ARIN and the economics of inter-RIR transfer politics

ARIN is examined through inter-RIR transfer politics as a registry-governance and institutional-economics problem for the North America region.

Article

ARIN is one of the cleanest test cases for a problem the Regional Internet Registry system has not yet solved. Cross-registry transfer rules expose a conflict between two ideas that were once easier to reconcile: regional stewardship and global asset mobility. The conflict is not mainly about public-meeting rhetoric or whether a registry publishes a neat page explaining its procedures. It is about the economic character of IPv4 after exhaustion. A finite identifier that once moved through administrative allocation now moves through contracts, acquisitions, leases, restructurings, network-continuity plans and private capital. The registry still calls the thing a number resource. The market increasingly treats recognised control over it as a scarce operating asset.

ARIN sits at the centre of that transition because North America is both a mature registry region and a large address economy. The region contains cloud platforms, national carriers, universities, data-centre networks, enterprise holders, brokers, address-rich legacy estates and buyers with enough capital to turn policy delay into a quantifiable cost. It also contains many smaller operators for whom a /24, a /22 or a modest inter-RIR transfer is not an abstract line in a manual but a constraint on customer growth. The result is a comparatively orderly registry environment in which the deeper issue can be seen without the distraction of institutional collapse. ARIN's rules show how a stable registry can still become a gate in a global capital market.

The core claim is simple. Inter-RIR transfer politics is market design. It decides whether IPv4 can move toward higher-valued use across administrative regions, how much friction a buyer must price, how much discretion a registry retains, and how far the old language of conservation can survive once the free pool is gone. ARIN's transfer page says transfers of IP addresses and ASNs issued by ARIN or its predecessors are governed by ARIN policies. It distinguishes merger, acquisition and reorganisation transfers, specified-recipient transfers inside the ARIN region, and inter-RIR transfers with other Regional Internet Registries. Its Number Resource Policy Manual supplies the legalistic vocabulary: registration, conservation, routability, stewardship, justified need, transfer approval and resource review.

Those materials are useful exhibits. They are not the conclusion. Official RIR, NRO and ICANN language can describe institutional posture, policy text and service claims; it cannot be allowed to decide the economic question by framing it as a matter of stewardship alone. The conclusion has to be drawn from institutional economics. A registry can be indispensable as a ledger while being dangerous as a gatekeeper. It can preserve uniqueness, contactability, RDAP and Whois records, reverse DNS delegations, RPKI-adjacent status, transfer history and dispute metadata without also deciding whether capital should be allowed to cross a regional boundary.

Market-side critics make that distinction more sharply. Lu Heng's public notes argue for protecting the ledger rather than protecting the incumbent gatekeeper; the Number Resource Society frames number-resource governance as a decentralisation and portability problem; LARUS presents registry risk as a business-continuity problem for operators exposed to contract and policy discretion. These are not neutral oracles, and their incentives should be read openly. They are nevertheless useful counterweights to official language because they ask the question the official record often avoids: what power does a registry really need in order to keep the Internet working?

ARIN is not a villain in this story. It is a revealing institution. Its published procedures are clearer than many. It allows transfers. It publishes compatible inter-RIR counterparties. It has a policy development process. It recognises legacy resources in ways that preserve some services even without a full ARIN agreement. It offers transfer pre-approval, a qualified facilitator programme and public statistics. Yet it also retains the allocation-era premise that a registry should make prudent decisions about whether a transfer should be approved, that resources are not sold, that recipients must demonstrate need, and that cross-registry movement depends on reciprocal compatible needs-based policy. In a world where the asset is globally useful but the registry remains regionally framed, that is the political economy.

Exhaustion changed the meaning of stewardship

Before IPv4 exhaustion, the strongest argument for needs-based allocation was administrative fairness. A registry had a finite pool. Applicants wanted addresses. The registry needed a method for deciding who received what. Conservation, efficient utilisation and operational need were imperfect but intelligible rules for distributing unallocated supply. They reduced obvious waste, discouraged purely speculative grabbing and made it possible to explain why one applicant received space while another waited.

After exhaustion, the same rules operate in a different market. ARIN's IPv4 Addressing Options page says the ARIN free pool of IPv4 address space was depleted on 24 September 2015. It points applicants toward reserved-policy exceptions, a waiting list for returned or recovered space, and transfers to specified recipients under NRPM 8.3 or 8.4. That is a structural break. The marginal address is no longer usually distributed from an administrative pool at registry cost. It is acquired from an existing holder through a market whose participants must price capital, diligence, operational history, reputation, routing-security state and settlement risk.

Stewardship therefore becomes ambiguous. In the allocation era it meant prudent distribution of a scarce common pool. In the transfer era it can mean one of two things. In the narrow sense, stewardship protects the registry function: unique registration, accurate records, valid authority, clean transfer settlement, reliable publication and non-destructive dispute handling. In the broad sense, stewardship becomes a discretionary claim over movement: who qualifies, whether a buyer has enough need, whether a block may leave the region, whether another registry's policy is compatible, whether a legacy holder should enter a contract, and whether a market transaction fits the registry's moral view of efficient use.

The broad version carries economic consequences. If ARIN recognises a transfer only after the buyer satisfies a need test, a private bargain becomes conditional on public-style approval. If a region can export its need test through inter-RIR compatibility, the rule becomes a cross-border capital-control instrument. If fee payment and agreement status condition access to settlement or modern registry services, contract and fee policy become part of asset valuation. If review powers can reach resources already maintained in the database, diligence must price audit risk. None of this requires bad faith. It is enough that one institution controls a settlement layer for a scarce resource.

IPv4's scarcity makes these costs visible because the market no longer treats addresses as administrative leftovers. Addresses are embedded in customer networks, hosting businesses, cloud deployments, access systems, payment flows, security controls, allowlists, data-centre migrations and merger valuations. Renumbering is not a clerical task. It is disruption. Delay in recognition is not a minor inconvenience. It can alter closing conditions, financing, routing-security timing and customer commitments. A compliance request is not a neutral formality when it determines whether a valuable block moves.

This is why ARIN's inter-RIR rules matter beyond ARIN. They test whether the RIR system can distinguish regional recordkeeping from regional asset control. A registry region is an administrative convenience. IPv4 addresses are global identifiers. The packets do not respect RIR borders. A route announcement does not carry a theory of North American stewardship. The economic value of a block depends on worldwide reachability, not regional sentimental ownership. When a registry border controls transferability, the old service-region boundary becomes a new market boundary.

ARIN's inter-RIR rule is a compatibility gate

ARIN's inter-RIR transfer rule is explicit. Under NRPM 8.4, inter-regional transfers of IPv4 addresses or ASNs may occur only through RIRs that agree to the transfer and share reciprocal, compatible needs-based policies. ARIN's transfer materials list APNIC, LACNIC and RIPE NCC as approved for inter-RIR transfers with ARIN, while AFRINIC is not approved. The same materials say inter-RIR transfers involving ARIN may not include IPv6 addresses. For outbound transfers from ARIN, source organisations must be current registered holders, must not be involved in a dispute over the resources, must provide a signed and notarised officer acknowledgement, must meet minimum transfer-size and timing conditions, and must avoid reserved-pool addresses. For inbound transfers to ARIN, recipients inside the ARIN region must satisfy current ARIN policy and demonstrate need for up to a 24-month supply of IPv4 addresses.

This looks like technical compatibility. It is better understood as institutional compatibility. ARIN is not saying that an address from an unapproved region cannot route. It is saying that ARIN will not recognise the cross-registry movement unless the other registry sits inside a compatible policy relationship. That turns policy into a border. A block's technical function is global, but its recognised movement depends on a club of mutually acceptable registry regimes.

The political economy is straightforward. If compatibility is required, each RIR becomes both a registry and a customs authority. A resource can cross only when the source and destination institutions share enough policy DNA. If one region has a strict needs-based regime, it can preserve that regime by requiring reciprocal compatibility. If a region wants liquidity with ARIN, it must keep its transfer rules close enough to ARIN's expectations. If a region lacks compatibility, the addresses under its administration carry an economic handicap even though the addresses are technically equivalent. Policy affects price.

This is the crucial point often hidden by administrative language. Transfer compatibility is not simply a consumer-protection rule. It is a market-access rule. It decides which regional assets can enter which pools of demand. It affects seller bargaining power, buyer choice, broker strategy, settlement timing and the value of legacy holdings. It may encourage formal compliance. It may also suppress liquidity, trap capital in lower-value uses, and push participants toward structures that imitate transfer without clean registry movement.

There are defensible reasons for some compatibility checks. A registry should not accept forged transfers. It should not create duplicate recognition. It should not ignore a valid dispute. It should verify the identity and authority of source and recipient. It should coordinate security-state transition. It should ensure the destination registry can maintain accurate public records. It may have to comply with court orders and legal sanctions. Those are ledger protections.

Needs-based compatibility is different. It asks whether the buyer deserves the resource according to a policy theory inherited from free-pool rationing. Once the resource is privately transferred rather than newly allocated, the question should change. The buyer's willingness to pay, deploy, finance and absorb business risk is powerful evidence of economic need. A registry may be useful in verifying that the buyer is real and the transaction is valid. It is less obviously competent to judge whether the buyer's future business plan is adequate.

ARIN's rule therefore creates an inter-RIR version of the same old tension. The ledger role requires compatibility of records. The gatekeeper role requires compatibility of ideology. ARIN's test combines them.

Needs assessment is rationing logic inside a market

The most important surviving allocation-era instrument in ARIN's transfer architecture is needs assessment. NRPM 8.5 requires a receiving entity to sign an RSA covering transferred resources unless a current RSA is already on file. It states that ARIN allocates or assigns number resources through transfer solely for use on an operational network. It sets the minimum IPv4 transfer size at a /24. It allows an organisation without an ARIN IPv4 allocation to qualify for an initial minimum-size block. For a larger initial block or additional block, the organisation must document use of at least 50% of the requested IPv4 block size within 24 months. Existing ARIN holders must have efficiently used at least 50% of cumulative IPv4 holdings to receive more. An alternative path permits additional IPv4 blocks for organisations demonstrating 80% utilisation, with a /16-equivalent limit within a six-month period.

In a free-pool world, these requirements are allocation criteria. In a transfer world, they are market entry criteria. A buyer can find a seller, agree on price, arrange operational transition, prepare routing-security changes and still wait for a registry judgment about need. The source can be valid. The buyer can be real. The block can be clean. The transaction can be commercially rational. Yet the settlement layer remains conditional.

The cost is not merely paperwork. Needs assessment forces market participants to translate business plans into registry-readable utilisation narratives. It favours buyers whose use resembles historical address-consumption models. It rewards organisations with counsel, registry experience and documentation staff. It burdens newer firms whose growth path may be real but uncertain. It can make sellers prefer counterparties with easier qualification rather than counterparties who value the asset most. It turns brokers and facilitators into navigators of institutional psychology, not only matchmakers of supply and demand.

The burden falls unevenly. A large incumbent can absorb transfer counsel, compliance staff, pre-approval, documentation and delay. A smaller operator may find the same steps material. A data-centre customer may need continuity quickly. A hosting business may face customer demand before it can express demand in a 24-month utilisation forecast. A security firm, VPN operator, content platform or enterprise network may value reserve capacity because future customer commitments are uncertain. Registry need tests are not well designed to price optionality. Markets are.

The usual answer is that needs assessment prevents hoarding and speculation. That answer is incomplete. Hoarding in a scarce market is not solved by making transfers hard. If a holder believes future prices will rise and transfer is administratively difficult, the holder may simply sit on the asset longer. If transfer is predictable, the holder has a clearer route to release unused space to a higher-valued user. Speculation is not automatically waste. It can supply inventory, discover prices and move risk to parties willing to carry it. The question is not whether speculation sounds unattractive. It is whether administrative forecasting allocates scarce resources better than price and contract under objective anti-fraud rules.

Needs assessment also creates confidential information problems. The buyer must show business plans to the registry. The seller may not know how that review is going. ARIN's transfer page notes that source and recipient requests are ticketed separately and that ARIN cannot provide information about another organisation's ticket; the parties are expected to coordinate directly. That may protect privacy, but it also produces settlement opacity. The seller has to price buyer approval risk. The buyer has to price source documentation risk. Each side is waiting on a monopoly recordkeeper whose view of the other side's file is not fully visible.

The more objective the transfer standard, the lower this friction becomes. A registry can verify source authority, officer approval, corporate succession, absence or scope of dispute, minimum block size, security-object transition and fee status. These are observable. A registry forecast of future need is less objective. It is a judgment about business reality made by an institution that does not bear the opportunity cost of being wrong.

Regional stewardship becomes capital control at the border

The regional frame is attractive because it sounds modest. Each RIR serves a region. Each region develops policy. Each region balances local conditions. In the allocation era, that was administratively convenient. It also reflected real differences in network development and participation. The problem arises when a regional registry treats service territory as a claim over the economic movement of globally useful assets.

The Internet's number space is global. IPv4's scarcity is global. Demand for addresses moves with data-centre expansion, cloud migration, content delivery, telecom consolidation, AI infrastructure, security products, access networks and enterprise continuity. A North American block can serve customers in other regions. A buyer in another region can value an ARIN-origin block. A company can change control through merger or restructuring and retain operational dependence on the same addresses. Regional stewardship may still matter for records, contacts and policy participation. It does not automatically justify restricting asset mobility.

ARIN's NRPM principles illustrate the conceptual carryover. Registration guarantees uniqueness and contactability. Conservation aims at efficient distribution to organisations with technical need. Routability recognises that ARIN cannot guarantee that any network will route a particular number resource. Stewardship applies these principles and acknowledges that goals may conflict with the interests of individual operators. Those are coherent principles for a registry manual. But once applied to inter-RIR transfers, they become a theory of capital movement.

The phrase "regional stewardship" can then perform two incompatible jobs. It can mean careful maintenance of records for resources currently administered in the region. That is legitimate. Or it can mean that resources associated with a region should not move freely unless the destination regime accepts the region's policy assumptions. That is capital control. It does not have to be absolute to matter. Compatibility filters, need tests, waiting-list penalties, reserved-pool exclusions, agreement requirements and documentation demands all affect mobility.

The result is a world in which registry geography influences price. A block's value depends not only on size, reputation, routing history and legal chain, but also on the ease with which it can move across a policy boundary. If a region has smooth compatibility with ARIN, its resources are more liquid to ARIN-region buyers. If it lacks compatibility, its resources may be less attractive even where technical deployment would be simple. That discount is not created by the network layer. It is created by the governance layer.

This is why AFRINIC's non-approved status on ARIN's transfer page matters even without adopting any claim about AFRINIC's separate controversies. The fact itself is enough. A region's resources may be technically global but not ARIN-compatible for recognised transfer. That is an institutional border. It may be justified by policy design; it may be defended as stewardship. But economically it means that registry compatibility changes asset mobility.

The better model is not regional lawlessness. It is portability. Records should move with the resource, and the destination registry should be able to maintain accurate publication, security and contact data. Transfer should fail where authority is false, where a court or independent dispute decision blocks movement, where the same resource would be recognised twice, or where the destination registry cannot preserve the record. Transfer should not fail merely because the buyer's region has chosen a different moral theory of need.

Regional stewardship should therefore be narrowed to what regional administration can actually justify: public records, operational contacts, security-state continuity, dispute notations, transparent policy for services, and coordination with other registries. Global asset mobility should be presumed for already allocated IPv4 unless a narrow ledger risk is shown.

The legacy and RSA boundary reveals the title problem

ARIN's legacy resource treatment is one of the best windows into the difference between ledger and gatekeeper. ARIN's legacy resource page explains that legacy holders not under an ARIN agreement can still maintain unique registration in Whois and RDAP, update publicly available data, manage reverse DNS delegations, maintain registry records through ARIN Online and access DNSSEC. The same page says RPKI and IRR access require an ARIN agreement. It also says the legacy fee cap expired on 31 December 2023, while active LRSAs entered before 1 January 2024 retain fee limits for covered legacy resources.

This is a quietly important distinction. It shows that ARIN can maintain a core public registry record without requiring every legacy holder to enter the full contemporary contract stack. The database must remain true because uniqueness and contactability matter to everyone. That is the ledger function. At the same time, access to modern routing-security and routing-registry services can be tied to agreement status. That is the service and contract layer. The two layers are related, but they are not identical.

The market sees the distinction as title uncertainty. "Title" is not an easy word in the RIR system. ARIN's NRPM says number resources are not sold under ARIN administration and are assigned for exclusive use under policy and agreement conditions. RIRs generally resist property framing. But the market does not need ordinary property doctrine to treat recognised control as valuable. Buyers pay for blocks. Sellers receive consideration. Lenders and acquirers diligence the record. Operators build customer systems around continuity. Lawyers draft warranties. The registry entry is not the whole asset, but it is a decisive public signal of recognised control.

Legacy resources therefore carry both value and complexity. They may have historical claim strength, but they may also require careful diligence around corporate continuity, agreement status, service eligibility, transfer path, RPKI availability and fee treatment. A legacy holder that has not signed an agreement may still have core record services. But if operational expectations increasingly demand RPKI and IRR access, the holder may face pressure to enter an RSA. That pressure may be commercially reasonable. It also shifts bargaining power. The holder remains formally free; the operational cost of remaining outside the contract rises.

ARIN's agreements page states that legal contracts define and bind the relationship between ARIN and its customers, and that ARIN must receive a signed RSA before approving creation of an Org ID in ARIN Online. It also says the agreement guarantees services while ARIN reserves the right to modify it at any time, with or without notice. Both existing and newly approved customers must sign the current RSA for each resource request. For a low-value administrative service that might seem ordinary. For scarce IPv4 capacity, the contract becomes a control surface.

The RSA boundary matters in inter-RIR transfers because movement often means new agreement obligations. An inbound recipient must have an updated and signed RSA and pay applicable fees. A legacy resource may cross into a new contractual environment. A buyer must ask not only whether the resource routes, but what registry obligations attach after settlement. A seller must ask whether a historical block's value changes when it leaves one status and enters another. The cross-registry transaction is therefore a legal and institutional migration as much as a change in recognised holder.

The ledger-first view would make the core record durable and the service stack explicit. A registry should be able to charge for services, require accurate contacts, prevent fraud and maintain routing-security systems. It should not blur the line between running a database and owning the asset's economic future. Legacy resources prove that this separation is possible. They also show how easily service dependence can become contract leverage.

Fees and facilitators reveal compliance burden

Transfer politics is not only about permission. It is about the cost of permission. ARIN's 2026 fee schedule lists non-refundable transfer fees: $500 for 8.2 merger, acquisition and reorganisation transfers; $500 for source requests in 8.3 specified-recipient transfers; and $500 for source requests in 8.4 inter-RIR transfers. Recipient transfer processing fees vary by aggregate IPv4 transfer size, ranging from small amounts for /24-scale transfers to much larger sums for very large blocks. The same schedule lists a $10,000 annual Qualified Facilitator fee and states that the $500 transfer processing fee does not guarantee approval.

These amounts are not necessarily excessive in isolation. A registry needs revenue to operate. Transfer review consumes staff time. A qualified facilitator programme may improve competence and reduce confusion. The economic point is cumulative. Fees sit on top of legal diligence, broker fees, escrow costs, documentation preparation, internal engineering work, routing-security cleanup, reverse DNS transition, blocklist remediation, geolocation updates, contract review and timing uncertainty. The visible fee is only the tip of the compliance cost.

Compliance burden changes market structure. Large buyers can treat it as transaction cost. Small buyers may treat it as a barrier. A seller deciding between counterparties may prefer the buyer most likely to pass review quickly, not necessarily the buyer with the highest economic use. A broker may add value by predicting registry reactions. A qualified facilitator may become part of the expected transaction architecture. The market becomes less about matching address supply to address demand and more about matching participants to institutional approval capacity.

This is where liquidity suffers. Liquidity is not merely the existence of a legal transfer route. A market is liquid when participants can transact at predictable cost, with reliable settlement and manageable information risk. A rule that technically permits transfer but imposes high uncertainty, delay and documentation demands can still suppress liquidity. It can widen bid-ask spreads. It can reduce seller willingness. It can make large clean blocks more valuable than fragmented or legally complicated blocks beyond what routing considerations justify. It can push users toward leasing, nominee arrangements or corporate structures that avoid clean transfer.

The compliance burden also complicates cross-registry transfers. An inter-RIR movement must satisfy both sides' policy architecture. ARIN may request certification from the receiving RIR that the recipient request is compatible with ARIN's needs-based policy. The receiving RIR has its own standards. The parties may need to coordinate timing across institutions, agreements, fees and security-state changes. The more discretion embedded in each layer, the more the market depends on parties who can absorb uncertainty.

The irony is that compliance burden is often defended as a protection for the community. In practice it may protect incumbents. Sophisticated market participants can hire specialists. Smaller operators face proportionally greater cost. If the policy objective is to preserve access for real networks, a friction-heavy transfer architecture may do the opposite. It may make scarce IPv4 more available to those with institutional capacity and less available to those with operational need but weak administrative muscle.

Member power is real, but not asset-owner consent

ARIN's governance model is not empty. Its membership page says there are Service Members, General Members and Trustee Members. Membership is not required to obtain direct number resources from ARIN, nor is it required to participate in policy discussions, submit suggestions or take part in public consultations. General Members in Good Standing can vote in ARIN elections through a designated Voting Contact. The Policy Development Process says policies must be developed through open and transparent processes with meaningful opportunity for public participation, and that significant support may be demonstrated by a subset of the active community rather than unanimity.

This is a serious governance design. It is more open than a closed regulator. It gives resource holders and interested parties channels to propose, discuss and contest policy. It creates elected bodies and public archives. It gives ARIN a stronger accountability story than a purely staff-driven bureaucracy.

But member power is not sovereign consent. A service region is not a polity. A voting contact is not a citizen. The active portion of a mailing-list discussion is not the entire affected economic base. Downstream customers, leased-address users, acquisition lenders, enterprise buyers, security-dependent customers and many small operators may be affected by transfer rules without appearing in policy forums. A policy can have significant support among those who show up and still impose costs on those who do not.

This is not a procedural defect unique to ARIN. It is a structural feature of technical governance. The cost of participation is attention. Those with time, expertise, institutional memory and motivation shape the rules. Those running networks may rationally remain silent until a rule affects a transaction. Those outside the region may care about inter-RIR mobility but have weaker incentives or fewer practical routes to influence ARIN policy. The result is a participation economy, not a full representation system.

Mandate laundering begins when this difference is hidden. A technical coordination body adopts rules through a visible process. The process is described as community-supported. Community support is then treated as a mandate over scarce assets. Regional stewardship is invoked to give the rules moral weight. The registry's narrow recordkeeping function becomes a broader authority over capital movement. Nobody has to say "we control capital." The language of policy, consensus and stewardship performs that work.

ARIN's own PDP acknowledges that policy must not create unreasonable fiduciary or liability risk for ARIN and must be consistent with its corporate documents and law. That is sensible. It also reminds the reader that ARIN is an institution with its own risk constraints. Its policies are not pure expressions of an abstract Internet community. They are choices made through a process operated by a corporation that must manage its mission, legal exposure, fees and staff implementation. That does not invalidate the process. It makes accountability more important.

Registry accountability should therefore be measured by more than openness. It should ask how often transfer requests fail and why, how long inter-RIR transfers take by counterparty region, how often need documentation is challenged, how many requests are abandoned after review begins, how often compatibility issues arise, how many source disputes are recorded, how fee and agreement status affect timing, and how smaller organisations experience the system. Public transfer statistics help, but the market needs more than counts. It needs risk signals.

Ledger versus gatekeeper is the institutional test

The ledger-versus-gatekeeper distinction is the cleanest way to judge ARIN's inter-RIR transfer politics. The ledger protects uniqueness, accuracy, authority, contactability, publication, security-state transition and dispute isolation. The gatekeeper decides whether a transaction should be allowed according to broader policy judgments. Some gatekeeping is unavoidable because a ledger cannot protect itself without rejecting invalid updates. The question is where the boundary sits.

ARIN's transfer rules include ledger-protective elements. Requiring the source to be the current registered holder protects the record. Requiring proof of corporate succession in merger and acquisition cases protects the chain of authority. Requiring that the source not be in a dispute over the resources protects against corrupting the ledger. Requiring officer acknowledgement reduces fraud. Requiring coordination of ROAs, IRR objects and reverse DNS protects operational transition. Requiring the destination registry to maintain records is sensible.

Other elements look more like gatekeeping. Needs assessment in private transfer judges the buyer's future use. Reciprocal compatible needs-based policy exports ARIN's allocation-era logic across borders. Waiting-list penalties tie transfer behaviour to residual rationing. RSA dependence makes agreement status part of settlement. Resource review powers can turn current database maintenance into audit exposure. Fee and facilitator systems can reward institutional familiarity. The market may accept some of these as the cost of order. But they should be named for what they are: controls over movement.

The line is not always obvious. Minimum transfer size may serve routing and administrative practicality. Reserved-pool exclusions may protect special-purpose allocations. Waiting-list restrictions may prevent arbitrage of recovered space. A signed RSA may define service obligations. The issue is proportionality. Does the rule protect the ledger from corruption, or does it preserve administrative power inherited from allocation scarcity? Does it reduce fraud, or does it suppress liquidity? Does it protect operational networks, or does it make them dependent on a discretionary office?

Lu Heng's registry-continuity note argues that continuity of records, RDAP, Whois, reverse DNS, RPKI publication, running networks and independent dispute handling should not be confused with continuity of an incumbent institution's authority claims. Applied to ARIN, the point is not that ARIN should disappear. It is that ARIN's legitimacy is strongest when it is most ledger-like. Its legitimacy weakens when technical-continuity language justifies broader control over asset movement.

The ledger test is especially important for inter-RIR transfer because cross-border mobility exposes hidden assumptions. If ARIN's role is to preserve the accuracy of resources under ARIN administration, then an outbound transfer should be recognised when the source is valid, the recipient registry can maintain the record, the security-state transition is clean and no narrow block applies. If ARIN's role is to enforce a regional theory of need even after a buyer and seller have agreed, then ARIN is not merely a ledger. It is a gatekeeper with global market influence.

Inter-RIR compatibility can launder mandates

Mandate laundering is a subtle institutional mechanism. A narrow function is passed through broader language until it emerges as a wider mandate. In the RIR context, the narrow function is maintaining a unique and reliable registry. The broader language is community, stewardship, conservation, technical need, regional policy and stability. The resulting mandate can include discretion over transfers, reviews, agreements, fees and cross-registry movement.

ARIN's inter-RIR compatibility rule is a mandate-laundering machine because it converts regional policy preference into a condition of global mobility. ARIN does not need to claim ownership over all IPv4. It only needs to say that inter-RIR transfers involving ARIN require reciprocal compatible needs-based policies. Because ARIN is a large market, that condition influences how other regions write and maintain transfer rules. Compatibility becomes a passport. Lack of compatibility becomes isolation.

The laundering is not necessarily intentional. Institutions often preserve old assumptions because old assumptions made sense in an earlier environment. Needs-based policy was built for allocation scarcity. It survived exhaustion because the policy machinery, vocabulary and consensus habits already existed. Inter-RIR transfer then carried the assumption outward. A rule that once rationed free-pool distribution now filters capital mobility.

The problem is that the official language hides the economic effect. "Compatible needs-based policy" sounds like a technical standard. It is not merely technical. It is a view about who may acquire scarce capacity and on what terms. "Community-developed policy" sounds democratic. It may be participatory, but it is not the same as consent by all affected asset holders and downstream users. "Stewardship" sounds neutral. It can also preserve institutional discretion. "Not sold" sounds principled. It does not prevent market pricing; it merely keeps the market inside an administrative recognition frame.

The market sees through the language because it has to price outcomes. A transfer route that depends on compatibility is worth more than one that does not exist. A block subject to clear settlement is worth more than one trapped in a region without accepted transfer paths. A buyer with obvious policy need is a safer counterparty than one with complicated future demand. A legacy holder with modern service eligibility is easier to transact with than one whose status raises agreement questions. The market translates mandate into discount.

This is why ARIN's stability matters. If a visibly troubled registry imposes controls, critics can blame crisis. If a stable and well-documented registry imposes controls, the structure itself is revealed. ARIN shows that the gatekeeper problem is not limited to failure. It can exist in orderly form, with clean pages, polite meetings, visible fees and procedural fairness. That makes it more important, not less.

Market liquidity and the shadow of alternative structures

When a transfer market is too constrained, economic demand does not disappear. It moves. Some demand shifts to leasing. Some moves through corporate acquisitions rather than pure address purchase. Some is satisfied by reassignment, customer arrangements, BYOIP contracts, merger structures, service contracts, private warranties or operational workarounds. Some shifts toward IPv6, NAT, address sharing or renumbering projects. Some simply remains unmet.

This is not an argument against every alternative structure. Leasing can be a rational continuity product. Acquisitions can reflect real business combinations. IPv6 deployment is essential. NAT and sharing may be practical in some environments. The point is that restricted transferability changes the menu. If clean registry transfer is slow, uncertain or unavailable, users will find less clean ways to obtain the economic function they need. The registry may then preserve the appearance of conservation while driving activity into less transparent channels.

ARIN's qualified facilitator programme and public transfer route reduce this risk by making formal transfer possible. That is a strength. But the persistence of need tests and inter-RIR compatibility filters means the formal route is not fully market-neutral. Participants still ask whether the buyer qualifies, whether the source faces restrictions, whether the destination RIR can certify compatibility, whether fees and RSAs are in order, whether security objects can move cleanly, and whether confidential tickets will align. The formal route exists, but it is administered.

Liquidity matters because IPv4 scarcity is not evenly distributed. Some holders have more than they can productively use. Some operators need more than they can obtain from residual pools. Some regions have higher marginal demand. Some older blocks sit inside institutions whose current operations no longer match their historical allocations. Transfer is the mechanism that turns idle or lower-valued holdings into productive capacity. Every unnecessary friction reduces that mechanism's effectiveness.

The compliance-heavy market also rewards scale. A large operator can run parallel diligence, maintain relationships with facilitators, absorb failed tickets and negotiate across regions. A small operator may need a single clean block and cannot afford an unpredictable process. If policy is meant to serve operational networks, that asymmetry should worry ARIN. A market that is formally open but practically easier for the largest participants is not an efficient stewardship outcome.

The ledger-first alternative would not eliminate regulation. It would focus regulation on settlement integrity. Inter-RIR transfer should require verified source authority, recipient identity, record continuity, security-state migration, disclosure of dispute status, applicable legal compliance and destination registry capacity. It should not require a registry to forecast whether the buyer will use 50% of the block within 24 months. It should not require a policy compatibility treaty where technical record compatibility would suffice.

That would increase liquidity without turning the number system into chaos. The Internet already operates through decentralised routing decisions. Routability is not guaranteed by ARIN. Network operators decide what to route. The registry's job is to maintain a truthful registry state. A truthful state can support markets better than a discretionary one.

What accountability would look like

ARIN's published materials provide process clarity, but accountability in a post-exhaustion market requires more granular economic visibility. A transfer market needs to know not just what the rules say, but how the rules behave. How long do 8.4 transfers take by counterparty RIR? How often does need documentation create additional rounds? How many inter-RIR transfers are abandoned after submission? How many are delayed because of receiving-registry certification? How often are source disputes the decisive issue? How often does RSA execution delay closing? How often do fees or unpaid invoices block evaluation? How often do small organisations receive transfer pre-approval and then fail to complete a transaction?

Such data could be published in aggregate without exposing confidential commercial terms. It would improve market planning, reduce rumour, discipline discretion and help policy participants see the economic effects of their rules. If needs assessment is truly low-friction and proportionate, aggregate data would show it. If it is a liquidity burden, the data would show that too. Accountability is not hostility to the registry. It is a condition of trust in a settlement layer.

Resource review deserves similar transparency. NRPM 12 allows ARIN to review current usage of resources maintained in its database when new resources are requested, when fraud or policy violation is suspected, when reassignment or reallocation noncompliance is suspected, or at any other time without having to establish cause unless a full review has been completed in the preceding 24 months. It permits ARIN to request or require return of resources when material noncompliance is found, and to revoke ARIN-issued resources if voluntary return does not occur, with safeguards around judgment, aggregation and timing.

Those powers may be necessary in some cases. Fraudulent procurement and false records damage the ledger. But resource review is economically significant after exhaustion. It can affect asset confidence, transaction warranties and buyer diligence. A market participant does not need to believe ARIN will abuse review power to price the fact that review power exists. Aggregate reporting on review triggers, duration, outcomes and appeals would make the risk less opaque.

The same principle applies to policy development. ARIN's PDP is open, but openness does not measure economic exposure. A transfer rule may receive strong support among active participants while imposing costs on absent parties. Policy assessments should therefore include transaction-cost analysis: expected effects on liquidity, small-operator burden, cross-registry settlement, legacy resources, compliance expenses and incentives for alternative structures. Technical soundness should not be limited to conservation, aggregation and registration. In a post-exhaustion IPv4 market, market design is technical infrastructure.

ARIN has the institutional maturity to do this. It has public documentation, experienced staff, a visible governance structure and a large enough market to produce meaningful aggregate signals. The question is whether it will treat economic accountability as part of registry accountability, rather than as an external complaint by buyers and sellers.

A narrower stewardship model

A better ARIN model would separate four functions that are currently entangled. The first is registry truth: uniqueness, holder recognition, contacts, reverse DNS, RDAP and Whois, routing-security publication and transfer history. The second is fraud and dispute control: authority verification, forged-document rejection, corporate succession, court orders, dispute notations and preservation of last verified state. The third is residual allocation: waiting-list space, reserved pools and any future returned-address distribution. The fourth is market settlement: recognising private transfers of already allocated resources.

Needs assessment belongs primarily in the third category. It is a rationing tool for common-pool distribution. It should not dominate the fourth. Market settlement should be governed by objective record and authority tests. If a buyer is real, the seller is valid, the resource is not blocked by a narrow dispute or legal order, and the destination registry can maintain the record, recognition should be routine. The registry can require accurate contact data, security-state transition and service agreement terms proportionate to the services actually provided. It need not approve the buyer's business model.

Inter-RIR compatibility should likewise shift from policy ideology to operational interoperability. The necessary questions are practical: can the receiving registry publish accurate data; can the source registry remove or redirect its record without duplicate recognition; can RPKI, IRR and reverse DNS transitions be handled; can disputes be recorded; can historical transfer records remain auditable; can legal orders be honoured where properly applicable? These are hard questions, but they are ledger questions. They do not require reciprocal needs-based ideology.

Legacy resources should remain a standing reminder that registry functions can be unbundled. Core recognition and publication are public-reliance functions. Advanced services can carry terms, but those terms should not become coercive paths to broader control. As routing-security expectations rise, ARIN should consider whether some security publication should be treated as part of ledger continuity rather than as a premium boundary for contract leverage.

Member power should remain, but it should be supplemented by economic accountability. Policy participants should see transfer-cost data. Small operators should have visible metrics. Cross-registry delays should be measured. Failed and abandoned requests should be studied. The fact that a rule is community-developed should not end the analysis; it should begin the obligation to show what the rule does.

This is not deregulation. It is better regulation. A ledger-first registry would still reject fraud, preserve records, coordinate with other registries, maintain fees for real services, process disputes carefully and defend uniqueness. It would simply stop treating post-exhaustion transfers as if they were free-pool allocations in disguise.

The wider lesson

ARIN's inter-RIR transfer politics matters because it shows the RIR system at a crossroads. IPv4 scarcity has created a global asset whose settlement remains regional. The official vocabulary still says stewardship, community, conservation and need. The market says liquidity, compliance burden, title certainty, option value, contract exposure and cross-border mobility. Both vocabularies describe something real. The danger is pretending that the first cancels the second.

ARIN demonstrates that the RIR system can be orderly and still economically interventionist. It can allow transfers and still condition them on rationing logic. It can publish compatibility lists and still turn regional policy into a market border. It can preserve legacy records and still use modern services as agreement leverage. It can invite public participation and still fall short of full asset-owner consent. It can call itself a steward and still act as a gatekeeper.

The strongest defence of ARIN is that the registry function is real. It is. The Internet needs unique number records. It needs accurate contacts. It needs transfer history. It needs reverse DNS and routing-security continuity. It needs a way to prevent forged movement and duplicate claims. It needs dispute handling that does not destroy running networks. The strongest criticism is that none of those functions requires a registry to preserve allocation-era control over global asset mobility.

Cross-registry transfer rules expose the contradiction because they force the question into the open. Is an RIR a regional bookkeeper for a global asset, or a regional authority deciding whether that asset may move? If it is the former, compatibility should mean interoperable records and clean settlement. If it is the latter, compatibility will continue to mean policy conformity and needs-based permission. ARIN's current architecture contains both models. That is why it is such a revealing test case.

The economics point in one direction. Scarce assets need trustworthy ledgers, not discretionary chokepoints. Market liquidity needs clear rules, not moralised rationing. Regional stewardship needs humility, not territorial capital control. Member power needs evidence of economic effects, not only process legitimacy. Legacy resources need service clarity, not quiet contract migration. Inter-RIR transfers need portability, not policy diplomacy dressed as compatibility.

ARIN is not the only registry facing this transition. It is the clearest North American version of it. The institution can continue to be useful if it narrows its authority to the work the Internet actually requires: preserve the ledger, publish accurate records, verify authority, isolate disputes, maintain security continuity and recognise legitimate movement. If it keeps treating the transfer market as an extension of allocation-era stewardship, it will remain a gatekeeper over capital while calling itself a steward of numbers.

That is the political economy of inter-RIR transfers. The issue is not whether ARIN has rules. It does. The issue is what kind of market those rules create. In the IPv4 era after exhaustion, the answer will determine whether number resources remain trapped inside regional administrative ideology or become portable infrastructure assets whose records can move as efficiently as the networks that depend on them.