Summary

  • APNIC-recognised IPv4 holdings give incumbents real options: they can wait, reserve, reassign, lease, sell, migrate, pledge continuity, or use scarce public reachability as insurance around customers, cloud moves, M&A, financing and network change.
  • That option value is not the same as hoarding and not the same as balance-sheet capitalisation. It is the value of choices kept open before management decides which path to exercise.
  • APNIC's official policy materials create the operating tension: resources are licensed rather than owned, transfer conditions and fee rules make movement visible, NIR arrangements create layered recognition, and review activity can protect the ledger or narrow the holder's choice set.
  • A disciplined APNIC role would preserve uniqueness, public registration, transfer recording, account accuracy and dispute clarity while avoiding discretionary judgement about whether an incumbent's reserve, lease book, migration pace or customer-continuity buffer is commercially virtuous.

The Board Pack With Too Many Doors

The board pack is not titled "IPv4 strategy". It sits under a more ordinary heading: network continuity and commercial options. The first page is a map of access regions, mobile packet-core growth, enterprise customers, ageing hosting platforms, cloud-migration projects and regulatory service obligations. The second page is a table of public IPv4 holdings recognised in the APNIC system, with enough internal notes to make the numbers intelligible to finance, engineering, legal and sales at the same time.

That table changes the meeting. If the operator did not have recognised IPv4, every growth decision would begin with purchase, lease, translation or denial. With recognised holdings, management has choices. It can keep part of the block unrouted as reserve for a public-sector customer. It can reassign addresses from a declining legacy platform into a higher-margin enterprise service. It can lease some capacity while testing demand in a new market. It can sell a clean block if a capital project needs cash more than future address slack. It can migrate a customer group into cloud while preserving public addresses that are already trusted by banks, payment processors and suppliers. It can delay renumbering until a product cycle gives customers a reason to accept disruption. It can show a lender that continuity does not depend entirely on buying addresses at the worst possible moment.

This is incumbent optionality. The word matters. It is not a moral defence of waste. It is not a claim that a registry line is a perfect property deed. It is not an accounting essay about whether an asset should be measured at cost, fair value or impairment. It is narrower and more practical. A recognised IPv4 holding gives an established operator the right, or at least the practical ability, to decide later. The option may expire if policy changes, if records become defective, if fees are not paid, if a dispute arises, if the holder sells, if customers move, or if APNIC review narrows the range of recognised uses. But while the option remains credible, it changes management's bargaining position.

The operator does not need to exercise every choice today. That is the value. Scarce capacity does not have to be consumed immediately to be useful. A spare suballocation can be customer insurance. A clean prefix can be migration collateral. A routable block can be an exit path from an unfavourable cloud architecture. A reserve can be a promise that a high-value customer will not be placed behind shared translation at launch. A transfer-ready file can make a divestment easier. A registry-recognised position can let the operator say no to a supplier, no to a hurried sale, no to a badly priced lease, or no to a migration schedule set by someone else.

In an abundant world, those choices would be routine. In the Asia-Pacific IPv4 economy, they are increasingly strategic. APNIC's region contains large incumbents with old allocations, National Internet Registry layers, fast-growth markets, cloud and content demand, public-service dependency, dense urban broadband, island networks, under-connected rural areas and companies that arrived after the free-pool era was effectively over. In that mix, recognised IPv4 is not merely a technical input. It is a management option under registry recognition.

Recognition Is The Option Surface

The option does not come from a private spreadsheet. It comes from a recognised record that other parties can rely on. APNIC's policy materials describe the registry's core goals in terms of uniqueness and registration: every assignment and allocation should be globally unique, and directly made assignments and allocations are to be registered in a public database to support troubleshooting. The policy also defines a transfer as the reallocation of current address blocks, or the reallocation of historical resources claimed and transferred to an APNIC account holder. The source is the organisation that was the legitimate holder before the transfer; the recipient is the organisation that will be the legitimate holder after it.

That language is dry, but it is where the option surface is formed. An incumbent's address holding is valuable partly because counterparties can see who APNIC recognises, who must be contacted, which resources are subject to current policy, and which transfers have been processed and recorded. The registry record does not install routers. It does not guarantee global routability. APNIC's own policy says routability cannot be guaranteed by any single account holder. Yet the record helps other actors decide whether a block can be accepted, financed, transferred, migrated, filtered, placed into a cloud onboarding file, or attached to customer service continuity.

The same record also defines the boundaries. APNIC policy says Internet number resources are licensed for use rather than owned, and that delegation or registration does not confer ownership. It treats account holders as custodians rather than owners and says they are not entitled to sell or otherwise transfer resources outside the provisions of the policy document. Licences are normally renewed annually. Renewal depends on the original basis of delegation remaining valid and address space being properly registered. The membership agreement then adds the contractual surface: the agreement is effective for one year, renewal accepts the agreement as it exists at renewal, APNIC documents can change and bind the member, fees must be paid, and rights including delegated resources can be revoked through stated mechanisms.

This is why optionality in APNIC space is neither simple ownership nor mere paperwork. It is recognised control under a continuing service and policy relationship. The incumbent cannot rationally treat a block as a risk-free commodity. Nor should an observer pretend the block has no economic force because the policy document avoids property language. The market already behaves as if recognised IPv4 matters. The registry record is the public layer that turns a private technical fact into a usable option.

The Asia-Pacific incumbent with recognised holdings therefore owns something more subtle than an unrestricted asset and more valuable than a database label. It has a set of choices that third parties will take seriously as long as the record is credible. The strategic question is how those choices are preserved, priced and limited.

Scarcity Makes Waiting Valuable

An option is a right to decide later. Scarcity makes that right valuable because the future purchase price, supply condition and operational need are uncertain. An incumbent with a recognised block can wait while a new customer segment is tested. It can keep addresses for a product whose launch depends on licensing, spectrum, procurement or data-centre availability. It can delay a transfer until market terms improve. It can hold back capacity from a sale because the avoided future purchase may be worth more than the cash received today.

This is not an exotic financial theory. It is how operators already think about routers, ducts, towers, spectrum, racks, submarine capacity and spare power. A network with no spare capacity is efficient only until demand arrives. Then it is fragile. IPv4 is harsher because public addresses cannot be manufactured by capex. A company can order fibre, buy servers, lease floorspace or add batteries. It cannot produce new globally unique IPv4. It can only obtain recognition of existing resources, source a transfer, enter a lease, compress customers through translation, redesign services, or make customers accept a worse product.

APNIC's remaining-pool policy makes this visible. The current minimum IPv4 delegation is a /24, and each account holder is eligible to receive only up to a /23 from the APNIC 103/8 pool. Recovered non-103/8 resources are treated together with 103/8 addresses for remaining-pool delegation, with a waiting list to be created once all IPv4 addresses run out. Initial LIRs must demonstrate criteria such as prior use or immediate need, and a detailed plan for use of at least a /23 within a year. Subsequent delegations depend on verified usage rate, documented plans and compliance, and the eighty percent rule asks that customer delegations account for at least eighty percent of total held address space before another delegation.

Those rules are not the centre of this article, but they explain why an old holding behaves differently from a fresh request. An incumbent that already has recognised IPv4 holds a timing instrument. It does not need to enter the remaining-pool queue for every customer contingency. It does not need to buy at the same moment that a contract is won. It does not need to reveal every commercial uncertainty to prove immediate need. It can decide whether to consume, preserve, convert, lease, sell or move capacity when the business case is clearer.

Waiting can be abused. A holder can hide a false record, misstate use, block market movement or treat the registry as cover for dead inventory. Those are record and conduct problems. They do not make waiting itself illegitimate. In a scarce market, the ability to wait is often the difference between disciplined investment and forced reaction. The policy danger is confusing prudent optionality with stockpiling simply because both can look like unused space on a rough utilisation chart.

Reserve Is Not Automatically Waste

APNIC's policy framework carries conservation language from the allocation era. It says address space should be distributed according to actual need and immediate use, and that stockpiling and maintaining reservations are contrary to the conservation goal. It also says account holders are custodians, not owners. Those statements made administrative sense when a registry was distributing addresses from a common free pool. They become more complicated after exhaustion, when recognised holdings are already embedded in operating companies, customer promises, financing files and secondary-market choices.

An incumbent reserve can be waste. It can also be insurance. A mobile operator may need public IPv4 for enterprise APNs, fixed-wireless customers, roaming arrangements, fraud-sensitive services and lawful-request precision. A data-centre operator may need addresses for customers that cannot accept shared translation. A national carrier may need a reserve for emergency services, banks, government platforms and regulated wholesale customers. An island operator may need spare public addresses to keep a backup path usable when a submarine route fails. A cloud-adjacent provider may need clean blocks for bring-your-own-address use or for services that have accumulated reputation around existing public identity.

The economic point is that reserve has function even before full utilisation. A customer-continuity buffer may sit idle until a migration weekend. A disaster-recovery block may route only when a primary site fails. A reputationally clean prefix may be saved for a product where blacklisting would destroy trust at launch. A block kept for enterprise segmentation may not appear efficient under a crude average-use measure, but it may be what allows the operator to win or retain high-value customers without placing them behind shared address translation.

The official policy language itself contains a clue. APNIC recognises that minimised overhead is a resource-management goal, and that repeated small successive expansions can create overhead compared with fewer larger expansions. That logic does not vanish when the free pool ends. A company that holds enough address space to avoid repeated emergency sourcing is not automatically doing something antisocial. It may be reducing the transaction overhead, support burden and customer risk that scarcity otherwise imposes.

The APNIC boundary should therefore be practical. The registry may ask whether records are accurate, whether contacts work, whether the holder is real, whether a transfer source is the currently registered holder, whether there is a dispute, and whether policies that apply to new requests are being followed. It should be careful about treating every reserve as failed virtue. In a scarce address economy, reserve is one of the ways an incumbent keeps promises in an uncertain future.

Leasing Converts An Option Into Yield

Leasing is where incumbent optionality becomes most visible and most politically sensitive. A holder with recognised IPv4 can use the addresses itself, keep them as reserve, sell them, or allow another operator to use them under a contract. The public policy vocabulary is often uncomfortable with that last option because it sounds like monetising scarcity. But the operational alternative is not abundance. The alternative may be idle recognised capacity on one side and a network forced into poorer service, expensive purchase, or opaque arrangements on the other.

Heng Lu's doctrine is direct on this point: scarcity is not hoarding, and registry bodies should not police leasing as if commercial structure were a violation of the Internet. In this frame, IPv4 is a productive input and, after exhaustion, an internal capital asset for the networks that hold it. Leasing can turn dormant capacity into usable connectivity while preserving the holder's future option to reclaim, sell, pledge or use the block later. It can also shift registry-layer risk depending on who holds the recognised relationship and who carries the operating use.

The public APNIC materials do not set out a general leasing code in the same way they define transfers. That absence is itself part of the option. If an incumbent leases capacity without changing the recognised holder, the transaction may sit largely in private contract, routing practice, customer assignment, reputation management and operational disclosure rather than in a full registry transfer. That can be efficient, but it can also create opacity. The registry record still names the holder. Abuse complaints, route reputation, reverse-DNS expectations, RPKI arrangements, customer disputes and law-enforcement requests may still flow toward the recognised party or through the lessor's operating structure.

The option is therefore not free money. A serious incumbent must ask whether a lessee will damage route reputation, whether subdelegations are documented, whether customer data is protected, whether contract remedies are enforceable, whether the lease conflicts with APNIC documents, and whether the arrangement can be unwound if the holder needs the addresses for its own customers. It must decide whether yield today is worth reducing future freedom.

That is exactly why leasing belongs in an optionality article rather than an asset-capitalisation article. The lease value is not merely a number on a valuation schedule. It is the choice to convert unused capacity into current income while retaining some future control. If APNIC recognition remains stable and review remains bounded, that option can improve address circulation. If registry discretion expands into commercial-structure policing, the lease option becomes less reliable, and holders respond by keeping more capacity private or moving through less transparent channels.

Sale Is Exercise, Not The Whole Story

Selling a block is the most obvious way to monetise recognised IPv4, but it is only one exercise of a broader option. An incumbent that sells gives up future capacity in exchange for cash and risk reduction. It may do so because a business line is shrinking, because cloud migration releases public addresses, because an acquisition leaves duplicate ranges, because debt service matters more than reserve, or because the market price is attractive. It may refuse to sell because the block supports customer continuity, future expansion, credit capacity, reputation, or bargaining power.

APNIC's transfer materials make the sale path administratively visible. The transfer page describes a transfer as movement of IP addresses or AS numbers from a source legal entity to a recipient legal entity. It identifies three permitted transfer types: merger-acquisition, historical Internet number resources, and unused IPv4 or AS numbers. It says APNIC processes transfer requests according to policy and updates the Whois Database to reflect transfer results. It also warns that an APNIC account is generally needed, supporting information must be provided, requests may be delayed if information is not supplied, and transfer conditions and fees may apply.

The policy details are more specific. For IPv4 transfers within the APNIC region, the source must be the currently registered holder and not involved in any dispute over the resources. Recipients are subject to current APNIC policies. Recipients that do not already hold IPv4 must show a detailed plan for use within twenty-four months; recipients that already hold IPv4 must also show past usage rate and evidence of compliance with APNIC policies for past delegations. Inter-RIR transfers add counterpart-RIR compatibility and a minimum /24 size, and APNIC records the transfer only when the counterpart RIR has an inter-RIR policy that permits the movement. Addresses delegated from the 103/8 free pool cannot be transferred for five years after original delegation.

Those conditions mean a sale option is not simply "find buyer, sign contract, receive money". It is a managed exercise through a registry-recognised channel. That has benefits. It reduces false-holder risk, gives counterparties a public record, helps maintain uniqueness and makes later diligence easier. It also affects option value. A holder that can sell quickly and predictably has a stronger option than a holder whose transfer may be slowed by uncertainty, recipient review, fee calculation, NIR coordination or dispute status.

This is why the incumbent's internal board pack treats sale as one door among several. Selling may be rational, but sale is irreversible relative to reserve. The question is not whether holding is morally suspect or sale is morally pure. The question is when exercise is worth more than keeping the option alive.

Reassignment Is The Quiet Option

The least dramatic option may be the most common: reassign addresses inside the incumbent's own customer base. A recognised holder with address slack can move capacity from an old hosting product to a managed security service, from a legacy broadband tier to enterprise fixed wireless, from a retired internal platform to a new data-centre customer, or from low-value generic use to a customer with strict public-reachability needs. No public sale occurs. No acquisition headline appears. Yet economic value moves.

APNIC policy distinguishes allocated address space, assigned address space and end users. Allocated space is distributed for subsequent distribution; assigned space is delegated to an LIR or end user for exclusive use within the infrastructure they operate. The registration goal allows account holders that receive allocations to choose whether customer assignment registrations should be publicly available, subject to privacy considerations and applicable law. If customer assignment records are hidden, Whois queries return allocation details.

That structure gives the incumbent room to manage its customer portfolio. It can protect privacy, segment customers, aggregate routing where possible and shift internal address plans as products evolve. The option lies in not needing to acquire every time a customer category changes. A company with recognised holdings can match scarce public reachability to the customers that value it most, while moving less sensitive customers to private addressing, IPv6-first service, CGNAT, shared hosting, proxy architectures or managed platform models.

This can improve efficiency. It can also create market power. If the incumbent can reserve public addresses for premium customers while smaller rivals must lease or buy at high prices, public IPv4 becomes a product differentiator. The incumbent may not need to say "we have a scarcity advantage". It can price static public addresses, faster onboarding, cleaner reputation, direct reachability, cloud portability or regulated-customer continuity into service tiers. The option is embedded in product design.

APNIC should not pretend this does not exist. Nor should it try to set the product mix. The registry's proper interest is whether the allocation and assignment records remain accurate enough for uniqueness, contactability, troubleshooting and policy compliance. Once it attempts to decide which customer class deserves public addresses, it stops being a ledger and starts acting like an industrial planner. In an incumbent-optionality frame, the discipline is to see the commercial power clearly without making the registry the pricing authority for that power.

Migration Turns Address Stock Into Patience

Network migration is rarely an engineering event alone. It is a customer-support event, a contract event, a security event, a procurement event and sometimes a public-service event. Public IPv4 holdings give incumbents patience during migration. They can move customers in stages, keep old allowlists alive, operate parallel platforms, test a cloud cutover, preserve a trusted source address during a fraud-monitoring transition, or give enterprise customers a long renumbering window.

APNIC's own materials remind readers that address space is not globally routable by guarantee and that small portable assignments may face filtering realities. That makes continuity more dependent on recognised, reputation-bearing, operationally accepted address space. A new block can be clean on paper and still slow in practice if counterparties need to update filters, allowlists, reverse-DNS expectations, abuse contacts, geolocation data, route-origin records and internal risk systems. An old recognised block can carry trust because it has already been seen by the world.

For an incumbent, that trust creates migration optionality. The operator can keep a legacy block attached to a banking customer while moving the rest of a platform. It can bring a prefix into a cloud environment where the provider supports customer-supplied address ranges, preserving public identity during infrastructure change. It can avoid forcing a hospital, port, government agency or payment processor to renumber in the same week as an application change. It can turn address continuity into a feature: not faster bandwidth, but fewer breakages.

This is not the same as cloud NAT and platform power, the neighbouring topic in the round. The question here is not how cloud providers price or control public address products. The question is how an incumbent's own recognised address stock gives it a cloud outside option. If a platform's public IPv4 pricing, NAT design or exit rules are unattractive, the holder of recognised space has more room to negotiate, bring addresses, stage migration or preserve identity across platforms. The address option reduces supplier dependency.

The registry layer still matters. If records are stale, if transfer history is unclear, if the holder's APNIC standing is weak, if NIR evidence is hard to reconcile, or if a review casts doubt on the recognised status of a block, migration patience shrinks. The board pack then reads differently. A block that looked like a continuity option becomes a question mark. The incumbent may still route it, but the customer and the lender may not price it the same way.

Pledge Means Continuity, Not Simple Collateral

The word "pledge" can mislead. This article is not about the detailed law of collateral. That is a later, narrower question. Here the point is simpler: an incumbent can use recognised IPv4 to make credible commitments. It can pledge continuity to a lender financing a data-centre upgrade. It can pledge public-address availability to an enterprise customer. It can pledge recoverability in an acquisition discussion. It can pledge that a service line will not depend entirely on a thin lease or on shared translation. The pledge may not be a perfected security interest, but it still changes confidence.

APNIC recognition is central to that confidence. A bank may not know BGP from a billing system, but it can understand a public registry record, an account standing file, a transfer history, a fee schedule, a list of clean prefixes, a route-origin practice and a documented customer assignment policy. A procurement team may not value every address, but it can ask whether the supplier has enough public reachability to meet the service requirement. A buyer may not want to litigate ownership theory, but it can ask whether the seller's recognised resources can continue after closing.

The APNIC Membership Agreement makes that confidence conditional rather than absolute. The member must pay fees, avoid false or misleading information, update material information, comply with APNIC documents and respond to notices. If APNIC reasonably believes there is a breach, it must send a notice describing the breach, remedy path and intended action. If the breach is not remedied, APNIC may send a further notice or revoke rights including delegated resources and terminate the agreement; the member has an appeal route to the Executive Council. A member receiving certain notices must immediately cease using the specified resources, subject to possible court remedies. The agreement also excludes APNIC liability to the extent permitted by law for matters connected to the agreement, APNIC documents or delegated resources.

That contractual surface does not destroy option value. It defines the risk around it. A serious incumbent cannot pledge continuity by saying "we have addresses" and stopping there. It must show that registry recognition, account standing, internal controls, fee payment, contact accuracy, transfer history, NIR files and customer assignment practices are all boring. Boring is a compliment in this context. The more boring the registry file, the more credible the option.

This also explains why a registry should avoid becoming a capital allocator. The holder and its counterparties can price pledge strength through diligence. APNIC's comparative advantage is not to decide how much a lender should value a block. It is to maintain the record and process boundaries that make diligence possible.

NIR Layers Add Local Knowledge And Layered Finality Risk

APNIC's National Internet Registry layer gives incumbent optionality a regional character. The operational policy for NIRs says APNIC provides for NIRs within economies of the region to improve allocation and registration services in local language and culture. It also says the NIR structure historically added complexity to APNIC's ability to ensure efficient resource utilisation, and that NIRs must implement applicable APNIC address-management policies while any additional local policies must not conflict with regional or global rules.

For incumbents, the upside is obvious. Local registry service can lower the cost of maintaining records, explaining documents, understanding customer evidence and translating operational practice into accepted registry form. A Japanese, Korean, Vietnamese, Indonesian, Taiwanese, Indian or Chinese holder may be better served by local-language registry interaction than by a single regional interface. Local knowledge can reduce accidental friction and help distinguish normal domestic business records from suspicious or incomplete files.

The option value rises when the NIR layer is trusted. An incumbent can hold reserve, reassign customers, prepare transfers, document history and support migration with a local evidence base that counterparties can understand. If APNIC and the NIR agree on recognition, the holder's option surface is wider.

The risk is finality. The same policy says APNIC keeps allocation windows for NIRs. Within a window, the NIR can send APNIC an allocation request that includes registration information but not justification; the NIR must keep justification permanently. Larger requests require a second-opinion request with full justification and a summary of the NIR's evaluation. If APNIC disagrees, it may ask for more information from the NIR and possibly for more information to be collected from the applicant.

That design is workable, but it means historical option value may depend on files held at two levels. A mature incumbent may have satisfied its NIR years earlier, only to need APNIC-level clarity later for a transfer, financing discussion, review or merger. If the record is smooth, optionality survives. If APNIC and the NIR differ, or if old justification is difficult to retrieve, the option narrows. The holder's ability to sell, lease confidently, pledge continuity or migrate under time pressure becomes a layered-recognition problem.

This is not a reason to attack the NIR model. It is a reason to define finality carefully. If NIR recognition supports APNIC's registry facts, incumbents should not have to relitigate settled commercial history each time they exercise an option. The public ledger must be accurate. But optionality requires a point at which a recognised file can be relied on.

Review Can Protect Or Shrink The Option

APNIC's Resource Delegation Review Program makes this tension current. APNIC says it began planned reviews across the APNIC registry and NIRs at the end of 2023 to ensure registry accuracy. Following preliminary investigations, the program expanded in 2025 to include steps intended to strengthen policy compliance and registry integrity. Its activities include aggregated analysis of APNIC and NIR allocation and transfer data, policy-compliance spot checks, review of APNIC delegation activity, account-accuracy checks, NIR support and training, and NIR agreement review.

The July 2026 APNIC Blog update gives the latest concrete status. APNIC reported that the primary activity is reviewing all IPv4 delegations and transfers made by each NIR and APNIC over a ten-year period. It said data analysis of TWNIC and KRNIC delegations and transfers had been completed, with minor queries and clarifications resolved, joining JPNIC as completed. Work continued with VNNIC and IDNIC; initial data reviews for IRINN and CNNIC were in progress; APNIC's own registry data analysis was expected to begin in the third quarter of 2026.

For incumbent optionality, that review can be beneficial. It can clean records before a sale. It can resolve old NIR inconsistencies before a lender asks. It can reduce uncertainty over transfers made during the years when IPv4 scarcity became economically material. It can improve account contacts and make future migration or customer continuity easier. A well-run review raises option value because counterparties trust the record.

The same review can shrink option value if it becomes retrospective business-plan inspection. A ten-year lookback across delegations and transfers touches precisely the period in which incumbents learned to treat IPv4 as scarce capacity. Some reserves were built for migration. Some transfers were made for customer continuity. Some address books were rationalised after consolidation. Some NIR-held files may reflect local practice rather than the style of an APNIC reviewer reading them years later. If review remedies are bounded to accuracy, finality and fraud prevention, the option surface becomes cleaner. If remedies reopen commercial judgement, the option surface becomes smaller.

This is the difference between ledger discipline and capital control. Ledger discipline says: prove the holder, fix the contact, remove contradiction, record the transfer, clarify the dispute, maintain the evidence. Capital control says: explain why your reserve is acceptable, why your lease book is morally tolerable, why your migration is fast enough, why your customer geography deserves public reachability, why your internal reassignments fit an outsider's view of need.

An incumbent can survive more review than an entrant. That does not make review harmless. If review is unpredictable, incumbents will treat addresses as less liquid, less pledgeable and less safely leasable. They may hold more reserve, not less, because exercising options becomes harder.

Fee Standing Is A Carrying-Cost Signal

Option value has carrying cost. APNIC's 2026 Member Fee Schedule makes that visible in the official record. It has a sign-up fee, annual fee, transfer fee, ASN fees, temporary assignment fees and a reactivation fee for terminated accounts. For annual fees, APNIC calculates the fee from the number of address bits held, with IPv4 and IPv6 holdings assessed separately and the larger amount determining the annual fee. Transfers are charged at twenty percent of the annual fee applicable to the resources being transferred, paid by the recipient except for transfers to other RIRs, where the source APNIC member pays; specified cases such as initial IPv4 transfers to members holding no IP addresses, transfers to NIR members and membership changes to or from an NIR are excluded.

That fee surface matters because it prices one part of holding the option. An incumbent keeping a large IPv4 position is not holding a completely costless instrument. It pays annual fees, carries internal administration, maintains contacts, responds to review, manages customer assignment data and preserves routing and reputation hygiene. If it transfers, a fee may attach. If membership is not renewed, the membership agreement gives APNIC revocation paths.

Carrying cost can discipline idle holding without requiring moral accusation. If a block has no internal value, no lease value, no sale value, no continuity value and no reserve value, the holder has reason to sell, lease or release it. If the holder keeps paying, that is evidence that management believes the option is worth more alive than exercised. It may be wrong, but it is not automatically antisocial.

The cleaner fee principle is dull and important: fees should keep the registry service sustainable and predictable, not become a hidden lever to punish reserve, discourage leasing or favour one kind of customer over another. Holders should decide how to use the options they carry.

Incumbency Advantage Is Real, But This Is Not The Entrant Story

A different analysis would begin with what a new operator lacks: inherited inventory, registry history, proof archives, routable reputation, upstream trust, financeable customer evidence and time. Incumbent optionality begins with what an established holder can choose to do before any external transaction is forced.

The distinction matters because incumbency advantage can be real without every incumbent action being suspect. A mature operator with recognised IPv4 can stage customer launches, decide which product tier receives public reachability, use old blocks as migration cushions, hold clean addresses for regulated accounts, lease capacity, sell at a chosen time, or wait. Those choices make it harder for a newcomer to compete. But the existence of advantage is not proof that the holder is hoarding. It may be proof that scarcity has turned old operational capacity into strategic flexibility.

There is a policy temptation to respond to this advantage by making the registry more interventionist. If incumbents have options, perhaps APNIC should narrow them. If reserve creates bargaining power, perhaps reserve should be challenged. If leasing earns yield, perhaps leasing should be discouraged. If transfer timing creates market advantage, perhaps transfer discretion should be limited. That temptation is understandable and dangerous.

The registry cannot erase historical allocation by pretending to equalise all future choices. Attempts to do so usually produce discretion. Discretion favours the parties best able to navigate procedure, document need, lobby policy rooms and absorb delay. In many cases that is the incumbent again. A rule designed to weaken incumbent optionality may therefore strengthen the largest incumbents while hurting smaller holders, fast-growing regional networks and less procedurally fluent operators.

A better approach is transparency, portability of evidence, accurate records, bounded review, clear transfer paths, visible fees, reliable NIR finality and low-friction correction. Those measures do not eliminate incumbent advantage. They make the advantage legible and reduce avoidable registry friction for everyone else. The market can then see what incumbents hold, what they sell, what they lease, what they reserve and what risks attach to their records.

Incumbent optionality is not a slogan of fairness. It is a description of power created by scarce recognised capacity. The editorial task is to describe that power without turning the registry into the instrument for redistributing it by administrative instinct.

Asset Capitalisation Is The Next Door, Not This Room

A separate asset-capitalisation analysis would ask how market value changes the meaning of a registry entry for boards, auditors, buyers, lenders and balance sheets. This article stops one step earlier. It is about the holder's choice set before value is formally recognised, measured, financed or impaired.

That boundary is easy to lose because optionality and capitalisation touch. A recognised IPv4 block has value partly because it can be sold, leased, pledged, reserved or used. But the conceptual order matters. The option comes first. The valuation follows. A network manager can know that a block is strategically precious before an accountant decides how it appears in a report.

Keeping the boundary clear also avoids property-law absolutism. APNIC policy says resources are licensed, not owned. The membership agreement creates continuing obligations and revocation paths. Those facts complicate capitalisation. They do not erase optionality. A licence, concession, capacity right or long-term customer contract can create choices without being fee-simple land. The question here is what choices the recognised holder can credibly exercise and how stable the recognition surface is.

APNIC recognition makes the choices visible enough for operators to plan around them. The registry should make those choices safer by keeping the ledger reliable.

The Ledger Must Not Price The Option

The strongest APNIC role is narrow. It should protect uniqueness. It should maintain accurate public registration within privacy and law. It should process and record transfers through clear rules. It should support account accuracy, NIR alignment and dispute clarity. It should ensure that source and recipient identities are not fictional. It should make the registry useful enough that a customer, buyer, lender, upstream, cloud provider or regulator can understand who is responsible for the resource.

That is already significant power. It becomes dangerous when the registry also tries to price the option. Pricing does not always mean setting a number. It can mean delaying transfers until a business plan feels acceptable. It can mean treating lease structures as suspicious because they monetise scarcity. It can mean interpreting reserve capacity as misconduct. It can mean deciding that customer geography, product tier, cloud strategy or migration pace is the registry's concern. It can mean allowing old conservation language to override current market reality.

Heng Lu's thin-coordination doctrine is useful here because it separates running-code needs from institutional ambition. The registry layer needs uniqueness, proof of control, registry accuracy, security assertions, transfer records, auditability and replacement paths. It does not need to be the judge of a telecom board's capital plan. It does not need to decide whether an incumbent should lease, sell, hold, reassign or reserve a block. It does not need to punish scarcity because scarcity makes some actors uncomfortable.

The counterargument is that without strong registry intervention, incumbents will sit on valuable holdings. Some will. But the cure is not to make APNIC a capital-control authority. The cure is to reduce friction around legitimate movement, make records reliable, support transparent transfers, avoid needless uncertainty, let leases and sales be priced by the parties that bear risk, and keep review targeted at false records, disputes and policy-defined conditions. Liquidity moves resources better than suspicion does.

The ledger must therefore remain a ledger. A ledger that is accurate can support options without blessing every exercise. A ledger that is predictable can make incumbents more willing to lease or sell because they trust the execution path. A ledger that turns into a discretionary gate can make incumbents hoard more, not less, because every exercise becomes an opportunity for loss of control.

This is the central APNIC governance issue inside incumbent optionality. Recognition creates power. But the institution that recognises should not confuse recognition with ownership of the choices that follow.

What A Disciplined APNIC Option Regime Would Look Like

A disciplined option regime would be boring in public and consequential in practice. Holder identity, contacts, organisation names, transfer history, NIR context, dispute flags, fee standing and relevant resource status would be easy to verify. Customer-assignment privacy would be respected, but accountability would not disappear. Historical resources, M&A changes and inter-RIR transfers would have known documentation expectations.

It would separate free-pool allocation logic from already-recognised option management. If APNIC is delegating from a remaining pool, demonstrated need, slow start, immediate use and eighty percent rules have a rationing purpose. If a recognised holder is deciding whether to hold, lease, transfer, reassign, migrate, pledge continuity or reserve, the registry's question should narrow to record integrity, applicable transfer conditions, dispute status, account standing and policy-defined obligations.

For NIR-held resources, local records should be usable evidence, not a source of endless reopening. APNIC may need to review and reconcile NIR data, especially in the current ten-year resource-delegation review. But once a file is corrected or confirmed, the holder and counterparties should be able to rely on it. Option value depends on the ability to say: this record is settled enough for business.

For leasing, APNIC would resist two bad extremes. It would not pretend leasing does not exist. It would not become the commercial police for leasing either. The relevant registry concerns are contactability, record accuracy, abuse-handling clarity, route-security consistency, customer-assignment documentation where policy requires it, and the avoidance of false holder claims. The yield, price, customer segment and reserve strategy are for the parties bearing the economics.

For transfers and reviews, APNIC would keep execution predictable. It would verify source, recipient, dispute status, required plans and policy compliance where policy asks for them, while avoiding the conversion of a twenty-four-month use plan into an investment committee. Review should make the option surface cleaner, not make recognised holdings feel less reliable simply because they are valuable.

The Incumbent's Real Asset Is Time

The board meeting ends without a single heroic decision. That is usually how optionality works. The operator does not sell the entire block. It does not lease everything. It does not renumber all customers. It does not move every service to cloud. It does not freeze its holdings forever. It approves a smaller set of actions: reserve one range for a government-service migration, lease a limited block under a tighter reputation covenant, prepare one sale file but wait for better terms, keep another prefix for enterprise continuity, update NIR evidence, and ask finance to treat APNIC standing as part of the annual risk review.

The value in that outcome is time. The incumbent has time to see whether a customer signs, whether cloud pricing shifts, whether a merger closes, whether APNIC review clarifies a file, whether demand grows in one economy faster than another, whether a block is more valuable leased than sold, whether a lender will recognise continuity evidence, or whether a product can move to IPv6 without hurting revenue. Public IPv4 scarcity makes time expensive. Recognised holdings make time available.

That is why incumbent optionality should be analysed without resentment and without romance. Established holders did not all obtain their address positions in the same way, and not every reserve is virtuous. But scarcity does not make holding a sin. It makes holding a choice with opportunity cost. An operator that preserves recognised IPv4 must carry fees, records, review risk, reputation risk and the risk of misjudging future demand. An operator that sells too early may lose customer continuity. An operator that leases carelessly may damage reputation. An operator that waits too long may miss capital that could have built the network. The option is valuable precisely because the future is uncertain.

APNIC's duty is to keep that uncertainty from becoming registry-made uncertainty. The region needs a registry that can record who holds what, who may transfer, who must be contacted, what history can be relied on, which NIR evidence is final enough, and where disputes or policy limits sit. It does not need a registry that decides whether an incumbent has too much patience.

In the Asia-Pacific address economy, recognised IPv4 holdings are not merely legacy inventory. They are a set of doors. Some doors lead to customer continuity, some to sale, some to lease income, some to migration, some to credit support, some to reserve, and some to mistakes. The disciplined role for APNIC is to keep the corridor legible, not to choose the door.

Sources and Further Reading