|
||
|
||
The Regional Internet Registries (RIRs) were designed to manage Internet number resources fairly, regionally, and through community consensus. Yet transfer statistics tell a different story: some registries are becoming net gainers of IPv4 addresses, while others are steadily leaking resources into the global market.
This imbalance is not just a matter of accounting. It reveals how policy choices, fee structures, and community dynamics shape the incentives driving IPv4 transfers. The sustainability of the RIR system may depend on recognizing—and adapting to—these emerging realities.
Over the past three years, RIPE NCC has consistently gained millions of IPv4 addresses while ARIN, APNIC, and more recently LACNIC, have leaked supply. Several factors explain RIPE’s unique position:
The result is that RIPE has become the global center of IPv4 liquidity. Hyperscalers, notably Microsoft, have favored RIPE for large-scale acquisitions because it combines speed, low costs, and legal clarity. Leasing markets also thrive, creating secondary revenue streams for holders.
In contrast, ARIN remains burdened by needs-based justification and size-based fees that escalate sharply with larger blocks. APNIC has imposed restrictions on its “final /8” pool and high membership costs, discouraging speculative holdings.
For years, LACNIC maintained neutrality in address transfers. But in 2025, it became a net leaker, losing more than three million IPv4 addresses—the steepest single-year outflow on record.
The reason is clear: LACNIC’s strict prohibition on leasing and size-based fee model made surplus addresses liabilities rather than assets. Large members facing growing costs and limited monetization options chose to exit, moving resources to more flexible regions like RIPE.
This episode underscores a key lesson: rigid policy can backfire, triggering the very instability it aims to prevent.
AFRINIC is the only registry that does not support inter-RIR transfers, cutting itself off from the global IPv4 economy. Combined with years of governance crises, lawsuits, and board suspensions, the region has become effectively irrelevant for large-scale acquisitions.
Even with moderate fees, AFRINIC is viewed as a dead end. Resources there are considered stranded assets—unattractive to hyperscalers, telcos, or enterprises. The result is a stagnant, inward-looking market with little hope of revival without major reform.
The way RIRs charge for membership is more than accounting—it directly shapes market behavior.
This divergence explains why large holders migrate to RIPE while smaller networks in ARIN or APNIC often stay put. In LACNIC, the exponential growth of fees at higher tiers made it one of the most expensive regions for big operators, accelerating the 2025 outflows.
For operators, these differences are not theoretical—they define real-world strategy.
The net result: RIPE encourages accumulation and leasing, while other RIRs create exit incentives for large holders.
The financial stability of RIRs depends heavily on their revenue models.
In ARIN and APNIC, 10—20% attrition among top-tier members could wipe out millions in annual revenue. In LACNIC and AFRINIC, even smaller shocks would be destabilizing given their limited budgets.
Compounding the issue is policy inertia. RIR communities are aging, with few new participants joining policy forums. Without renewal, bold reforms are unlikely. Governance disputes, as seen in AFRINIC, further erode trust.
The danger is a feedback loop:
Once this cycle begins, it is difficult to reverse.
To avoid long-term instability, RIRs must evolve beyond their traditional role as custodians of scarce resources. Several steps are essential:
Transfer statistics show a clear reality: RIPE is the consistent net gainer, while ARIN, APNIC, and now LACNIC leak supply year after year. AFRINIC remains cut off entirely.
These trends are not accidents—they are the predictable outcomes of policy, fee models, and governance choices. If left unchecked, they threaten the financial stability of registries and the balance of the global IPv4 ecosystem.
The path forward is clear. RIRs must modernize their technology, harmonize their compliance processes, and renew their communities. Only then can they remain sustainable and relevant in an internet that demands flexibility, transparency, and trust.
This article is part of a bigger study. Read the full text here.
Sponsored byVerisign
Sponsored byWhoisXML API
Sponsored byDNIB.com
Sponsored byCSC
Sponsored byIPv4.Global
Sponsored byVerisign
Sponsored byRadix
Vincentas — thanks for raising these issues; they’re important ones for the RIR community to examine closely. That said, your analysis still overlooks several substantial factors that shape the actual dynamics of the IPv4 ecosystem.
First, migration risk cuts both ways. Tiered pricing models (like ARIN’s) may incentivize some large holders to move to RIPE, but they also offer far lower costs for small and mid-size holders — for example, under $300/year for a /24 — compared to RIPE’s flat €1,800 fee. In a fully portable market, those smaller holders would have just as much reason to migrate away from RIPE, which your analysis doesn’t consider.
Second, “financial weight” isn’t limited to top-tier holders. There are far more organizations in those lower/mid tiers, and their collective contribution matters very significantly to RIR revenue models.
Third, the “downward spiral” scenario is overstated. Most RIRs maintain six months or more of operating reserves precisely to cushion membership or revenue fluctuations, giving years for deliberate adjustments — not months before collapse.
Finally, current market behavior reflects decades of needs-based policies that have limited speculative participation. If those safeguards were removed, you’d likely see very different behavior from new entrants (e.g., financial actors) than the current market suggests.
It’s good to surface these issues, but they need to be discussed with the full set of dynamics on the table, not just one side of the ledger.
Good article.
Hi John,
It’s hard to argue with the facts that addresses under RIPE are growing and the other RIRs are losing addresses. But you bring up an interesing point about it being cheaper at ARIN for smaller blocks than it is at RIPE. You would need a lot of /24 holders to make up for a large legacy holder who finds another RIR less threatening or less costly, though.
As you know, needs-based transfers were jettisoned by RIPE a decade ago, meaning anybody who wants to speculate is free to do so, and current lease returns on purchased addresses are unusually attractive due to dramatically lower IPv4 purchase costs.
Finally, we are seeing the new entrants (financial actors) enter the market through investment vehicles (Cogent) even in ARIN, where need-tests persist.
There are indeed many issues to surface, one that I find interesting is how the intersection of the IPv4 market, inter-regional transfers, and no routing police has created the potential for registry shopping/registry competition.
I don’t think the numbers will be significant to any RIR’s finances, but a little competition won’t hurt the consumer.
Thanks, Mike — agreed, it’s a valuable discussion to have.
What I want to push back on is the idea advanced in the blog that policy changes are necessary because “if left unchecked, they threaten the financial stability of registries and the balance of the global IPv4 ecosystem.” That framing sounds urgent, but it doesn’t hold up under more than a superficial analysis.
If the registry system were to shift to a far more liquid and portable environment, the effects wouldn’t be one-sided. There are more than ten thousand RIPE customers currently paying the flat €1,800 annual fee who would see real savings from migrating in the other direction. The point is that we can’t assume “reducing friction” automatically solves one problem without potentially creating others — and without fully considering the impacts in both directions.
Furthermore, the RIRs aren’t fragile. Most maintain substantial reserves, diversified revenue sources, and multi-year adaptability. A revenue shortfall doesn’t lead to collapse — it leads to budget planning. And the supposed “balance of the ecosystem” is a vague notion unless we’re talking about actual operational coordination, which isn’t under threat.
If some resource holders want policy changes — to allow more leasing, remove needs-based justification, or encourage portability — they should absolutely make their case in the appropriate policy forums.
Open debate is healthy. But we should resist narratives that oversimplify or exaggerate risks to what is, in fact, a rather robust RIR ecosystem — and instead evaluate proposed policy changes on their merits, not out of manufactured claims that the IPv4 system is on the brink of collapse.