The ISP Column

A column on things Internet

Other Formats:

---

** IP Addresses through 2025 ** January 2026

Geoff Huston

It’s time for another annual roundup from the world of IP addresses. Let’s see what has changed in the past 12 months in addressing the Internet and look at how IP address allocation information can inform us of the changing nature of the network itself.

Back around 1992, the IETF gazed into their crystal ball and tried to understand how the Internet was going to evolve and what demands that would place on the addressing system as part of the “IP Next Generation” study. The staggeringly large numbers of connected devices that we see today were certainly within the range predicted by that study. The assumption made at the time was that we would continue to use much the same IP protocol architecture, including the requirement that each connected device was assigned a unique IP address, and the implication was that the 32-bit address field defined in version 4 of the IP protocol was clearly going to be inadequate to cope with the predicted number of connected devices. A span of 4 billion address values was just not large enough.

We concluded at the time that the only way we could make the Internet work across such a massive pool of connected devices was to deploy a new IP protocol that came with a massively larger address space. It was from this reasoning that IPv6 was designed, as this world of abundant silicon processors connected to a single public Internet was the scenario that IPv6 was primarily intended to solve. The copious volumes of a 128-bit address space were intended to allow us to uniquely assign a public IPv6 address to every such device, no matter how small, or in whatever volume they might be deployed.

But while the Internet has grown at amazing speeds across the ensuing 33 years, the deployment of IPv6 has proceeded at a more measured pace. There is still no evidence of any common sense of urgency about the deployment of IPv6 in the public Internet, and still there is no common agreement that the continued reliance on IPv4 is failing us.

Much of the reason for this apparent contradiction between the addressed device population of the IPv4 Internet and the actual count of connected devices, which is of course many times larger, is that through the 1990’s the Internet rapidly changed from a peer-to-peer architecture to a client/server framework. Clients can initiate network transactions with servers but are incapable of initiating transactions with other clients. Servers are capable of completing connection requests from clients, but cannot initiate such connections with clients. Network Address Translators (NATs) are a natural fit to this client/server model, where pools of clients share a smaller pool of public addresses, and only require the use of an address once they have initiated an active session with a remote server. NATs are the reason why a pool of excess of 30 billion connected devices can be squeezed into a far smaller pool of some 3 billion advertised IPv4 addresses. Services and Applications that cannot work behind NATs are no longer useful in the context of the public Internet and no longer used as a result. In essence, what we did was to drop the notion that an IP address is uniquely associated with a device’s identity, and the resultant ability to share addresses across clients largely alleviated the immediacy of the IPv4 addressing problem for the Internet.

However, the pressures of this inexorable growth in the number of deployed devices connected to the Internet implies that the even NATs cannot absorb these growth pressures forever. NATs can extend the effective addressable space in IPv4 by up to 32 ‘extra’ bits using mapping of the 16-bit source and destination port fields of the TCP and UDP headers, and they also enable the time-based sharing of these public addresses. Both of these measures are effective in stretching the IPv4 address space to encompass a larger client device pool, but they do not transform the finite IP address space into an infinitely elastic resource. The inevitable outcome of this process, if it were to be constrained to operate solely within IPv4, is that we would see the fragmenting of the IPv4 Internet into a number of disconnected parts, probably based on the service ‘cones’ of the various points of presence of the content distribution servers, so that the entire concept of a globally unique and coherent address pool layered over a single coherent packet transmission realm would be foregone.

Alternatively, we may see these growth pressures motivate the further deployment of IPv6, and the emergence of IPv6-only elements of the Internet as the network itself tries to maintain a cohesive and connected whole. There are commercial pressures pulling the network in both of these directions, so it’s entirely unclear what path the Internet will follow in the coming years, but my (admittedly cynical and perhaps overly jaded) personal opinion lies in a future of highly fragmented network, as least in terms of the underlying packet connectivity protocol.

Can address allocation data help us to shed some light on what is happening in the larger Internet? Let’s look at what happened in 2025.

IPv4 in 2024

It appears that the process of exhausting the remaining pools of unallocated IPv4 addresses is proving to be as protracted as the process of the transition to IPv6, although by the end of 2021 the end of the old registry allocation model had effectively occurred with the depletion of the residual pools of unallocated addresses in each of the Regional Internet Registries (RIRs).

It is difficult to talk about “allocations” in today’s Internet. There are still a set of transactions where addresses are drawn from the residual pools of RIR-managed available address space and allocated or assigned to network operators, but at the same time there are also a set of transactions where addresses are traded between network in what is essentially a "sale". These address transfers necessarily entail a change of registration details, so the registry records the outcome of a transfer, or sale, in a manner that is similar to an allocation or assignment.

If we want to look at the larger picture of the amount of IPv4 address space that is used or usable by Internet network operators, then perhaps the best metric to use is the total span of allocated and assigned addresses, and the consequent indication of annual change in the change in this total address span from year to year.

What is the difference between "allocated" and "assigned"?

When a network operator or sub-registry has received an allocation it can further delegate that IP address space to their customers along with using it for their own internal infrastructure. When a network operator has received an assignment this can only be used for their own internal infrastructure. [https://help.apnic.net/s/article/Using-address-space]

I personally find the distinction between these two terms somewhat of an artifice these days, so from here on I’ll use the term “allocation" to describe both allocations and assignments.

The total IPv4 allocated address pool contracted by some 237 thousand addresses in 2025, with some 3.687 billion allocated addresses at the end of the year. This represented a contraction of some 0.01% for the total allocated IPv4 public address pool through 2025 (Table 1).

| | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | Address Span (B) | Annual Change (M) | Relative Growth | | | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | —————— | —————–– | ————— | | 3.395 | 3.483 | 3.537 | 3.593 | 3.624 | 3.643 | 3.657 | 3.657 | 3.682 | 3.684 | 3.685 | 3.687 | 3.686 | 3.687 | 3.687 | | 168.0 | 88.4 | 53.9 | 55.9 | 30.6 | 19.4 | 13.2 | 0.6 | 24.9 | 2.2 | 1.1 | 1.6 | -0.4 | 1.2 | -0.2 | | 5.2% | 2.6% | 1.5% | 1.6% | 0.85% | 0.53% | 0.36% | 0.02% | 0.68% | 0.06% | 0.03% | 0.04% | -0.01% | 0.03% | -0.2% |

Table 1 - IPv4 Allocated addresses by Year

Have we exhausted all the available sources of further IPv4 addresses? The address management model is that unallocated addresses are held in a single pool by the Internet Assigned Numbers Authority, and blocks of addresses are passed to RIRs, who then allocate them to various end entities, either for their own use or for further allocation. The IANA exhausted the last of its available address pools some years ago, and these days it holds just 3 /24 address prefixes, and has done do for the past 13 years. Because the option of dividing this tiny address pool into 5 equal chunks of 153.6 individual address is not viable, then these 768 individual IPv4 addresses are likely to sit in the IANA Recovered Address registry for some time.

That is, until one of more of the RIRs return more prefixes recovered from the old “legacy" allocated addresses to the IANA, who would then be able to divide the pool equally and distribute them to each the 5 RIRs. This is unlikely to occur.

There are also addresses that have been marked by the IANA as reserved for "special uses"". This includes blocks of addresses reserved for Multicast use. At the top end of the IPv4 address space registry there is a set of addresses that are marked as reserved for "Future Use"". This is a relatively large pool of 268,435,456 addresses (the old former “Class E" space) and if ever there was a “future" for IPv4 then it has well and truly come and gone. But exactly how to unlock this space and return it to the general use pool is a problem that so far has not found a generally workable solution, although efforts to do so have surfaced in the community from time to time.

The topic of releasing the Class E space for use in the public Internet as globally routable unicast address space has been raised from time to time over the past 15 years or so. Some Internet drafts were published for the IETF’s consideration that either directly proposed releasing this space for use, or outlined the impediments in various host and router implementations that were observed to exist in 2008 when these drafts were being developed.

The proposals lapsed, probably due to the larger consideration at the time that the available time and resources to work on these issues were limited and the result of effort spent in ‘conditioning’ this IPv4 space for general use was only going to obtain a very small extension in the anticipated date of depletion of the remaining IPv4 address pools, while the same amount of effort spent on working on advancing IPv6 deployment was assumed to have a far larger beneficial outcome.

From time to time this topic reappears on various mailing lists and blogs (see https://www.potaroo.net/ispcol/2024-09/2404.html, for example), but the debates tend to circle around this same set of topics one more time, and then lapse.

As the IANA is no longer a source of addresses, then we need to look at the RIR practices to see the life cycle of addresses from the registry’s perspective. When IP address space is returned to the RIR or reclaimed by the RIR according to the RIR’s policies it is normally placed in a RIR-reserved pool for a period of time and marked as reserved by the RIR. Marking returned or recovered addresses as reserved for a period of time allows various address prefix reputation and related services, including routing records, some time to record the cessation of the previous state of the addresses prefix, prior to any subsequent allocation. Following this quarantine period, which has been between some months and some years, this reserved space is released for re-use.

The record of annual year-on-year change in allocated addresses per RIR over the same fourteen-year period is shown in Table 2. There are some years when the per-RIR pool of allocated addresses shrunk is size. This is generally due to inter-RIR movement of addresses, due to administrative changes in some instances and inter-RIR address transfers in others.

| | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | APNIC | RIPE NCC | ARIN | LACNIC | AFRINIC | TOTAL | | | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | –– | —– | –––– | –– | —— | —–– | ——— | | 101.0 | 0.6 | 1.2 | 4.6 | 7.4 | 6.7 | 3.2 | 0.4 | 10.5 | 1.7 | 1.5 | 0.8 | -1.1 | -0.8 | 0.6 | | 40.5 | 37.8 | 1.0 | 33.8 | 4.7 | 4.1 | 3.7 | 0.3 | 12.0 | 0.4 | 2.5 | 4.7 | 6.2 | 0.5 | 8.9 | | 53.8 | 24.3 | 19.0 | -14.1 | 2.3 | -4.8 | -2.3 | -0.3 | -10.1 | -0.9 | -1.7 | -3.8 | -5.5 | -3.0 | -2.9 | | 13.6 | 17.3 | 26.3 | 18.7 | 1.2 | 1.5 | 1.4 | 0.1 | 2.4 | 1.2 | -0.2 | -0.3 | -0.1 | 0.0 | -7.0 | | 9.4 | 8.5 | 6.3 | 12.8 | 15.0 | 11.9 | 7.1 | 0.2 | 10.1 | -0.2 | -0.9 | 0.2 | 0.1 | 0.0 | 0.2 | | 218.3 | 88.5 | 53.8 | 55.8 | 30.6 | 19.4 | 13.1 | 0.7 | 24.9 | 2.2 | 1.2 | 1.6 | -0.4 | 1.2 | -0.2 |

Table 2 – Annual change in IPv4 Allocated addresses (millions) - Distribution by RIR

Each of the RIRs are running through their final pools of IPv4 addresses. At the end of 2025, across the RIR system there are some 3.9 million addresses are in the Available Pool, held mainly in APNIC (3.1 million) and AFRINIC (773 thousand). Some 11.2 million addresses are marked as Reserved, with 5.6 million held by ARIN and 4.5 million addresses held by AFRINIC. As seen in Table 3, there has been a reduction in the Reserved Pool for all RIRs, except AFRINIC, and the major reductions were seen in APNIC (1.7M) and ARIN (600K) in ARIN (98K).

Available Reserved RIR202320242025 202320242025 APNIC2,469,1203,647,4883,107,392 2,202,624416,256465,152 RIPE NCC1,0242561,536 708,872677,160782,440 ARIN8,9603,84066,560 5,213,1844,609,7925,424,640 LACNIC2561,5362,304 151,296118,528118,528 AFRINIC1,201,664990,976773,631 4,186,1124,443,6484,480,512 TOTAL3,681,0244,644,096****3,951,424 12,462,08810,265,38411,271,272 Table 3 – IPv4 Available and Reserved Pools, December 2023 – December 2025

The RIR IPv4 address allocation volumes by year are shown in Figure 1, but it is challenging to understand precisely what is meant by an allocation across the entire RIR system as there are some subtle but important differences between RIRs, particularly as they relate to the handling of transfers of IPv4 addresses.

In the case of ARIN, a transfer between two ARIN-serviced entities is conceptually treated as two distinct transactions: a return of the addresses to the ARIN registry and a new allocation from ARIN. The date of the transfer is recorded as the new allocation date in the records published by the RIR. Other RIRs treat an address transfer in a manner analogous to a change of the nominated holder of the already-allocated addresses, and when processing a transfer, the RIR’s records preserve the original allocation date for the transferred addresses. When we look at the individual transaction records in the published RIR data, and collect then by year, then in the case of ARIN the collected data includes the volume of transferred addresses that were processed in that year, while the other RIRs only include the allocations performed in that year.

In order to provide a view across the entire system, it’s necessary to use an analysis approach that can compensate for these differences in the ways RIRs record address transactions. In this study, an allocation is defined here as a state transition in the registry records from reserved or available to an allocated state. This is intended to separate out the various actions associated with processing address transfers, which generally involve no visible state change, as the transferred address block remains allocated across the transfer, from address allocations. This is how the data used to generate Figure 1 has been generated from the RIR published data, comparing the status of the address pools at the end of each year to that of the status at the start of the year. An allocation in that year is identified as allocated in that year if the address block was not registered as allocated at the start of the year.

Figure 1 – IPv4 Address Allocations by RIR by year

The number of RIR IPv4 allocations by year, once again generated by using the same data analysis technique as used for Figure 1, are shown in Figure 2.

Figure 2 – IPv4 Allocations by RIR by year

It is clear from these two figures that the average size of an IPv4 address allocation has shrunk considerably in recent years, corresponding to the various IPv4 address exhaustion policies in each of the RIRs.

What’s Left in the IPv4 Address Pools?

To recap, when addresses are held by an RIR they are classified into one of three states:

• Available, indicating that the address block is available for allocation under the terms of the prevailing address allocation policies adopted by the community that is served by that RIR,
• Allocated, indicating that the address block has been allocated to an entity, and
• >Reserved, indicating that the address block is held by the RIR, but is not available for allocation at this point in time. The Reserved category covers a number of scenarios, depending on the RIR’s procedures, including the holding of an address block in a form of quarantine after its recovery by the RIR before declaring the address to be available.

The pool size of available addresses over the past five years for each RIR is shown Figure 3.

Figure 3 – IPv4 Available Pool Sizes by RIR by day – 2020 - 2026

Only APNIC and AFRINIC are operating with relatively large pools of available addresses. At the start of 2026 APNIC had some 2,849 address blocks in its registry that were marked as available, with a total pool size of 3.095M addresses. AFRINIC has 19 address blocks similarly marked, with a total pool size of 0.765M addresses. For both of these RIRs, the allocation rates from these pools are small, and even without any further returns of addresses these available address pools will likely last from some years to come at current allocation rates.

There are some 11.169M addresses in the RIRs’ reserved address pools. Between 2020 and 2025 APNIC reduced the size of its reserved address pool by some 4M addresses, and the current reserved pool is 0.454M addresses in APNIC. Both RIPE NCC and LACNIC have similarly small reserved address pools these days. The majority of the total pool of reserved address space lies with AFRINIC, which has 543 separate reserved address blocks with a total span of 4.481M addresses, and ARIN, which has 3,765 such address blocks with a total span of 5.2865M addresses. The AFRINIC pool size has been slowly increasing in size, while the ARIN pool size was declining up to 2025, and increased in size through 2025 (Figure 4).

Figure 4 – IPv4 Reserved Pool Sizes by RIR by day – 2020 - 2026

At the start of 2026, 45% of the total pool of 3.687B allocated IPv4 addresses are held in ARIN’s registry, 24% in APNIC’s registry, 23% in the RIPE NCC, 5% in LACNIC and 3% in ARINIC.

IPv4 Address Transfers

The RIRs permit the registration of IPv4 transfers between address holders, as a means of allowing secondary re-distribution of addresses as an alternative to returning unused addresses to the registry. This has been in response to the issues raised by IPv4 address exhaustion, where the underlying motivation as to encourage the reuse of otherwise idle or inefficiently used address blocks through the incentives provided by a market for addresses, and to ensure that such address movement is publicly recorded in the registry system.

The number of registered transfers in the past eleven years is shown in Table 4. This number of transfers includes both inter-RIR and intra-RIR transfers. It also includes both the merger and acquisition-based transfers and the other grounds for of address transfers. Each transfer is treated as a single transaction, and in the case of inter-RIR transfers, this is accounted in the receiving RIR’s totals.

Receiving RIR2013201420152016201720182019202020212022202320242025 APNIC180307451840845491533820785745796752706

RIPE NCC1711,0542,8362,3732,4513,7754,2214,6965,7424,6404,9375,2154,196

ARIN 3222626689415014197185703

LACNIC 2 3917201712

AFRINIC 172726805814152

Total3511,3613,2903,2353,3224,3114,8495,6396,7665,6015,8646,1845,619 Table 4 - IPv4 Address Transfers per year

The differences between RIRs reported numbers are interesting. The policies relating to address transfers do not appear to have been adopted to any significant extent by address holders in AFRINIC and LACNIC serviced regions, while uptake in the RIPE NCC service region appears to be very enthusiastic!

A slightly different view is that of the volume of addresses transferred per year (Table 5).

Receiving RIR	2013	2014	2015	2016	2017	2018	2019	2020	2021	2022	2023	2024	2025	APNIC	RIPE NCC	ARIN	LACNIC	AFRINIC	Total
2.3	4.1	6.6	8.2	4.9	10.0	4.3	16.6	6.5	3.7	2.7	2.5	2.4
2.0	9.6	11.6	9.2	24.6	19.5	26.9	18.2	16.2	36.9	20.8	23.0	22.3
		0.1	0.3	0.2	-	0.3	0.2	0.2	3.1	1.6	4.5	8.4
					-		-	-	-	0.0	0.0	0.1
					0.2	0.5	1.2	3.4	0.5	0.1	0.1	0.0
4.3	13.7	18.2	17.6	29.6	29.7	31.9	36.2	26.4	44.3	25.3	30.2	33.4

Table 5 – Volume of Transferred IPv4 Addresses per year (Millions of addresses)

A plot of these numbers is shown in Figures 5 and 6.

Figure 5 – Number of Transfers: 2013 - 2025

Figure 6 – Volume of Transferred Addresses: 2013 - 2025

The volumes of transferred addresses reached a peak in 2022 and declined in 2023. In the case of APNIC the peak occurred in 2020, and the APNIC 2024 volume is comparable to the volume transferred in 2013. In the ARIN region address transfers are growing in total volume, while in APINC the volume of IPv4 address transfers has largely waned.

The aggregate total of addresses that have been listed in these transfer logs since 2012 is some 342 million addresses, or the equivalent of 20.4 /8s, which is some 9.3% of the total delegated IPv4 address space of 3.7 billion addresses. However, that figure is likely to be an overestimate, as a number of address blocks have been transferred multiple times over this period.

Are Transfers Performing Unused Address Recovery?

This data raises some questions about the nature of transfers. The first question is whether address transfers have managed to be effective in dredging the pool of allocated but unadvertised public IPv4 addresses and recycling these addresses back into active use.

It was thought that by being able to monetize these addresses, holders of such addresses may have been motivated to convert their networks to use private addresses and resell their holding of public addresses. In other words, the opening of a market in addresses would provide incentive for otherwise unproductive address assets to be placed on the market. Providers who had a need for addresses would compete with other providers who had a similar need in bidding to purchase these addresses. In conventional market theory the most efficient user of addresses (here “most efficient” is based on the ability to use addresses to generate the greatest revenue) would be able to set the market price. Otherwise unused addresses would be put to productive use, and as long as demand outstrips supply the most efficient use of addresses is promoted by the actions of the market. In theory.

However, the practical experience with transfers is not so clear. The data relating to address re-cycling is inconclusive. In the period 2000 to 2010, the pool of unadvertised assigned IP4 addresses increased in size from 600M to 900M addresses, which was almost one third of the assigned address pool. In the ensuring 11 years to pool of unadvertised assigned addresses fell to around 800M addresses, with the bulk of that reduction occurring in 2015. There was a substantial reduction in the size of this unadvertised address pool at the start of 2021, due to the announcement in the Internet’s routing system of some seven /8s from the address space originally allocated to the US Department of Defence in the early days of the ARPANET. At the end of 2021 AS749 originated more IPv4 addresses than any other network, namely some 211,581,184 addresses, or the equivalent of a /4.34 in prefix length notation, or some 5% of the total IPv4 address pool. Across 2022 and 2023 the previous trend of an increasing pool of unadvertised addresses resumed. On December 12 2024, a total of some 81,224,704 addresses (the equivalent of 4.8 /8s) was advertised by ASes operated by Amazon, mainly AS16509, bringing the total pool of unadvertised addresses down to a level last observed in the year 2000. Across 2025 the pool of unadvertised assigned IP4 addresses has increased slightly (Figure 7).

Figure 7 – IPv4 Unadvertised Address Pool Size

The larger picture of the three IPv4 address pool sizes, allocated, advertised and unadvertised address pools since the start of 2000 is shown in Figure 6. The onset of more restrictive address policies coincides with the exhaustion of the central IANA unallocated address pool in early 2011, and the period since that date has seen the RIRs run down their address pools.

Figure 8 – IPv4 Address Pools 2000 - 2025

We can also look at the year 2025, looking at the changes in these address pools since the start of the year, as shown in Figure 9. The total span of advertised addresses fell by a total of 10M addresses through the year.

Figure 9 – IPv4 Address Pool changes through 2025

That change in the blue trace in Figure 9, the total allocated address pool, can be attributed to LACNIC, where some 7M addresses were market as available through March 2025, and then passed back into the advertised pool in early April 2025.

Four of the major step changes through the year in the advertised address span (5 January, 5 May, 5 November and 11 December) can be attributed to AS16509 (Amazon 02), and the day-by-day record of the total address span advertised by this AS is shown in Figure 10. Amazon has advertised a total of an additional 12M addresses through 2025, peaking at 168M addresses at the start of November. On the 5th November, Amazon stopped announcing a collection of prefixes with a total span of 15.744M addresses. A month later, on the 10th December some 5.039 M additional addresses were announced bringing the total span of addresses announce by AS16509 Amazon at the year’s end to 157.425 addresses, just 2.777M more than the 154.763M addresses that were announced at the start of 2025.

Figure 10 – Total advertised address span for AS16509 (Amazon-O2) through 2025

In relative terms, expressed as a proportion of the total pool of allocated IP addresses, the unadvertised address pool peaked at 38% of the total assigned address pool in early 2003, and then declined over the ensuing 15 years to a relatively low value of 22% at the start of 2018. The ratio has been steadily climbing since that date, with abrupt falls due to the advertisement of the legacy US Department of Defence address space in 2021, and the Amazon address announcements in December 2024 (Figure 11). The unadvertised address space now sits at some 16% of the total assigned address pool.

Figure 11 – Ratio of Unadvertised Pool Size to Total Pool Size: 2000 - 2026

The data behind Figure 11 gives the impression of a steady effort to recycle otherwise idle IP addresses over the past twenty-five years, and to a certain extent this has been the case. However, the large drop in this ratio in early 2021 was due to the moves by the US DoD to advertise their address holdings to the public Internet, and the large drop in late 2024 was due to Amazon announcing address space that it had previously acquired.

The transfer data points to a somewhat sluggish transfer market. The number of transfer transactions is rising, but the total volume of transferred addresses is falling for most RIRs, with the exception of the RIPE NCC (Tables 4 and 5). The address market does not appear to have been all that effective in flushing out otherwise idle addresses and re-deploying them into the routed network. However, as with all other commodity markets, the market price of the commodity reflects the balancing of supply and demand and the future expectations of supply and demand. What can be seen in the price of traded IPv4 addresses over the past 10 years?

One of the address brokers, Hilco Streambank’s IPv4.Global, publish the historical price information of transactions (if only all the address brokers did the same, as a market with open price information for transactions can operate more efficiently and fairly than markets where price information is occluded). Figure 12 uses the Hilco Streambank IPv4.Global transaction data to produce a time series of address price.

Figure 12 – IPv4 Price Time Series (data from Hilco Streambank IPv4.Global>)

There are a number of distinct behaviour modes in this time series data. The initial data prior to 2016 reflected a relatively low volume of transactions with stable pricing just below $10 per address. Over the ensuing 4 years, up to the start of 2019, the unit price doubled, with small blocks (/24s and /23s) attracting a price premium. The price stabilised for the next 18 months at between $20 to $25 per address, with large and small blocks trading as a similar unit price. The 18 months from mid-2020 up to the start of 2022 saw a new dynamic which was reflective of an exponential rise in prices, and the unit price lifted to between $45 and $60 per address by the end of 2021. The year 2022 saw the average market price drop across the year, but the variance in prices increased and trades at the end of the year were recorded at prices of between $40 to $60 per address.

This price decline continued across 2023, and by the end of 2023 IPv4 addresses were traded at unit prices of between $26 to $40. The prices of addresses across 2024 were relatively stable, but the price decline resumed across 2025, with a low of $9 per address (for a /14) and a mean of $22 per address in the most recent 40 days (up to the 10th of January 2026). The average monthly prices for each prefix size in the most recent 25 months is shown in Figure 13.

Figure 13 – Average unit price per prefix per month over 2024 - 2026

If prices are reflective of supply and demand it appears that the initial period from 2014 to 2022 saw demand increase at a far greater level than supply, and the price escalation reflected some form of perceived scarcity premium being applied to addresses. However, the subsequent price slump shows that this perception was short-lived. These days the low price of $9 per address is back to the same price that was seen in 2014. The difference this time is the range of prices if far greater, and while the mean price is around $22 per address, the price in an individual transaction as high as $34 per address. Generally, larger address blocks fetch a lower price per address in the market, and the January 2026 sale for $9 per address was for a /14 address block.

What is this price data telling us? If you were hanging onto some idle IPv4 address hoping for the price to rise, then you may have missed out! Equally if you were looking to fund your costs in transition from a IPv4-only platform to a dual-stack through the sale of part of your IPv4 address holds, then that opportunity may have already passed you by. If you are forecasting a future demand for more IPv4 addresses in your enterprise then there is no urgency to hit the market and secure IPv4 addresses. If you wait, then the price is likely to drop further.

The largest buyer on in the IPv4 market was Amazon, and it appeared that they were meeting demands from enterprise customers of their cloud-based products, a sector that has been very conservative in their moves to transition into a dual stack situation. I think it’s reasonable also make the supposition that they saw the price escalation in the period 2014 to 2018 as a signal of a longer-term trend, so securing as much as their forecast future need for IPv4 addresses made sense for them in a rising market. However, once the big data centre buyers had secured their address inventory they then existed the market, and the remainder of the buyers had insufficient volume to sustain the price. Demand fell off and the price slumped from the start of 2022 onward.

It’s not as if this IPv4 address market has collapsed completely, and Figure 6 shows that in 2025 some 33M IPv4 addresses were transferred within the RIR registry system. But the declining price suggests that supply is running higher than demand and while buyers appear to be willing to pay a price premium to purchase from a preferred registry or with a preferred provenance, the average price per address has dropped by some 50% across 2025.

Are there any supply-side issues in the market? Is the supply of tradable IPv4 address declining? One way to provide some insight into answering this question is to look at the registration age of transferred addresses. Are such addresses predominately recently allocated addresses, or are they longer held address addresses where the holder is wanting to realise the inherent value in otherwise unused assets? The basic question concerns the age distribution of transferred addresses where the age of an address reflects the period since it was first allocated or assigned by the RIR system.

The cumulative age distribution of transferred addresses by transaction is shown on a year-by-year basis in Figures 14 and 15.

In the period 2019 – 2021 a visible subset of address holders appeared to hold recently allocated addresses for the policy-mandated minimum holding period of some 2 years and then transfer these addresses on the market. In previous years some 8% of addresses that were transferred were originally allocated up to 5 years prior to the transfer. In 2022 this number has fallen to 4%, which is presumably related to the smaller volumes of address allocations in 2022 rather than any change in behaviours of address holders, and in 2023 and 2024 this behaviour has all but disappeared, due to the very small volume of address allocations by the RIRs rather than any change in the behaviour of address holders.

The bulk of transferred addresses in 2025 (more than 55% of the total volume) were originally allocated between 13 and 25 years ago, or between 2000 and 2012.

Figure 14 – Age distribution of transferred addresses

Figure 15 shows the cumulative age distribution of transfer transactions (as distinct from the volume of transferred addresses), and the disparity between the two distributions for 2025 show that recent individual allocations have been far smaller in size but are still being traded. Some 20% of the recorded transfer transactions in 2025 refer to an address prefix that was allocated within the past 7 years, yet these transactions encompass less than 2% of the inventory of transferred addresses in 2025. Some 40% of the volume of transferred addresses were originally allocated 20 or more years ago, while these transactions are recorded in just 28% of the transfers recorded in 2025.

Figure 15 – Age distribution of Transfer Transactions

There appear to be a number of motivations driving the transfer process.

One is when demand is outstrips supply and price escalation is an inevitable consequence. This may motivate some network operators to purchase addresses early, in the expectation that further delay will encounter higher prices. This factor also may motivate some address holder to defer the decision to sell their addresses, in that delay will improve the price. Taken together, these motivations can impair market liquidity and create a feedback loop that causes price escalation. This factor appeared to be a major issue in the period between 2019 and 2022, but these days the opposite is the case and supply far outstrips demand in the addrtess market.

The second factor is IPv6 deployment. Many applications prefer to use IPv6 over IPv4 if they can (the so-called “Happy Eyeballs” protocol for protocol selection). For a dual stack access network this means that the more the services that they use are provisioned with dual stack, then the lower the traffic volume that uses IPv4, and the lower the consumption pressure on their IPv4 CG-NAT address pools, which reduces their ongoing demand for IPv4 address space. This reduced demand for additional IPv4 addresses has an impact on the market price. A falling market price acts as a motivation for sellers to bring their unused address inventory to market sooner, as further delay will only result in a lower price.

The overriding feature of this address market is the level of uncertainty within the market over the state of the IPv6 transition, coupled with the uncertainty over the further growth of the network. This high degree of uncertainty may lie behind the very high variance of individual transfer transaction prices, as shown in Figure 12 for 2025. Have we finally managed to deploy enough network infrastructure in both IPv4 and IPv6 to get ahead of the demand pressures? Are we now looking at a market which is currently saturated with sufficient addresses and associated service platform infrastructure?

Do Transfers Fragment the Address Space?

The next question is whether the transfer process is further fragmenting the address space by splitting up larger address blocks into successively smaller address blocks. There are 56,629 transactions described in the RIRs’ transfer registries from the start of 2012 until the start of 2026, and of these 14,831 entries list transferred address blocks that are smaller than the original allocated block. In other words, some 26% of transfers implicitly perform fragmentation of the original allocation.

These 14,831 transfer entries that have fragmented the original allocation are drawn from 9,231 original allocations. On average the original allocation is split into 1.9 smaller address blocks. This data implies that the answer to this question is that address blocks are being fragmented as a result of address transfers, but in absolute terms this is not a major issue. There are some 253,021 distinct IPv4 address allocation records in the RIRs registries as of the end of 2025, and the fragmentation reflected in 14,831 more specific entries of original allocation address blocks represents around 5.9% of the total pool of allocated address prefixes.

Imports and Exports of Addresses

The next question concerns the international flow of transferred addresses. Let’s look at the ten economies that sourced the greatest volume of transferred addresses, irrespective of their destination (i.e. including ‘domestic’ transfers within the same economy) (Table 6), and the ten largest recipients of transfers (Table 7), and the ten largest international address transfers (Table 8). We will use the RIR-published transfer data for the year 2024 as basis for these tables.

Rank	CC	Addresses	Source Economy
1	US	11,747,840	USA
2	BR	7,113,216	Brazil
3	DE	4,375,552	Germany
4	GB	2,706,432	Ukraine
5	IT	778,240	Italy
6	ES	655,872	Spain
7	IR	488,960	Iran
8	NL	479,232	Netherlands
9	RU	451,072	Russian Federation
10	JP	436,224	Japan
11	AU	369,408	Australia
12	CN	295,936	China
13	UA	283,136	Ukraine
14	HK	280,576	Hong Kong (SAR)
15	IN	277,248	India
16	IE	236,544	Ireland
17	SE	187,136	Sweden
18	KR	184,064	Republic of Korea
19	FR	182,272	France
20	CZ	180,224	Czech Republic

Table 6 – Top 20 Countries Sourcing Transferred IPv4 addresses in 2025

Rank	CC	Addresses	Destination Economy
1	GB	7,581,696	Ukraine
2	US	5,460,736	USA
3	DE	5,201,664	Germany
4	BR	4,082,944	Brazil
5	SG	1,248,768	Singapore
6	IT	1,021,696	Italy
7	RU	777,216	Russian Federation
8	ES	669,184	Spain
9	NL	622,080	Netherlands
10	SE	482,816	Sweden
11	IN	381,952	India
12	HK	343,040	Hong Kong (SAR)
13	FR	329,728	France
14	UA	264,192	Ukraine
15	SA	256,512	Saudi Arabia
16	AE	256,256	United Arab Emirates
17	KR	250,368	Republic of Korea
18	CN	243,712	China
19	ID	240,128	Indonesia
20	JP	229,120	Japan

Table 7 – Top 20 Countries Receiving Transferred IPv4 addresses in 2025

There are many caveats about this data collection, particularly relating to the precise meaning of this economy-based geolocation. Even if we use only the country-code entry in the RIRs’ registry records, then we get a variety of meanings. Some RIRs use the principle that the recorded country code entry corresponds to the physical location of the headquarters of nominated entity that is the holder of the addresses, irrespective of the locale where the addresses are used on the Internet. Other RIRs allow the holder to update this geolocation entry to match the holder’s intended locale where the addresses will be used. It is generally not possible to confirm the holder’s assertion of location, so whether these self-managed records reflect the actual location of the addresses or reflect a location of convenience is not always possible to determine.

When we look at the various geolocation services, of which Maxmind is a commonly used service, there are similar challenges in providing a geographic location service. At times this is not easy to establish, such as with tunnels used in VPNs. Is the “correct” location the location of the tunnel ingress or tunnel egress? Many of the fine-grained differences in geolocation services reflect the challenges in dealing with VPNs and the various ways these location services have responded. There is also the issue of cloud-based services. Where the cloud service uses anycast, then the address is located in many locations at once. In the case where the cloud uses conventional unicast, the addresses use may be fluid across the cloud service’s points of presence based on distributing addresses to meet the demands for the service. The bottom line is that these location listings are a “fuzzy” approximation rather than a precise indication of location.

With that in mind let’s now look at imports and exports of addresses of 2025 transfers where the source and destination of the transfers are in different economies. Some 2,421 transfers appear to result in a movement of addresses between countries, involving a total of 18,729,216 addresses. The 20 largest country pairs are shown in Table 8.

Rank	From	To	Addresses (M)	Source	Destination
1	US	GB	5,880,576	USA	UK
2	BR	US	2,682,880	Brazil	USA
3	US	DE	935,936	USA	Germany
4	DE	US	560,384	Germany	USA
5	BR	SG	458,752	Brazil	Singapore
6	US	SG	454,400	USA	Singapore
7	US	ES	395,264	USA	Spain
8	JP	US	286,720	Japan	USA
9	GB	SE	273,408	UK	Sweden
10	GB	US	235,008	UK	USA
11	US	SA	232,704	USA	Saudi Arabia
12	AU	US	219,648	Australia	USA
13	US	LT	214,016	USA	Lithuania
14	GB	DE	213,248	UK	Germany
15	US	IT	202,752	USA	Italy
16	US	PA	196,864	USA	Panama
17	US	RU	196,608	USA	Russian Federation
18	US	BR	196,608	USA	Brazil
19	US	NL	187,136	USA	Netherlands
20	US	HK	180,224	USA	Hong Kong SAR

Table 8 – Top 20 Economy-to-Economy IPv4 address transfers in 2025

The 2025 transfer logs also contain 3,198 domestic address transfers, with a total of 14,261,856 addresses, with the largest activity by address volume in domestic transfers in the Brazil (4M), Germany (4M), UK (1.5M), US (1.1M), Italy (0.8M) and the Russian Federation (0.4M).

An outstanding question about this transfer data is whether all address transfers that have occurred have been duly recorded in the registry system. This question is raised because registered transfers require conformance to various registry policies, and it may be the case that only a subset of transfers are being recorded in the registry as a result. This can be somewhat challenging to detect, particularly if such a transfer is expressed as a lease or other form of temporary arrangement, and if the parties agree to keep the details of the transfer confidential.

It might be possible to place an upper bound on the volume of address movements that have occurred in any period is to look at the Internet’s routing system. One way to shed some further light on what this upper bound on transfers might be is through a simple examination of the routing system, looking at addresses that were announced in 2025 by comparing the routing stable state at the start of the year with the table state at the end of the year (Table 9).

Jan-25Jan-26DeltaUnchangedRe-HomeRemovedAdded Announcements955,9631,049,90953,946878,09229,69988,172142,118

Root Prefixes:469,272505,73136,459411,63420,62330,32159,826 Address Span (M)3,117.653,106.37-11.282,944.9937.97127.8695.77

More Specifics:526,691544,17817,487294,1269,07657,85182,292 Address Span (M)899.75872.16-27.59773.5019.8481.4989.05 Table 9 – IPv4 BGP changes over 2025

While the routing table grew by 53,946 entries over the year, the nature of the change is slightly more involved. Some 88,172 prefixes that were announced at the start of the year were removed from the routing system at some time through the year, and 142,118 prefixes were announced by the end of the year that were not announced at the start of the year. More transient prefixes may have appeared and been withdrawn throughout the year of course, but here we are comparing two snapshots rather than looking at every update message. A further 29,699 prefixes had changed their originating AS number, indicating some form of change in the prefix’s network location in some manner.

If we look at the complete collection of BGP updates seen from an individual BGP vantage point (AS 131072) across all of 2025 we see a larger collection of transient address prefixes. A total of 1,270,968 distinct prefixes were observed through 2025. That implies that some 221,059 additional prefixes were seen at some point through the year, starting from the initial set as January 2025.

We can compare these prefixes that changed in 2025 against the transfer logs for the two-year period 2024 and 2025. Table 10 shows the comparison of these routing numbers against the set of transfers that were logged in these two years.

TypeListedUnlistedRatio Re-Homed All1,99126,7226.9% Root Prefixes92712,7116.8%

Removed All3,17285,0003.6% Root Prefixes1,86928586.8%

Added All14,457127,57110.2% Root Prefixes12,68647,14021.2% Table 10 – Routing changes across 2025 compared to the Transfer Log Entries for 2024 - 2025

These figures show that some 7% of changes in advertised addresses from the beginning to the end of the year are reflected as changes as recorded in the RIRs’ transfer logs. This shouldn’t imply that the remaining changes in advertised prefixes reflect unrecorded address transfers. There are many reasons for changes in the advertisement of an address prefix and a change in the administrative controller of the address is only one potential cause. However, it does establish some notional upper ceiling on the number of movements of addresses in 2025, some of which relate to transfer of operational control of an address block, that have not been captured in the transfer logs.

Finally, we can perform