What 2026 IPv4 Transfer Data Reveals About Address Reuse and Market Demand

IPv4 transfer data shows that address scarcity has not made IPv4 static. Instead, millions of previously allocated addresses continue to move between organisations, regions, and networks. The data suggests that reuse—not new global IPv4 supply—has become an important part of how existing IPv4 capacity is redistributed.

That distinction matters.

The global IPv4 address space is finite, but the operational use of that address space is constantly changing.

Companies grow.

Networks consolidate.

Infrastructure moves.

Older address holdings become underused.

Some organisations require additional public IPv4 capacity while others no longer need everything they hold.

The result is a secondary transfer system in which existing IPv4 resources can move to new organisations and, eventually, new operational environments.

Data from Regional Internet Registries provides a useful window into this process.

It does not reveal every commercial arrangement in the IPv4 market, and transfer volume should not be treated as a perfect measurement of demand. But it can tell us something important about how existing address space is being redistributed and reused.

For the LARUS Foundation, that makes IPv4 transfer data more than a market indicator.

It is also a way to understand the changing lifecycle of one of the Internet's most constrained technical resources.

Why IPv4 Transfer Data Matters

IPv4 contains approximately 4.3 billion possible addresses because the protocol uses a 32-bit address space.

Not all of those addresses are available for ordinary public allocation, and the major pools of previously unallocated IPv4 space have long been constrained or exhausted.

LARUS Foundation explains the underlying distribution process in What Is IP Allocation? and the broader scarcity issue in IPv4 Addresses: A Scarce Resource.

Scarcity, however, does not mean an address remains permanently tied to its first operational use.

IPv4 transfer mechanisms allow previously allocated resources to be registered to different organisations under applicable registry processes.

APNIC describes transfers as a way for organisations requiring Internet number resources to receive IPv4 addresses from organisations with unused resources, while keeping the resources correctly registered to the organisations using them.

See: APNIC – Transfer of Unused IP and AS Numbers

That creates an important distinction:

IPv4 supply is fixed at the protocol level, but usable IPv4 capacity can still be redistributed.

Transfer data helps us observe that redistribution.

The 2025 Baseline: More Than 33 Million Addresses Transferred

Before looking at 2026, it is useful to establish a baseline.

An APNIC analysis of RIR transfer records reported approximately 33.4 million IPv4 addresses transferred during 2025.

The receiving-RIR totals reported in that analysis were approximately:

The same analysis estimated that transfer logs had recorded about 342 million IPv4 addresses since 2012, equivalent to roughly 9.3% of the delegated IPv4 address space.

That cumulative figure needs an important qualification: some address blocks have transferred more than once, so it should not be interpreted as 342 million unique addresses permanently moving exactly once.

Source: APNIC Blog – IP Addresses Through 2025

Even with that limitation, the scale is significant.

The transfer system is no longer a marginal feature of IPv4 administration.

It has become one of the mechanisms through which existing address resources move between organisations.

What 2026 Data Shows So Far

The 2026 data continues to show substantial transfer activity.

The RIPE NCC publishes monthly information about the number of IPv4 addresses transferred in its service region.

For January through July 2026, its member updates reported:

These figures are calculated from the RIPE NCC's monthly member updates for the first seven months of 2026.

See: RIPE NCC Member Updates

The July figure is particularly notable: 5.94 million IPv4 addresses were transferred in a single month in the RIPE NCC service region.

Source: RIPE NCC Member Update – August 2026

For context, RIPE NCC received about 22.3 million transferred IPv4 addresses over all of 2025 according to APNIC's cross-RIR analysis.

The first seven months of 2026 therefore already represent roughly three-quarters of that 2025 volume.

This does not mean 2026 will necessarily finish above 2025. Transfer activity can be uneven from month to month, as July itself demonstrates.

But it does show that large-scale redistribution remains active.

What Does This Tell Us About IPv4 Demand?

Transfer volume is often described as a measure of IPv4 demand.

That is useful, but incomplete.

A completed transfer reflects several things at once:

• a recipient willing and able to obtain additional IPv4 space;
• a source willing and able to release or transfer address space;
• applicable registry requirements being satisfied;
• a transaction or organisational change reaching completion; and
• the registry recording the resulting change.

Therefore:

Transfer volume measures completed registered redistribution, not total underlying IPv4 demand.

Demand may exist without producing a completed transfer.

A network may use leased addresses.

An organisation may delay expansion.

A company may use carrier-grade NAT.

A cloud provider may reorganise existing address inventory.

A network may deploy IPv6 alongside IPv4.

Another organisation may remain on a waiting list for available IPv4 space.

This is why transfer statistics should be interpreted together with other data.

Waiting Lists Provide Another Demand Signal

Registry waiting lists help show that some organisations still seek IPv4 resources outside ordinary secondary-market transfers.

ARIN states that its IPv4 free pool depleted on 24 September 2015. Its current options include a waiting list, specified-recipient transfers, and limited reserved pools for particular purposes.

See: ARIN – IPv4 Address Options

In January 2026, ARIN reported fulfilling 149 waiting-list requests using 59 available IPv4 blocks.

Source: ARIN – January 2026 Waiting List Announcement

RIPE NCC data tells a similar story from another angle.

Its August 2026 member update reported 757 LIRs on the IPv4 Waiting List, with the first LIR in the queue having waited 467 days.

Source: RIPE NCC Member Update – August 2026

These figures should not be added to transfer statistics as though they measure the same thing.

They do not.

But together they show that IPv4 demand has multiple expressions:

completed transfers, waiting-list requests, leasing, internal reuse, address sharing, and migration toward IPv6 can all coexist.

The Most Interesting Story Is Address Reuse

The headline market story is often price.

For Internet infrastructure, however, the more interesting story may be reuse.

The IPv4 protocol does not gain new addresses when a transfer happens.

Instead, an existing block changes administrative or organisational context and may eventually be redeployed.

Consider a simplified lifecycle:

Initial allocation
↓
Network deployment
↓
Years of operational use
↓
Reduced or changed requirement
↓
Transfer
↓
Registry update
↓
Routing and security preparation
↓
Redeployment in another network

This is fundamentally different from manufacturing new supply.

It is recycling existing Internet number resources into new operational uses.

That is why transfer data matters even to readers who are not buying or selling IPv4 addresses.

It provides evidence about how a fixed technical resource adapts to changing network demand.

Older IPv4 Space Is Re-entering Active Use

One of the strongest indicators of reuse is the age of transferred address space.

APNIC's analysis of 2025 transfer records found that approximately 40% of the volume of transferred addresses had originally been allocated 20 or more years earlier.

At the other end of the spectrum, about 20% of transfer transactions involved prefixes allocated within the previous seven years, but those transactions represented less than 2% of the total address volume transferred.

Source: APNIC Blog – IP Addresses Through 2025

That difference is revealing.

Many smaller, more recent resources are changing hands, but a significant share of transferred address volume comes from much older allocations.

This is consistent with a reuse model in which address space originally allocated during earlier phases of Internet growth is gradually being reassessed and, in some cases, moved into new operational environments.

It also shows why the age of an IPv4 allocation does not necessarily tell us how the address space is being used today.

Network requirements change over decades.

Registry data needs to evolve with those changes.

Transfers Can Split Larger Blocks Into Smaller Ones

Another useful question is whether IPv4 transfers change the shape of the address space.

They do, at least to some extent.

APNIC's analysis found 56,629 transfer transactions recorded from the start of 2012 through the start of 2026.

Of those, 14,831 entries—about 26%—involved a transferred block smaller than the original allocation.

In other words, some transfers effectively split larger historical allocations into smaller prefixes.

Those fragmented transfers came from 9,231 original allocations, with each original allocation split into an average of around 1.9 smaller blocks.

Source: APNIC Blog – IP Addresses Through 2025

This is an important market signal.

A network that no longer needs a large historical block does not necessarily have to move the entire block as one unit.

Depending on the applicable technical and registry arrangements, smaller portions may move separately.

That can allow address supply to better match the requirements of smaller recipients.

But fragmentation also has operational implications.

More-specific prefixes can affect:

• address planning;
• route management;
• registry records;
• RPKI configuration;
• reverse DNS;
• IPAM; and
• future aggregation options.

LARUS Foundation explores the relationship between address structure and long-term operational planning in How IPv4 Subnet Design Supports Network Continuity.

Does Fragmentation Create a Routing Problem?

The data requires careful interpretation.

A transfer record showing that an allocation was divided does not automatically mean every resulting block becomes a new global BGP announcement.

Registry fragmentation and routing-table fragmentation are related concepts, but they are not identical.

A transferred prefix may:

• remain aggregated in routing;
• be originated separately;
• be incorporated into an existing routing architecture;
• remain temporarily unrouted; or
• change operational use later.

This is why transfer data should not be used by itself to make claims about BGP growth.

Registration tells us that a resource relationship changed.

Routing data tells us how networks are announcing the addresses.

RPKI provides another layer of information about route-origin authorisation.

IPAM describes how the recipient manages the addresses internally.

Understanding the IPv4 lifecycle requires looking at these layers separately.

Transfer Does Not Automatically Mean Immediate Reuse

Another important limitation is that a completed transfer is not identical to a completed operational deployment.

When a registry records a transfer, several steps may still need to occur before the address space becomes fully integrated into its new environment.

Depending on the network, that can include:

• updating routing arrangements;
• preparing BGP announcements;
• creating or updating ROAs;
• updating Internet Routing Registry information;
• configuring reverse DNS;
• integrating the block into IPAM;
• reviewing address reputation;
• creating firewall policies;
• documenting new assignments; and
• migrating production workloads.

This matters because the administrative lifecycle and operational lifecycle are connected but not identical.

LARUS Foundation's guide to IP Address Registration explains why registration records are an important shared layer without being a substitute for routing or internal network management.

Address Reputation Can Follow a Previous Use

Reuse also creates a less obvious challenge: historical reputation.

An address may change registrant or operator, but external systems may still associate it with activity from its previous use.

ARIN has warned blocklist operators that reputation assumptions about returned addresses may no longer remain valid after those addresses are redistributed to new recipients.

See: ARIN Announce Archive

That is an important reminder about the lifecycle of an IP address.

The number can be reused.

The registry record can change.

The route can change.

But third-party databases do not always update at the same speed.

Operators receiving previously used IPv4 space may therefore need to review:

• blocklist status;
• geolocation records;
• reputation databases;
• reverse DNS;
• routing history; and
• other external dependencies.

Address reuse works best when the surrounding information ecosystem also catches up with the new operational reality.

Why Accurate Registry Records Matter During Reuse

If IPv4 addresses move between organisations, registry accuracy becomes more—not less—important.

A transfer creates a change in real-world resource relationships.

The shared record should reflect that change accurately.

RIPE NCC's public transfer database, for example, records information including the original block, transferred blocks, source and destination organisations, country codes, transfer type, and processing date.

It also distinguishes policy transfers from transfers associated with changes in business structure such as mergers or acquisitions.

See: RIPE NCC – IPv4 Transfer Statistics

That information supports transparency around resource movement.

Accurate records can help with:

• identifying responsible organisations;
• transfer auditing;
• operational troubleshooting;
• security coordination;
• resource due diligence; and
• understanding changes in address distribution.

LARUS Foundation discusses this broader principle in Why Accurate Registry Data Supports a Stable Internet.

The underlying idea is straightforward:

When operational reality changes, shared records should be able to reflect that change.

Market Demand Is Not Uniform Across Regions

Another lesson from transfer data is that there is no single global IPv4 market pattern.

The 2025 data showed substantial differences between receiving RIR regions.

RIPE NCC accounted for the largest transferred-address volume in the APNIC analysis, followed by ARIN and APNIC.

Source: APNIC Blog – IP Addresses Through 2025

APNIC's 2025 annual review also noted that transfer activity remained high relative to the longer-term trend, even though total transferred IPv4 volume was below 2023–2024 levels.

See: APNIC – Key Moments from the APNIC AGM at APRICOT 2026

Regional differences can arise from many factors:

• historical address distribution;
• network growth;
• available inventory;
• transfer policies;
• market maturity;
• organisational restructuring;
• cloud and hosting demand;
• IPv6 adoption; and
• differences in how operators obtain additional IPv4 capacity.

For this reason, a global aggregate can hide important regional changes.

A falling transfer volume in one region does not necessarily mean global IPv4 demand is disappearing.

Likewise, a large month in another region does not prove demand is accelerating everywhere.

IPv4 Scarcity and IPv6 Growth Can Exist at the Same Time

IPv4 transfers are sometimes framed as evidence that IPv6 adoption has failed.

That conclusion is too simple.

Networks can deploy more IPv6 while still requiring IPv4.

Many Internet services operate in dual-stack environments.

Customers, applications, enterprise networks, APIs, security systems, and legacy infrastructure may still require IPv4 reachability even when the same network supports IPv6.

IPv4 transfer demand and IPv6 deployment therefore do not have to move in opposite directions.

They can represent two simultaneous infrastructure realities:

IPv6 expands the long-term address architecture.

IPv4 transfers redistribute an existing finite resource that many networks still use.

This distinction is important for interpreting market data without turning it into an argument for or against either protocol.

What Does 2026 Data Tell Network Operators?

The data suggests several practical lessons.

1. IPv4 Has an Active Secondary Lifecycle

Previously allocated addresses continue to move.

The lifecycle of IPv4 does not necessarily end when the original holder no longer needs the resource.

2. Older Address Space Remains Economically and Operationally Relevant

A substantial portion of transferred volume comes from address space allocated many years ago.

Historical allocation does not mean permanently fixed use.

3. Smaller Blocks Matter

The fragmentation data shows that larger allocations are sometimes divided during transfer.

That allows the resource structure to adapt to different recipient requirements.

4. Registry Data Is Only One Layer

Transfer records document administrative movement.

Operators still need to consider routing, RPKI, reverse DNS, IPAM, reputation, and production deployment.

5. Demand Should Be Measured Using More Than One Indicator

Transfer volume, waiting lists, address utilisation, IPv6 adoption, leasing activity, and routing data describe different parts of the market.

No single statistic captures everything.

What the Data Does Not Tell Us

Good market analysis also requires recognising what the available data cannot prove.

Public RIR transfer records generally do not provide a complete picture of:

• private commercial pricing;
• all IPv4 leases;
• confidential temporary arrangements;
• unused space retained by existing holders;
• future demand;
• the profitability of IPv4 transactions; or
• every operational change affecting an address block.

Transfer records also contain mergers, acquisitions, and other organisational changes that may not represent a conventional open-market sale.

That means it would be misleading to interpret every transfer as an identical commercial transaction.

The data is strongest when used to answer a narrower question:

How much registered IPv4 address space is moving, and what does the structure of that movement tell us about resource reuse?

For that question, the evidence is valuable.

From Allocation to Reuse: The Modern IPv4 Lifecycle

The Internet's address system was originally organised primarily around allocation and assignment.

Scarcity added another stage.

Today, the lifecycle of an IPv4 block may look more like:

Allocation
The resource enters the number-resource system.

Deployment
An operator uses the block in a network.

Operational dependency
Servers, customers, applications, routing systems, and security tools become associated with the addresses.

Reorganisation
The original requirement changes.

Transfer or other reassignment
The resource relationship changes.

Registry update
Shared records are updated.

Technical transition
Routing, RPKI, DNS, IPAM, and other systems are adjusted.

Reuse
The address space supports a new operational environment.

This lifecycle connects several topics that are often studied separately.

Allocation matters.

Registration matters.

Routing matters.

Subnet planning matters.

Continuity matters.

Transfers matter.

What the market data helps reveal is that these are not isolated stages.

They form a continuing resource lifecycle.

Why This Matters for Internet Resource Management

IPv4 scarcity is often discussed as a simple shortage problem.

Transfer data shows a more complex reality.

The Internet is also managing a redistribution problem.

Existing resources need to move from places where their value or utilisation has changed toward networks that can put them to new use.

For that process to work well, several things matter:

• accurate registry records;
• clear resource history;
• reliable technical coordination;
• operationally manageable prefixes;
• routing-security updates;
• clean documentation; and
• continuity during infrastructure change.

This fits a broader principle in Internet number-resource management: the shared coordination layer should help maintain reliable information, while network operators remain responsible for the detailed architecture and operational use of their resources.

Frequently Asked Questions

What is the IPv4 transfer market?

The IPv4 transfer market refers broadly to the secondary redistribution of previously allocated IPv4 address space between organisations under applicable registry processes and commercial or organisational arrangements.

How many IPv4 addresses were transferred in 2025?

APNIC's analysis of RIR transfer records reported approximately 33.4 million transferred IPv4 addresses in 2025.

The figure includes different types of registered transfers and should not be interpreted as 33.4 million unique open-market sales.

Source: APNIC Blog – IP Addresses Through 2025

How many IPv4 addresses were transferred in 2026?

Complete global 2026 data is not yet available because the year is still in progress.

RIPE NCC monthly reports alone recorded approximately 16.72 million IPv4 addresses transferred between January and July 2026.

See: RIPE NCC News and Member Updates

Why are old IPv4 addresses being transferred?

Network requirements change.

An organisation may hold address space that is no longer fully required, while another network may need additional IPv4 capacity.

Transfer mechanisms allow existing resources to be redistributed rather than requiring new global IPv4 supply.

Does an IPv4 transfer create new addresses?

No.

A transfer changes how existing IPv4 address space is registered or controlled.

It does not increase the size of the IPv4 protocol's 32-bit address space.

Does a transfer mean the addresses are immediately routed?

Not necessarily.

Registration and routing are separate layers.

A recipient may still need to configure BGP, RPKI, reverse DNS, IPAM, security systems, and other infrastructure before production deployment.

Are IPv4 transfers increasing?

The answer depends on the region, year, and whether the measurement is transaction count or total address volume.

2025 saw approximately 33.4 million transferred addresses globally, while 2026 RIPE NCC data shows substantial but uneven monthly activity.

A single global trend should therefore be interpreted carefully.

Do IPv4 transfers fragment address space?

Sometimes.

APNIC's analysis found that about 26% of transfer entries from 2012 through the start of 2026 involved blocks smaller than the original allocation.

Registry fragmentation does not automatically mean identical fragmentation in global BGP routing.

Does continued IPv4 demand mean IPv6 adoption is not working?

No.

IPv4 demand and IPv6 deployment can coexist.

Many networks continue to require IPv4 compatibility while expanding IPv6 deployment.

Conclusion

The most important lesson from 2026 IPv4 transfer data is not simply that addresses still have market value.

It is that IPv4 has become a resource with an active secondary lifecycle.

Previously allocated addresses do not necessarily remain permanently attached to their first network.

They can be reassessed.

They can move.

They can be divided.

Their registry records can change.

Their routing can change.

And they can eventually support a new generation of services and users.

The 2025 global transfer volume of roughly 33.4 million addresses and the more than 16.7 million addresses recorded as transferred in the RIPE NCC service region during the first seven months of 2026 show that this redistribution process remains significant.

At the same time, transfer statistics should be interpreted carefully.

They do not capture every lease, every private arrangement, every unmet request, or every IPv4 address waiting to be reused.

They are one layer of evidence.

Combined with waiting-list data, allocation history, routing information, and operational records, however, they reveal an Internet adapting to scarcity through reuse.

For the LARUS Foundation, that is the more important story.

IPv4 scarcity is not only about how many addresses remain.

It is also about how effectively existing address space can move through its lifecycle while registry information, routing systems, and running networks remain understandable and reliable.

That makes address reuse not merely a market phenomenon, but an important part of modern Internet resource management.