with the President as a non-voting ex officio participant. These arrangements may be normal for a non-profit registry. They also mean that the same institution defines process, evaluates process, adopts policy, manages elections and implements outcomes.
Self-contained accountability can work when the community believes the institution is neutral. It becomes fragile when the institution is accused of capture. A public regulator can be challenged through politics or administrative law. A market actor can be disciplined by competition. A regional registry has no simple equivalent. Its authority rests on recognition. If the process is viewed as unfair, participants can complain, litigate, route around, ignore stale records, delay cooperation or support alternatives in extreme cases. Those are blunt instruments. Because they are blunt, grievances can accumulate without correction until legitimacy is already damaged.
Post-exhaustion economics intensify the problem. When ARIN distributes scarce recognition over transfers, waitlist allocations, routing-security eligibility, Whois records and agreements, its accountability is not merely technical. A transfer can involve large sums. A policy change can alter the value of holdings or the cost of growth. A fee schedule can shift burdens across organizations. A legacy agreement can determine whether an old allocation remains cheap to maintain. These are economic decisions made through a governance model that still speaks largely in the language of community stewardship.
The 2026 fee schedule shows the mixture of cost recovery and distributional consequence. ARIN says it recovers registry costs through annual fees charged to customers. The schedule sets Registration Services Plan categories from 3X-Small at $275 to 5X-Large at $282,240, with annual increases capped at no greater than 5% as approved by the Board. It imposes a $500 non-refundable transfer processing fee for each request for transfer of registration rights, recipient processing fees that scale by transfer size, a $10,000 annual Qualified Facilitator fee and a $5,000 annual Premier Support Plan for certain customers. These may be reasonable charges. They also shape who can afford which services, who can participate professionally and who finds the market too expensive to use.
Accountability requires more than publishing the schedule. It requires explaining how burdens fall on small holders, large holders, legacy holders, brokers, new entrants and regional minorities. A fee can be transparent and still regressive in practice. A transfer charge can be administratively justified and still deter small transactions. A facilitator fee can screen for seriousness and still privilege established intermediaries. The Board can approve charges properly while affected parties lack a strong channel to contest the economic policy implicit in them.
ARIN's consultation and suggestion channels address some non-policy issues. They can be useful for service dependencies, routing-security notifications, geofeed fields, member-list issues or registry-service improvements. But separating policy from services and fees can weaken accountability when the economic problem crosses categories. IPv4 transfer friction is not only policy, not only service, not only fee and not only legal risk. It is all of them. Fragmented channels can make each part appear too narrow to carry the whole critique.
True accountability in this setting would require ARIN to measure whether its processes produce representative contestation. That means publishing not only proposal status but participation patterns: who comments, which sectors are absent, how often small organizations participate, how remote participation affects outcomes, whether petitions are attempted or avoided, and how Advisory Council decisions correlate with proposal origin. It also means treating market structure as a governance issue rather than a private matter outside the registry's concern. ARIN does not set transfer prices, but its rules create tradable certainty. Accountability must follow power, not only formal mandates.
Post-exhaustion incentives
Institutions change when their environment changes, but they seldom rewrite their self-image at the same speed. ARIN's self-image is rooted in stewardship of number resources for Internet growth. Post-exhaustion, its incentives are more complicated. It no longer sits mainly above an expanding pool; it sits beside a market that depends on its records. Its customers need services beyond allocation: transfers, routing security, reverse DNS, Whois accuracy, automation, data access, fraud prevention and contractual clarity. Its revenue comes from fees attached to resource holdings and services, not from market prices. Its legitimacy comes from the claim that policy is community-developed. Each incentive can pull in a different direction.
One incentive is stability. The registry has reasons to avoid radical changes that could unsettle holders, buyers, routers, lawyers and coordination partners. Stability is valuable. The addressing system should not be a laboratory for constant redistribution. But stability can become incumbent protection. When the status quo gives large resource holders predictable market value, a bias toward stability is not neutral. It preserves a distribution created by past allocation and current scarcity.
A second incentive is service growth. As ordinary IPv4 allocation declines, registry relevance depends on services around existing resources and IPv6 adoption. Transfer processing, routing security, IRR, RDAP, Whois accuracy, automation and support become more salient. This can make ARIN more customer-service oriented. That is good if customers are understood broadly. It is risky if "customers" primarily means organizations with resources, fees and staff attention, rather than the wider set of networks and users affected by address-market outcomes.
A third incentive is legal caution. ARIN must avoid unreasonable corporate and liability risk, and the PDP requires policies to be consistent with law and governing documents. Legal caution is unavoidable where transfers and legacy resources are valuable. It is also a powerful conservative force. If every ambitious reform is filtered through possible liability while the existing settlement is treated as operational fact, legal caution becomes a ratchet. It lets advantages accumulated under older rules persist because changing them feels riskier than tolerating them.
A fourth incentive is participation maintenance. ARIN needs enough visible community activity to sustain legitimacy. Fellowships, outreach, public meetings, remote participation and open lists create on-ramps. They should not be dismissed. Yet there is a difference between bringing new people to the edge of the process and enabling them to alter outcomes. Outreach can become a legitimacy display if the core policy economy remains controlled by experienced participants. The institution may increase participation quantity without changing participation power.
A fifth incentive is global comparability. ARIN is one regional registry among five. It cannot ignore inter-regional policy compatibility, global identifier coordination or the expectations of the wider registry system. These constraints can be real. They can also become a vocabulary for avoiding regional accountability. If a proposal challenges transfer rules, compatibility can be invoked. If a proposal challenges recognition of resources, global stability can be invoked. Such arguments may be valid in particular cases. They should not become automatic defenses for continuity.
Post-exhaustion also changes the meaning of IPv6. ARIN and other registries have long encouraged IPv6 adoption, and the 2026 fee schedule includes a temporary IPv6 fee waiver for organizations in the 3X-Small service category, allowing more IPv6 resources while still paying the 3X-Small fee through 31 December 2026. IPv6 is the long-term answer to IPv4 scarcity at the protocol level. But it does not erase short- and medium-term IPv4 economics. Many networks still need IPv4 reachability for customers, applications, hosting and transition mechanisms. A registry can promote IPv6 and still govern IPv4 transfers fairly. It should not use one duty to avoid the other.
The most dangerous incentive is the desire to preserve legitimacy by preserving appearances. An open archive, a free meeting, a public consultation, a Board review and a published manual all signal procedural health. If ARIN becomes too invested in those signals, it may treat capture analysis as hostility to the community rather than as a warning about representation. That is how open systems decay. They confuse access with influence, documentation with accountability, and active-minority support with regional consent.
What a less captured ARIN would require
A less captured ARIN would not abandon consensus. It would make consensus more representative and more economically honest. The first step would be to describe transfer-market policy in distributive terms. When a rule changes transfer size, waiting periods, documentation, recipient need or inter-regional compatibility, the public assessment should state who is likely to bear the cost and who is likely to gain. It should not be enough to say that a proposal is technically sound or supported by the active community. The institution should ask whether that active community contains the affected parties in plausible proportion.
Participation costs should be treated as governance costs. If a proposal requires months of mailing-list engagement, meeting attendance and legal familiarity, the process should acknowledge that only certain actors can comply. Remote participation, fellowships and free registration are useful but incomplete. A more serious approach would include plain-language economic briefs for major proposals, structured calls for input from small networks and regional minorities, and post-decision reviews of who actually participated. The goal is not to replace expert deliberation with opinion polling. It is to prevent expert deliberation from becoming a proxy for consent.
The petition mechanism should be evaluated against actual use. A seven-day support window and 15 member-organization forms may be defensible if outsiders can realistically use them. If petitions are rarely attempted, that may mean Advisory Council decisions are trusted. It may also mean challengers know the threshold is socially difficult. ARIN should publish enough data to distinguish those possibilities. The legitimacy of a safety valve depends not on its existence but on whether affected parties can pull it.
Fee and transfer consultations should be joined when the economic issue crosses categories. If friction arises from the interaction of manual text, staff review, legal requirements, processing charges and facilitator economics, separating the issue into policy and services obscures the whole. ARIN should create deliberation formats for cross-cutting market issues without forcing participants to misclassify them. That would reduce the insider advantage created by knowing which door to knock on.
Legacy-resource certainty should be paired with public-interest accounting. ARIN may decide that fee caps, agreement protections and predictable transfer standing are necessary to keep records accurate and avoid legal conflict. If so, it should explain the public benefit and opportunity cost. How much address space is covered by legacy arrangements? How does fee treatment compare with ordinary holdings? What incentives encourage updated contacts, signed agreements and return or transfer of unused space? What safeguards prevent certainty from becoming a perpetual reward for historical luck? These questions are not anti-legacy. They are pro-legitimacy.
The Advisory Council and Board should be assessed not only by election mechanics but by diversity of economic position. Geography and biography matter, but so does resource position: small holders, non-holders seeking entry, access networks, enterprise users, public-interest operators, Caribbean networks, legacy holders, brokers, large platforms and security practitioners. A council dominated by people with similar procedural backgrounds may deliberate well while seeing too little. Representation in this context is not about symbolic balance. It is about the range of costs that deliberators understand without being taught.
ARIN should also avoid using IPv6 as an escape from IPv4 governance. IPv6 deployment is essential. It does not eliminate the need for fair transfer policy during the transition. A credible registry can promote IPv6 while recognizing that IPv4 scarcity creates rents, entry barriers and legitimacy risks. The transition period is not a pause in political economy. It is the period in which scarcity rents are allocated.
None of these reforms requires treating ARIN as corrupt or illegitimate. They require treating it as a powerful infrastructure institution operating in a market shaped by scarcity. That is a more demanding standard than procedural openness. It asks whether the rules produce fair contestation under unequal participation costs. It also asks whether the institution is willing to examine its market effects without hiding behind technical neutrality.
The 12-24 month risk map
Over the next 12-24 months, the main risk is not a single dramatic policy fight. It is a slow widening of the gap between procedural legitimacy and economic legitimacy. Procedural legitimacy will remain strong if ARIN continues to publish materials, hold meetings, archive lists and follow its PDP. Economic legitimacy will weaken if smaller networks and entrants conclude that transfer costs, legacy certainty and insider culture are fixed features they cannot influence. The danger is precisely that the public record will look healthy while trust thins outside the active core.
The first watchpoint is transfer liquidity. If IPv4 transfer activity remains central to growth while rules continue to require substantial documentation and procedural knowledge, the market will reward intermediaries and sophisticated buyers. That may be tolerable if fraud is low, records are accurate and entry remains possible. It becomes dangerous if small organizations experience the transfer system as expensive, opaque or biased toward known players. ARIN should be judged by the accessibility of successful transfer participation, not only by the number of processed tickets.
The second watchpoint is legacy-resource treatment. Fee caps, agreement incentives and transfer certainty should be monitored for their effect on public trust. If legacy holders appear to receive predictable benefits while entrants face high market prices and strict documentation, the fairness claim will weaken. The registry should not destabilize legacy records. It should make the public bargain visible: certainty in exchange for accurate data, accountable transfers and contribution to the common registry system.
The third watchpoint is election and membership engagement. A rule that reverts inactive General Members to Service Member status may keep the voting roll current, but it can also narrow voting power to the most attentive organizations. ARIN should watch whether the electorate becomes more concentrated, whether candidate pools reflect the economic diversity of the region, and whether voting-contact requirements shape participation. Governance of a scarce market should not depend on a thin electorate of insiders.
The fourth watchpoint is proposal origin and fate. A healthy policy system should show that proposals from outside the long-standing core can progress, change materially and sometimes win. If most consequential changes come from familiar participants, or if outsider proposals are remanded, narrowed or diverted, capture risk rises. The relevant measure is not politeness toward newcomers. It is whether newcomers can alter rules.
The fifth watchpoint is the treatment of economic argument. If transfer-market concerns are repeatedly dismissed as outside scope, insufficiently technical or better suited to another channel, the institution will be avoiding the central post-exhaustion question. A registry governing scarce identifiers cannot be neutral by refusing to discuss distribution. Refusal itself distributes power.
The sixth watchpoint is Caribbean and smaller-region participation. ARIN's region is not only the continental United States and Canada. Many Caribbean and North Atlantic territories fall within the service area. Outreach is useful, but representation must be visible in policy influence. If regional diversity appears mainly in fellowship language and meeting geography while outcomes reflect the priorities of large North American actors, legitimacy risk deepens.
The seventh watchpoint is rhetoric. Healthy institutions can distinguish between criticism of process and hostility to the mission. Insecure institutions treat capture analysis as an attack on community. ARIN should be able to say that the process is open, participation is unequal, scarcity changes incentives and evidence is needed that consensus is not simply the view of those who can afford to participate. If it cannot say that, the problem is larger than any single policy.
The conclusion ARIN should fear
The harsh conclusion is not that ARIN is closed. It is that openness may be insufficient for the market it now governs. The door is open, the list is public, the meeting is free, the manual is published, the board minutes exist, the petition path is written down and the transfer rules are visible. Those are real achievements. They are also insufficient safeguards against consensus capture in a post-exhaustion market.
The economics of capture are straightforward. Scarcity creates rents. Rents create incentives to monitor rules. Monitoring rules creates procedural expertise. Procedural expertise creates agenda power. Agenda power shapes which reforms are considered practical. Practicality becomes the language through which incumbent advantage defends itself. At no point does the process need to become formally closed. It only needs to remain costly enough that the same actors dominate.
ARIN's strongest defense is also its greatest vulnerability. It can say that anyone may participate. But if meaningful participation requires time, history, drafting skill, meeting presence, social networks and organizational attention, then "anyone" is a formal category rather than an economic reality. The relevant question is not who may speak. It is whose speech can change outcomes.
The North American IPv4 transfer-market setting makes this more than a procedural complaint. Address scarcity affects entry, competition, pricing, routing practices, customer experience and the distribution of historic windfalls. Legacy-resource certainty may be necessary, but it must not become a quiet settlement for the already fortunate. Registry accountability may be complex, but complexity must not blur responsibility. Member power may be orderly, but an attentive minority should not be mistaken for the region.
ARIN does not need to choose between stewardship and market realism. It needs to admit that stewardship now operates through a market. Once that is admitted, the standards change. A policy is not fair merely because it is technically coherent. A consensus is not legitimate merely because opposition was weak among active participants. A transfer rule is not neutral merely because it applies to everyone. A legacy accommodation is not justified merely because it reduces legal uncertainty. Each rule must be judged by how it distributes opportunity under scarcity.
The next phase of ARIN governance will be measured by whether it can make that judgment openly. If it can, the institution may renew its legitimacy for the post-exhaustion era. If it cannot, the likely future is not immediate revolt but quieter disengagement: smaller actors treating policy as theater, entrants treating transfers as a toll market, legacy holders treating certainty as entitlement, and procedural insiders mistaking orderly meetings for consent. That is how consensus capture succeeds. It wins not by closing the process, but by making openness too expensive to use.
Public record anchors for this analysis include ARIN's Policy Development Process at https://www.arin.net/participate/policy/pdp/, the Number Resource Policy Manual at https://www.arin.net/participate/policy/nrpm/, transfer guidance at https://www.arin.net/resources/registry/transfers/, membership materials at https://www.arin.net/participate/oversight/membership/, the 2026 fee schedule at https://www.arin.net/resources/fees/fee_schedule/, the bylaws at https://www.arin.net/about/corporate/bylaws/, the regional service-area page at https://www.arin.net/about/welcome/region/, the mailing-list page at https://www.arin.net/participate/community/mailing_lists/, the meetings page at https://www.arin.net/participate/meetings/, and the statistics and reporting page at https://www.arin.net/reference/research/statistics/. These materials are treated here as exhibits of institutional design and market rules, not as the measure of whether the design is legitimate.