A mobile operator in Lagos, Nairobi or Cairo does not meet IPv4 scarcity as a seminar topic about an old protocol. It meets it in the capital-allocation room, after the sales forecast has already moved. The commercial team has signed enterprise customers. A bank wants more fixed egress points for anti-fraud systems. A government agency wants a cloud-hosted service reachable by legacy networks. A data-centre partner wants public addresses for hosting customers that cannot rely only on private translation. A mobile-money platform wants predictable endpoints for settlement and compliance. A school-connectivity project and a hospital network have separate procurement deadlines. The radio upgrade, metro fibre build and backhaul order are moving. The public-address plan is still being negotiated.
That meeting has a distinctive emerging-market character. The problem is not simply that IPv4 is scarce everywhere. It is that many African networks are still in a phase of fast demand expansion while the public-address layer has already entered scarcity, rationing and legal uncertainty. The network needs optionality at the same time that the registry environment asks it to document, wait, prove, litigate, lease or buy. The address is no longer a cheap administrative input that arrives before growth. It has become a scarce complement to growth after much of the growth case has already been sold.
This is different from the position of mature incumbents in North America, Europe or parts of Asia. Many of them entered the commercial internet with large legacy inventories, deeper engineering departments, older peering relationships, established procurement files and balance sheets that could absorb market purchases. Their address stock became a portfolio advantage. A fast-growing African carrier, cloud edge, fintech platform, data-centre operator, enterprise network or managed-service provider faces the opposite timing. It must scale into a world where IPv4 is already rationed, where the free pool is narrow, where transfer and leasing markets carry legal and reputational friction, and where AFRINIC's own institutional uncertainty raises the cost of every plan that depends on recognised registration.
The economic mechanism is growth pressure: the burden created when demand accelerates faster than a network's assured public-address option set. In such a market, IPv4 is not merely a technical identifier. It is a timing option, a financing input, a procurement warranty, a customer-acquisition constraint and a hedge against future uncertainty.
AFRINIC is the sharpest case because its region combines late-stage demand growth with a strained registry institution. Public records and reporting show a sequence that investors cannot ignore: AFRINIC's limited share of global IPv4 space, its movement into IPv4 exhaustion controls, the move to soft-landing Phase 2, reported allegations of past address misappropriation, a contested resource dispute with Cloud Innovation, reported bank-account freezes in Mauritius, years without ordinary board continuity, receivership, election challenges, questions over board legitimacy and continuing concern over whether registry services can remain dependable through institutional stress.
No network planner needs to accept the rhetoric of any party to see the investment implication. A scarce input is becoming more valuable. A fast-growing region needs more of it at the margin. The institution recording it has been contested in court and governance. The result is an address risk premium that falls most heavily on expansion plans, because growth is where tomorrow's need becomes today's financing decision.
The address problem arrives at the investment committee
The first mistake is to treat IPv4 scarcity as a network-engineering nuisance rather than as an investment problem. Engineering teams can conserve addresses, extend translation, segment customers, add IPv6, redesign internal pools and operate around shortage. They do this constantly. But an expanding network is not only an engineering system. It is a bundle of signed contracts, forecast revenue, debt covenants, equipment orders, spectrum costs, fibre routes, cloud commitments, data-centre racks, enterprise migrations and public-service obligations.
When a company approves a growth plan, it is buying time. It pays upfront for infrastructure in the expectation that customers will arrive quickly enough to repay the capital. Address scarcity interrupts that timing. A planned service may be technically feasible but commercially slower because public endpoints, routing acceptance, reverse DNS, RPKI, abuse contacts, geolocation and customer due diligence take longer than the sales cycle assumed. A bank or hospital may not wait for a registry issue to settle. A merchant may move to a provider with cleaner address evidence. A data-centre customer may demand a discount if address continuity is not guaranteed.
This is why the address layer behaves like an option. A network with spare, recognised IPv4 inventory can exercise growth when demand appears. It can onboard customers, split services, move traffic, reserve public endpoints, support legacy counterparties and absorb mistakes. A network without inventory must acquire, lease, justify, document or route around the need at the moment demand becomes urgent. That is more expensive than buying the same input before demand appears.
The option value is highest in fast-growing markets because growth is uncertain but potentially large. A mature incumbent with flat residential demand and stable enterprise accounts can plan address use with more confidence. A market adding mobile-data customers, cloud adoption, digital payments, public platforms and local hosting capacity faces a wider range of possible demand. The difference between a /22 and a /19 is not just 7,168 addresses. It is the difference between serving a contained project and preserving the ability to respond to several projects at once.
AFRINIC's Phase 2 rules make this visible. The published soft-landing description says Phase 2 begins when no more than a /11 of non-reserved IPv4 space remains in the final /8, and that the minimum allocation or assignment size is /24 while the maximum is /22 per allocation or assignment. A /22 is useful. It is not a growth platform for a carrier, data-centre operator or cloud-heavy enterprise segment with multiple public-facing commitments. The rule is understandable as rationing. Economically, it shifts the growth problem from allocation size to portfolio management.
The investment committee then faces a choice. It can proceed and assume the address gap will be solved later. It can scale down the project. It can lease addresses, buy from the market, hold extra inventory, redesign services around translation, delay customer commitments or add contractual caveats. Each option has a cost. The cost is not measured only in address price. It is measured in lost speed, reduced confidence and weaker bargaining power.
Mature markets grow from inventory; emerging markets grow into scarcity
The unevenness of IPv4 distribution is an inheritance from the internet's early geography. The earliest and richest network economies accumulated large blocks before the commercial internet became globally universal. They then built institutions, engineering cultures, vendor relationships and transfer practices around that stock. When scarcity arrived, they were constrained, but not in the same way as markets whose main growth curve came later.
Public analysis of AFRINIC has often noted the scale imbalance. AFRINIC has held only a small fraction of global IPv4 space. Internet Governance Project reporting in 2021 described AFRINIC as having received four /8s and having only ever held about 2% of the IPv4 address space, while much larger shares had already been distributed elsewhere. Lu Heng's later notes cite delegation data showing the United States and China together holding more than half of delegated IPv4, while the AFRINIC zone accounts for only a small single-digit share. The exact measurement depends on dataset and classification. The structural point does not: the region's address base is small relative to its population and future demand.
That matters because growth in mature markets can be financed partly out of old abundance. A large incumbent may have underused legacy space. A cloud provider may have already accumulated address assets through acquisitions and global transfers. A carrier may be able to reassign internal stock, reclaim from old services or buy in larger blocks through established channels. Its address strategy is not painless, but it is anchored in inventory.
An emerging-market network is more likely to grow from a smaller base. It may have received addresses under older need assumptions that did not anticipate today's mobile broadband, enterprise cloud, public digital services, local hosting and AI demand. It may have a business case that is large enough to require addresses but not large enough to buy them at the best price or carry long legal diligence. It may also face currency, banking and procurement friction when trying to acquire dollar-priced inputs.
The result is a timing penalty. Mature incumbents can treat address stock as a cushion. Fast-growing networks treat address access as a bottleneck. The same global scarcity therefore has different economic incidence. It slows the marginal network more than the network that already has a stockpile. It raises the hurdle rate for the new growth plan more than for the mature estate.
This distinction is often obscured by moral language about stewardship or regional fairness. The relevant economic question is narrower. Does the registry environment help a network turn rising demand into service quickly enough to justify investment? If the answer is uncertain, capital will price the uncertainty even if the public language remains noble. Address scarcity then becomes a market-structure advantage for those who already hold inventory.
AFRINIC's small pool makes timing a structural disadvantage
AFRINIC's exhaustion page provides the administrative facts. Since 2005, AFRINIC has delegated internet number resources to organisations that could justify need under policy. It adopted a soft-landing policy to manage scarcity. Phase 1 began on 31 March 2017. AFRINIC announced it was approaching Phase 2 in August 2019 and entered Phase 2 on 13 January 2020. Under Phase 2, requests are first-come, first-served, complete applications move to evaluation, and the maximum IPv4 block is /22 per allocation or assignment. Members requesting additional IPv4 space must show efficient use of prior AFRINIC-delegated space.
Those rules are not merely procedural. They define the rhythm of investment. A first-come, first-served queue rewards readiness and paperwork discipline. Completion standards reward administrative capacity. Efficient-use requirements reward operators that can document utilisation cleanly. The /22 ceiling rewards plans that can be broken into small, defensible increments. None of this is irrational. Scarcity requires rationing. But rationing is not neutral across different growth profiles.
A fast-growing network often has lumpy demand. It signs a bank, a logistics platform, a public-service platform and a hosting customer in the same quarter. It opens a data-centre hall and discovers that several tenants need public IPv4 sooner than expected. It launches fixed-wireless access in a city and finds that customer growth exceeds the conservative plan. It brings a cloud provider to a local exchange and new enterprise customers ask for stable endpoints. It must reserve addresses for migration, testing, redundancy and customer-specific controls, not only for immediate live sessions.
Rationing forces that demand into smaller pieces. The network may apply for a small amount, then return later. It may lease for temporary demand. It may put more customers behind CGNAT even when that creates support and compliance burdens. It may prioritise high-revenue customers over public or low-margin services. It may keep less reserve, reducing resilience. It may delay products that require stable public identity. In each case, scarcity converts technical planning into economic triage.
The Phase 2 environment also changes negotiation. A seller, lessor or incumbent with inventory knows that AFRINIC's free-pool alternative is limited. A customer knows that the provider may have less flexibility. A lender knows that a business plan dependent on new public-address availability has more execution risk. The registry does not need to set a market price for its rationing to affect price. The existence of rationing is enough.
The growth penalty is therefore not that AFRINIC created scarcity. IPv4 scarcity is global. The penalty is that African demand acceleration is now mediated through a small regional pool and a troubled recognition layer. The region is asked to compound digital services from a thinner address base and with less institutional calm than richer markets enjoyed during their own scaling decades.
Growth demand is lumpy, not linear
Economic models often treat network growth as a smooth curve. Operators do not experience it that way. Demand arrives in clusters around infrastructure, regulation, anchor customers and technology adoption. A new submarine cable lowers wholesale cost. A metro-fibre build opens enterprise districts. A central bank licences new payment institutions. A government digitises tax or customs. A hyperscale cloud region or data-centre campus changes buyer expectations. An IXP becomes more useful as more members peer. A mobile operator's 4G or 5G upgrade changes data use. A fintech product becomes national infrastructure almost by accident.
Each event increases address demand differently. Mobile broadband pushes large user populations behind translation but still needs public addresses for gateways, enterprise products, lawful-intercept logging, operational segmentation and customer-facing services. Data centres need address plans for tenants, appliances, managed services, firewalls, monitoring and migration. Financial platforms need stable endpoints that counterparties can whitelist and audit. Public services need reachability across old devices, old networks and old procurement systems. Schools and hospitals need reliability more than elegance. Hosting and AI infrastructure need public connectivity for customers who may not control their counterparties' network stacks.
The common feature is that these demands do not wait politely for annual registry cycles. They are tied to commercial windows. A bank migration has a cutover date. A public tender has a delivery milestone. A data-centre tenant has a move-in schedule. A mobile campaign has a launch quarter. A cloud customer has a board-approved migration plan. A delay in addresses may not cancel the project, but it can change its economics. It can force temporary workarounds, higher support cost, extra translation, customer discounts or lost contracts.
This is why a small allocation can be economically large and operationally insufficient at the same time. A /22 might cover one product, one site, one customer segment or one emergency pool. It cannot provide the full option set for a network whose demand is accelerating across several sectors. The address portfolio needs to be larger than immediate consumption because growth needs slack. Slack is not hoarding when the market is moving. It is how networks keep promises when demand arrives unevenly.
The market treats slack as value. A provider with spare, clean, recognised IPv4 can say yes to customers faster. It can run pilots. It can segregate risky traffic. It can support bring-your-own-IP arrangements more credibly. It can reserve addresses for financial institutions, hospitals and public services that have stricter controls. It can absorb an abuse incident without contaminating unrelated customers. It can negotiate from strength. Scarcity removes that slack and turns each new customer into a planning dispute.
AFRINIC's institutional uncertainty makes the lumpy-demand problem sharper. If the registry environment were calm, networks could at least model the process. They would know what documents, timings, approvals and appeals to expect. When governance and court events repeatedly enter the public record, modelling becomes harder. The network does not simply ask how many addresses are available. It asks whether the institution that must recognise, update or defend the record will remain predictable during the growth window.
Phase 2 turns expansion into option management
Phase 2 should be read as an option-management regime. The registry still performs allocation and assignment, but the amounts are small enough that serious growth plans must combine multiple instruments: conservation, translation, leasing, purchase, transfer, reclaimed inventory, customer prioritisation and sometimes project redesign. The operator is no longer merely applying for addresses. It is managing a portfolio of future rights to keep expanding.
Options have value because they protect against uncertainty. A network does not know exactly which customers will sign, which public services will digitise, which cloud workloads will land locally, which fintech products will grow fastest or which AI-hosting demand will be real. If it holds address options, it can respond. If it lacks them, it must either pay spot prices later or decline opportunities. That difference can determine whether infrastructure investment compounds or stalls.
The cost of losing optionality is not always visible in accounts. It appears as conservative planning. A data-centre operator builds fewer public-address-dependent products. A mobile operator is slower to offer enterprise static services. A managed-security provider avoids address portability commitments. A local cloud platform steers customers toward architectures that fit its address limits rather than their business needs. A public-sector vendor narrows its bid because it cannot guarantee enough stable endpoints. Each decision is rational. Together they reduce the market's growth frontier.
AFRINIC Phase 2 also increases the value of timing. An operator that secured addresses before exhaustion holds an option against later demand. An operator that needs addresses after Phase 2 must compete with every other latecomer and must prove need under scarcity. This is where emerging-market timing becomes critical. Many African operators are not late because they were inefficient. They are late because the region's digital demand is now expanding after global IPv4 exhaustion.
The address market can help, but only if recognition is reliable. Leasing and transfers can convert idle or underused stock into productive use. Public criticism often portrays commercialisation as a threat to weaker regions. That comparison is misleading if it imagines a fair pre-market allocation that never existed. Need-based distribution in an unequal world rewarded already scaled networks. A transparent market at least allows demand to be financed. The problem is not that prices exist. The problem is when price is combined with registry uncertainty, title ambiguity, transfer friction, reputation risk and delayed recognition.
For growth networks, the best environment is not a romantic return to abundance. Abundance is gone. The best environment is a recognisable, low-friction way to assemble the address options needed for expansion. That requires the registry to behave like infrastructure: accurate records, predictable services, bounded review, secure routing evidence, reliable reverse DNS, usable dispute handling and continuity even when the corporate institution is under pressure.
Governance uncertainty taxes the option before it is exercised
The option value of IPv4 depends on whether it can be exercised. An address block that can be routed, certified, delegated, updated, financed and explained is more valuable than a block surrounded by recognition doubts. AFRINIC's governance crisis matters because it taxes the option before the network uses it. The tax is paid through diligence, delay, legal review, risk discounts and customer caution.
The public chronology is now long. AFRINIC faced serious allegations of address-record manipulation before the Cloud Innovation dispute became the dominant story. In 2021, the Internet Governance Project described a major conflict over Cloud Innovation's AFRINIC-held resources, out-of-region use and leasing. It reported that Mauritius court action provisionally froze up to US$50 million in AFRINIC funds, affecting operations. It also described multiple court cases and a conflict that threatened the viability of the registry. In 2023, AFRINIC was placed under receivership by order of the Bankruptcy Division of the Supreme Court of Mauritius. In 2025, public reporting described AFRINIC as having operated without a board since 2022, with board elections under receiver supervision and legal challenges over the election process.
These facts do not mean every AFRINIC-administered address is unusable. They mean every serious growth plan must ask a governance question. Who can approve? Who can sign? Who can update records? Who can defend continuity in court? Who speaks for the institution? What happens if another injunction, election dispute or receiver decision affects ordinary service? Which actions are routine maintenance and which are value-moving changes? What evidence will customers, banks, public buyers or lenders require?
The costs are asymmetrical. A mature operator with in-house counsel, established registry contacts and address inventory can absorb governance noise. A fast-growing operator preparing a financing round, customer migration or data-centre launch cannot treat the noise as background. Its future revenue depends on timely execution. A small uncertainty in recognition can become a large uncertainty in valuation if it affects the timing of customer onboarding.
This is why official continuity statements, by themselves, are insufficient. The fact that registry services continue is necessary. It does not remove the option tax if counterparties still worry about future recognition, authority and dispute handling. Investors do not price only today's uptime. They price the probability that tomorrow's update, transfer, certificate, routing record, reverse delegation or contact correction will be accepted when the business needs it.
Governance uncertainty therefore behaves like a higher discount rate on address-dependent growth. The cash flows may still arrive, but they are riskier and may arrive later. A network with a strong address position can borrow or invest on better terms than a network whose growth depends on uncertain future recognition. That is how a registry crisis becomes an economic-development issue without needing to become a dramatic outage.
Reported address-record manipulation changed the price of trust
Trust in registry records is valuable because it reduces transaction costs. A buyer, lender, lessee, customer or regulator can accept a record more readily if the institution maintaining it is perceived as accurate, auditable and constrained. AFRINIC's past address-theft allegations damaged that perception long before receivership.
The 2021 Internet Governance Project account described investigations by Ron Guilmette and journalists in South Africa into alleged manipulation of AFRINIC records. It said a longstanding staff member, Ernest Byaruhanga, was accused of gaining control of valuable IPv4 addresses through dubious shell companies or businesses no longer in existence. The reported market value was more than US$50 million. Public accounts said Byaruhanga was fired after the allegations surfaced and that the matter was referred to Mauritius authorities. That same analysis presented AFRINIC's later posture toward Cloud Innovation partly as overcompensation after earlier control failures; that interpretation remains an analysis of contested conduct, not a court finding that resolves every claim.
The economic importance is not only the alleged misconduct. It is the change in trust pricing after misconduct is alleged. A registry record becomes more expensive to rely on when counterparties ask whether records were historically accurate, whether old assignments were legitimate, whether dormant companies were abused, whether internal controls were adequate and whether the current institution has enough credibility to correct the past without overreaching in the present.
Fast-growing networks pay for that credibility deficit. They need clean diligence when acquiring or leasing addresses. They need customers to accept that their public-address story is stable. They need lenders to treat address-dependent revenue as financeable. They need regulators and public-sector buyers to believe that registry uncertainty will not interrupt service. If AFRINIC's records are perceived as fragile, every transaction carries more questions.
The danger is that a corruption response can itself become a source of uncertainty. A registry that ignored manipulation would be unsafe. A registry that reacts with broad discretionary reclamation power may also be unsafe. The market wants a narrower answer: correct forged, corrupt or unsupported records through evidence, audit trail, notice, adjudication and proportionate remedy, while preserving live network continuity unless there is a specific reason to disrupt it.
This distinction matters for emerging-market growth because addresses are capital inputs. Capital markets tolerate risk when the remedy is predictable. They fear risk when the remedy is discretionary. A network can price the chance that a record will be challenged if it knows the review path, evidence threshold and transition rules. It cannot easily price a regime in which reported past misconduct is used to justify open-ended institutional discretion over live assets.
Receivership preserved the office but did not erase the market signal
Receivership is best understood as both a stabiliser and a warning. The 2023 Internet Governance Project article argued that AFRINIC's move into receivership showed the resilience of private, contract-based internet governance because a court-appointed receiver could preserve organisational stability, maintain assets, oversee elections, facilitate formation of a proper board and keep critical internet-resource services going. That is one reading. The market reads an additional signal: ordinary governance failed badly enough that receivership was necessary.
Both can be true. A court backstop can prevent collapse. It can keep records, services and staff activity alive. It can restrain relocation, takeover or uncontrolled restructuring. It can preserve the value of the business while leadership is replaced. For operators depending on AFRINIC, that is better than institutional free fall.
But receivership does not restore the lost years of certainty. It does not erase the bank-account freezes, board absence, election delays, legal attacks, factional accusations or public doubts over who legitimately controls the institution. It does not by itself answer how a fast-growing network should price a five-year plan that depends on timely registry recognition. It proves that continuity can survive through court intervention, but also that continuity needed court intervention.
For investors, this matters because infrastructure investment is forward-looking. A network upgrade may have a payback period of five to ten years. A data-centre campus may be financed over longer horizons. A public-service digitisation project may depend on continuity for decades. Receivership that keeps the lights on today is valuable, but the cost of capital depends on whether tomorrow's registry environment will be ordinary.
There is also a reputational channel. A registry in receivership may still operate. Yet counterparties outside the internet-governance world may not distinguish between technical continuity and institutional distress. A bank's risk officer, a public procurement board or an infrastructure fund may see "receivership" and demand more diligence. That extra diligence is a cost imposed on the operator's growth plan.
AFRINIC's recovery therefore cannot be measured only by whether services remain online. It must be measured by whether ordinary growth transactions become easier to finance, document and close. A registry that survives but leaves every expansion plan carrying a receivership premium has preserved the office without fully restoring the economic function.
Board legitimacy matters because growth needs authority certainty
The board of a registry may look remote from everyday network growth. It is not. Board legitimacy affects the authority chain through which policy, budget, senior management, litigation strategy, service priorities, external representation and high-consequence decisions become credible. A fast-growing network does not want to know the personalities of a registry board. It wants to know that someone with lawful authority can govern the institution without every major action becoming contestable.
AFRINIC's board problem has been unusually visible. Public reporting in 2025 described the organisation as having operated without a board since 2022. It described an election under receiver supervision, legal challenges by TISPA and ICANN-related concerns over transparency, fairness and voting rights, and a Supreme Court of Mauritius ruling on 19 June 2025 that dismissed the reported challenges and allowed the election process to proceed. The reporting also noted disputes over Cloud Innovation's status in corporate records and described the court as rejecting the argument that the registration issue should be attributed to AFRINIC and the receiver in the way challengers suggested.
For growth economics, the point is not which candidate won or which camp was morally superior. The point is that control of the board became a live legal and strategic issue because control of AFRINIC carries power over critical internet resources in the region. If board authority is uncertain, members and counterparties must ask whether policies, reviews, contracts, settlements, appointments and service decisions will survive challenge.
That uncertainty is especially costly for expansion plans. A mature network can wait out a governance cycle. A fast-growing network may have customer deadlines that do not align with litigation. It may need assurances for a financing close. It may need a transfer processed before a data-centre opening. It may need routing-security support for a public-service cutover. It may need registry evidence for a large enterprise customer. If the board's legitimacy is still a contested market signal, every such action becomes harder to treat as routine.
The registry's legitimate function is narrow but important: preserving unique, accurate and updateable records around number resources. Board legitimacy is valuable only insofar as it supports that function. It is not valuable as institutional grandeur. A board that sees itself as a restrained steward of records lowers the cost of growth. A board that becomes a prize in a struggle over discretionary control raises it.
The next stage of AFRINIC's credibility will depend less on rhetoric than on ordinary authority certainty. Members need clear sign-off paths. Staff need stable mandates. Courts need identifiable representatives. Customers need confidence that routine changes will not become governance disputes. Investors need to believe that the registry will remain an infrastructure utility rather than a theatre of control.
Demand acceleration makes delay more expensive than price
The price of IPv4 is visible. Delay is harder to see and often more expensive. A network can finance an address purchase or lease if the price is known. It can pass some cost to customers or capitalise the asset. It cannot easily finance an indefinite wait, an uncertain registry decision, a contested record or a migration blocked by unclear authority. Delay consumes the one asset fast-growing markets cannot recover: timing.
Consider a data-centre operator preparing to host local AI workloads, content platforms and enterprise services. The address requirement may not be enormous in global terms, but the timing is critical. Customers want to know when servers, firewalls, load balancers, management interfaces and public endpoints will be ready. If address availability is uncertain, the operator may overbuild private-network workarounds, offer weaker service terms, lease at a premium or lose anchor customers. The market does not wait because a registry file is complicated.
Fintech is similar. Payment services, fraud monitoring, API access, bank integrations and regulatory controls often rely on stable network identity. If a fintech platform expands into new services or countries, address stability becomes part of counterparty trust. A delay in public-address readiness can delay banking approval or require temporary infrastructure that later must be unwound. The cost is not the number of addresses. It is the lost transaction volume and customer confidence during the delay.
Public services also turn delay into social cost. A tax portal, hospital network, education platform or customs interface cannot treat address uncertainty as a purely commercial inconvenience. If a migration is delayed, old vendors stay longer, security exceptions remain open, citizens encounter unreliable service, or public agencies pay for overlapping infrastructure. The address layer is invisible until it becomes the reason a public project cannot close cleanly.
This is why the emerging-market growth-pressure story is not a plea for sympathy. It is a timing analysis. Where demand is accelerating, the cost of waiting rises quickly. A mature market may spend years optimising an address estate. A fast-growing market may need to decide this quarter whether to sign customers, light capacity, open a data-centre hall, launch a product or bid for a public contract. Delay turns scarcity into lost compounding.
AFRINIC's challenge is to lower the delay premium. It cannot create a new global IPv4 supply. It can, however, make recognition, updates, routing evidence, reverse DNS, dispute markers and continuity rules predictable enough that networks can combine market acquisition with stable registry treatment. That would not eliminate price. It would make price financeable.
The market advantage of incumbent inventory
Scarcity transfers bargaining power to those who already hold usable inventory. This is not morally surprising. It is basic economics. The owner of a scarce input gains optionality over those who need it later. In IPv4, that advantage is magnified because address stock is not a generic commodity. Blocks carry history, reputation, registry region, transfer conditions, routing acceptance, geolocation memory, abuse history and customer-specific usefulness.
Mature incumbents benefit in at least four ways. First, they can serve new customers from internal stock without entering the market at every demand spike. Second, they can wait for better purchase opportunities. Third, they can monetise unused or underused space through sale, lease, internal redeployment or acquisition strategy. Fourth, they can present address continuity as part of their service quality. For customers, that can look like reliability. For competitors, it is an inherited barrier.
Fast-growing African networks face the mirror image. They must buy or lease into a seller's market. They must prove clean authority around addresses administered by a registry that has faced public institutional stress. They may need smaller blocks now and larger blocks later, which worsens pricing and fragmentation. They may lack the balance-sheet capacity to buy ahead of need. They may also be judged by customers against global providers whose address portfolios were assembled under earlier, easier conditions.
This creates an opportunity-cost problem. The capital used to acquire IPv4 could have bought fibre, power resilience, radio equipment, data-centre cooling, security tools, customer support or software development. The higher the address risk premium, the more growth capital is diverted from productive infrastructure into scarcity management. Some of that is unavoidable because IPv4 is scarce. Some is avoidable if registry friction is reduced.
A transparent market does not remove incumbent advantage, but it can reduce the penalty of late growth by making access clearer. A market with uncertain recognition, inconsistent transfer treatment and registry risk does the opposite. It lets incumbents enjoy both inventory and lower uncertainty while newer growth networks face both price and institutional friction.
AFRINIC's institutional crisis therefore has a competitive dimension. It does not merely affect members equally. It changes how customers compare providers. A large global cloud, carrier or hosting firm can reassure customers through scale and portfolio depth. A local African challenger may have better local knowledge, better service and better proximity, yet still lose because its address story is harder to underwrite. Growth pressure then becomes a route through which infrastructure markets concentrate.
The policy lesson is modest. If AFRINIC wants African network growth, it should not romanticise scarcity management. It should minimise the avoidable frictions around the scarce input. The more predictable the registry record, the more address markets can reallocate stock toward productive African growth rather than merely rewarding those who entered the game earlier.
Financing treats registry uncertainty as a discount rate
Infrastructure finance converts uncertainty into discount rates, covenants, reserves and conditions precedent. A lender or investor does not need a philosophical view of internet governance. It needs to know whether the projected cash flows are likely to arrive. If address access is essential to those cash flows, registry uncertainty becomes a financing input.
A mobile expansion may require public addresses for enterprise services, gateways, management, customer-specific products and compatibility with external systems. A data-centre project may require address capacity to attract tenants. A fintech platform may require stable endpoint identity to satisfy banks and regulators. A public-service vendor may require address continuity to win procurement. If any of these depends on future AFRINIC recognition, the financing file must address that dependency.
The questions are practical. Does the company hold recognised resources? Are the records accurate? Are the addresses free of known disputes? Can transfers or leases be recognised? Are reverse DNS and routing-security materials maintainable? Is there a record of prior assignments or downstream commitments? Could a registry review affect service? What happens if AFRINIC governance or court proceedings delay an update? Can the company substitute market purchases? At what price? In what currency? On what timeline?
Each unanswered question raises the discount rate. The project may still be financed, but with more equity, higher interest, tighter covenants, lower valuation or staged release of funds. In a fast-growing market, that matters. A higher cost of capital can make the difference between building ahead of demand and waiting until competitors have captured customers.
This is one reason IPv4 should be analysed as economic infrastructure. It is not just a line in a network plan. It enables revenue and affects financeability. Public market commentary around IPv4 as an investment asset sometimes pushes the claim further than a conservative operator would. Even so, it identifies a structural truth: IPv4 addresses enable services whose revenue value can far exceed the direct monthly charge historically attached to an address. If the enabling asset is scarce and institutionally uncertain, the businesses built on it carry that uncertainty into valuation.
AFRINIC cannot eliminate all valuation risk. No registry can. It can lower the institutional component by making the record auditable, services dependable, disputes isolated and authority chains clear. That would reduce the discount applied to address-dependent African growth. Conversely, if AFRINIC's next decade is dominated by renewed board fights, legal ambiguity, opaque reviews or contested enforcement, the discount will persist even if packets keep flowing.
The point is not that finance should control registry policy. The point is that registry policy already affects finance. A system that ignores this will still be priced by investors; it will simply be priced with less confidence.
The opportunity cost is foregone compounding
Opportunity cost is the right measure because the growth network loses more than an address. It loses the compounding that would have occurred if the address option had been available when demand appeared. A delayed enterprise product does not simply move revenue from March to June. It may lose a customer relationship, a reference account, a bank integration, a public tender score, a peering opportunity or a platform partnership that would have made the next sale easier.
This compounding effect is central to emerging markets. A new IXP becomes more valuable as more networks connect and exchange traffic locally. A data centre becomes more attractive as more tenants, carriers, cloud platforms and security vendors cluster around it. A fintech platform becomes more useful as more banks, merchants and public agencies trust its endpoints. A school or hospital programme becomes easier to expand when the first sites prove stable. Each additional participant lowers the cost of the next participant. Address uncertainty interrupts that sequence.
The interruption is not always dramatic. A local hosting provider may postpone a premium service that requires stable public addresses. An IXP participant may delay a peering upgrade because routing evidence and customer addressing are not ready. A bank may ask for a longer security review. A public agency may keep an old contractor because the new provider cannot prove address continuity. An AI-hosting start-up may choose foreign infrastructure because public-address capacity is easier to obtain there. None of these decisions alone defines a national market. Together they decide where the next layer of digital value is built.
The address shortage also changes product design. A provider with constrained IPv4 can still serve customers, but it may push them toward architectures that fit the provider's scarcity rather than the customer's best operating model. It may use more translation, shared egress, ports, proxies, overlays and exceptions. Those tools are useful. They are not free. They add logging burdens, support complexity, application failures, abuse-management problems and customer confusion. The customer may never know the constraint began at the address layer, but it pays through a less flexible service.
Foregone compounding is hardest to measure precisely because it appears as what did not happen. A data-centre hall that opened at lower occupancy. A local cloud service that never reached scale. A public-service contract that went to a foreign platform. A managed-security product that was not offered. A peering relationship that remained shallow. A network expansion that served retail traffic but not enterprise customers. Economic statistics rarely label these outcomes as IPv4 or registry costs. Yet the mechanism is real whenever address uncertainty changes the feasible set of investments.
This is the point at which growth pressure becomes distinct from ordinary scarcity. The network is already moving. Customers are already visible. Infrastructure has already been ordered or financed. The question is whether the operator can compound when demand arrives faster than its address portfolio, or whether each promising new contract forces a fresh round of scarcity management before revenue can appear.
AFRINIC's role should be judged against that compounding test. If the registry can make records dependable and recognition predictable, market purchases and leases can be turned into African capacity. If it cannot, address stock will still move, but more value will be captured by those with stronger legacy portfolios, better legal teams and easier access to external infrastructure. The difference is not only who owns addresses. It is where the next increment of network value is built.
Peering, hosting and AI make address certainty a local multiplier
Local interconnection changes the economics of growth pressure because it makes every stable endpoint more useful. When networks peer locally, content caches, enterprise services, payment platforms, security providers and public-sector applications can serve users with lower latency and lower transit cost. The address layer is not the whole story; fibre, power, routing, data-centre quality and commercial trust all matter. But public-address certainty helps turn local connectivity into usable local services.
An IXP, for example, does not need large public-address pools to exist. But its participants often do. Carriers need addresses for customer services and edge devices. Content and hosting providers need addresses for servers, appliances and management. Enterprise networks need stable identities for remote access, security policy and partner connectivity. Route servers and routing records must be maintained with enough clarity that participants can filter and accept traffic. If the address environment is uncertain, the exchange may still pass packets, but its ability to anchor higher-value local services is weaker.
The same is true for hosting and AI infrastructure. AI workloads are often discussed in terms of GPUs, power and cooling. Those inputs are critical, but the service still has to be reachable by customers, partners, APIs, developers and monitoring systems. Some traffic can live behind platforms and private arrangements. Some still needs stable public IPv4 compatibility because enterprise environments, old equipment, regulatory systems and partner networks remain mixed. A local AI or hosting provider that cannot offer predictable public connectivity is at a disadvantage against a foreign platform with deeper address inventory.
Public digital services create another multiplier. A hospital network, school system, tax portal or customs platform often depends on counterparties outside the immediate project: vendors, banks, logistics firms, identity providers, payment processors, regional offices and citizens using old networks. The more interconnected the service, the more valuable stable public network identity becomes. Address uncertainty does not stop digitisation in a single stroke. It forces more exceptions and workarounds, which slow scaling and make public confidence more fragile.
Submarine cables and backhaul upgrades intensify the point. More capacity entering a country should lower cost and support local services. But bandwidth alone does not create digital industry if the address, hosting, peering and trust layers cannot scale with it. A country can have better international capacity and still send value offshore if local providers cannot assemble the address options and registry evidence needed to host serious services. Growth pressure is therefore not solved by fibre alone.
This is also where mature-market inventory advantage becomes most visible. A global platform entering an African market may bring address assets, operational playbooks, legal teams and reputation with it. A local provider may bring proximity, language, customer knowledge and lower latency, but still struggle to match the public-address confidence embedded in the global platform's portfolio. If registry uncertainty increases local friction, it pushes demand toward the better-capitalised entrant even when local infrastructure is technically capable.
The policy objective should not be to insulate local providers from competition. It should be to ensure that the registry layer does not add avoidable friction to local compounding. Better records, predictable updates, clear dispute isolation and reliable routing evidence would make African hosting, peering and AI infrastructure easier to underwrite. That would not guarantee success. It would remove one unnecessary tax from the local multiplier.
Continuity must protect the record, not the institution
The most important distinction in AFRINIC's future is between continuity of the registry function and continuity of the institution's full authority claims. The registry function is real. Number uniqueness matters. Accurate registration matters. RDAP, Whois, reverse DNS, RPKI and routing records matter. Legitimate updates matter. Dispute records matter. Running networks and customers must be protected during conflict.
None of that proves that every institutional claim made by AFRINIC, its board, a receiver, a policy faction, a global coordination body or a litigation party deserves protection. The function is narrower than the rhetoric around it. Growth networks need the function to continue. They do not need institutional drama to be converted into a test of regional loyalty.
The continuity architecture should begin from the live network. The last verified operational state should be preserved during ordinary disputes unless there is fraud, compromise, court order or another specific high-risk reason to act. Disputed resources can be marked as disputed without disabling unrelated maintenance. Transfers can be paused without breaking reverse DNS. Holder changes can be restrained without preventing emergency contact correction. Routing-security materials can be maintained for unchanged origins while ownership questions are reviewed. The registry can protect evidence without turning customers into leverage.
This approach also helps growth. A network deciding whether to invest needs assurance that a future dispute will not automatically contaminate its entire address-dependent business. It can accept that fraud will be corrected. It can accept that fees must be paid and records must be accurate. It can accept that transfers require evidence. What it cannot easily finance is a system where any institutional conflict may interrupt the ability to maintain ordinary services.
Protecting the record rather than the institution would also make AFRINIC more credible after its crisis. Receivership showed that the service can be separated from ordinary board control in an emergency. Continuity planning should make that separation deliberate rather than improvised. Records should be versioned, auditable and backed by clear authority trails. Essential services should have failover and succession plans. Disputes should have independent review paths. Staff should know which services are maintenance and which are value-moving changes. Members should know what is protected even when the institution is under stress.
Such a design would not weaken AFRINIC's legitimate role. It would focus it. A registry that can say "the record will be protected, the network will not be needlessly disrupted, and disputes will be isolated" is more valuable to growth than a registry that insists its institutional prestige is indistinguishable from internet stability. Growth capital trusts boring infrastructure, not symbolic authority.
The better test is whether growth becomes easier to underwrite
AFRINIC's next phase should be judged by whether African network growth becomes easier to underwrite. That is a stricter and more useful test than asking whether the registry has survived. Survival is necessary, but it is not enough. A registry can survive while leaving a high risk premium on every address-dependent investment plan.
The practical indicators are observable. Are routine updates processed on predictable timelines? Are reverse-DNS and routing-security services insulated from governance disputes? Are transfer and lease-related recognition questions handled with clear evidence standards? Are disputed records preserved without collateral disruption? Are election and board processes credible enough that members know who can act? Are court-related constraints translated into narrow service rules rather than broad uncertainty? Are public records accurate enough for customers, lenders and procurement boards to rely on them?
If these answers improve, AFRINIC can reduce the growth-pressure penalty even under global IPv4 scarcity. Networks will still need to buy, lease, conserve and deploy IPv6 where appropriate. They will still face high address prices and legacy compatibility. But they will be able to treat the registry layer as an infrastructure utility rather than as an additional source of volatility. That difference matters when demand is accelerating.
If the answers do not improve, the region's growth networks will adapt in less efficient ways. They will overuse translation, avoid products that require stable public identity, accept worse leases, buy address stock at higher prices, rely on foreign platforms, delay local hosting, demand more customer lock-in or move value to providers with stronger inherited inventories. None of these adaptations looks like a dramatic policy failure on its own. Together they reduce the compounding of African digital infrastructure.
The issue is urgent because the region's demand curve is not waiting for institutional repair. Mobile data, cloud access, fintech, public services, data centres, submarine-cable and backhaul upgrades, enterprise networking, peering, hosting, AI workloads, schools and hospitals are all moving in the same direction. They require more reliable public network identity, not less. IPv6 will help over time, but it does not remove the medium-term demand for IPv4 compatibility in commerce, government and legacy interconnection.
AFRINIC's institutional crisis has therefore turned a global scarcity problem into a regional growth-finance problem. The address shortage is real. The free pool is narrow. The market is expensive. But the avoidable cost is uncertainty over the recognition layer. A fast-growing network can pay a known price for a scarce input. It can design around a known technical constraint. It can finance a known migration path. It cannot efficiently compound growth when the institution that records the input is itself a recurring uncertainty.
The economic conclusion is blunt. Emerging-market growth pressure makes registry restraint more important, not less. The faster the network economy expands, the less room there is for discretionary ambiguity around the addresses that let customers, banks, public agencies, cloud platforms and counterparties recognise each other. AFRINIC's legitimacy will not be restored by asking the market to admire the institution. It will be restored if the next operator's investment committee can treat the registry record as boring enough to build on.