Month: October 2026

IPv4 transfers in September 2026 increased by 39.3% across ARIN and RIPE NCC, excluding explicitly classified mergers and acquisitions (M&A). Address volume reached 2.13 million, compared with 1.53 million in August. CloudIP drove much of ARIN’s increase, while RIPE recorded growth across more recipients. Including M&A, total volume fell from 7.91 million to 3.71 million addresses.

InterLIR’s separate lease-price dataset also recorded monthly increases: ARIN rose to €0.43 per IP, while RIPE reached €0.42 per IP. To apply these figures to a purchasing or leasing decision, compare the relevant registry and block-size segment: aggregate transfer growth does not establish stronger sale prices or faster sales.

September’s key signal: transfer volume grew 39.3% outside explicitly classified M&A, while the number of unique recipients increased just 4.2%.

September IPv4 transfers: monthly totals with M&A separated

August included approximately 5.30 million addresses moving from Frontier Communications to Verizon Business under M&A. That single source–recipient pair was larger than September’s entire transfer total, making the comparison particularly sensitive to corporate transactions.

Excluding transfers explicitly classified as M&A gives a different picture:

Scope: ARIN and RIPE, August–September 2026, excluding explicitly classified M&A. One duplicate September record has been removed.

Address volume grew much faster than recipient numbers. The 39.3% increase reflects more addresses changing hands, with a smaller expansion in the number of organisations receiving them.

These categories are not a pure measure of independent market purchases: they can still include corporate transfers. An IPv4 transfer updates the registered rights to address resources; it does not necessarily represent a separately negotiated sale. Likewise, grouping records by date, source and recipient helps compare activity but does not establish the number of contracts.

CloudIP drove much of ARIN’s increase

CloudIP received six /16 blocks from IBM on September 15 and three from NCR Voyix on September 23. Together, these transfers covered 589,824 addresses, representing 43.5% of September’s ARIN RESOURCE_TRANSFER volume.

That total exceeded ARIN’s entire monthly increase of 526,080 addresses. IBM also transferred two /16s to Netflix and one to RCS Technologies, bringing its September total to nine /16s across three recipients.

Block size explains how a small number of transfers can dominate monthly totals. A /16 contains 65,536 addresses, equivalent to 256 /24 blocks. Our IPv4 subnet reference provides the address counts for other prefix sizes.

Did growth extend beyond the largest recipients?

CloudIP’s September volume was equivalent to approximately 98% of the combined net increase across ARIN and RIPE. However, removing CloudIP from September alone leaves all of August’s large recipients in the comparison.

Ranking recipients separately within each month provides a more balanced view:

Combined ARIN and RIPE address volume outside explicitly classified M&A. Recipients are ranked separately within each month.

The largest recipient was Sky UK in August and CloudIP in September. Across these three adjusted comparisons, growth remained between 11% and 17%. September’s increase therefore cannot be attributed to CloudIP alone.

Within ARIN specifically, removing its largest recipient from each month leaves growth of 9.6%. Removing only CloudIP from September produces a 7.7% decline against August’s full volume. The results answer different questions because only the first adjusts both months.

More registry entries did not mean more transfer groups

Across the two registries, entries outside M&A rose from 353 to 438, an increase of 24.1%. One transfer explains much of that rise: Canada Post moved 101,888 addresses to Element IP Management Services on September 15, spread across 74 entries.

After grouping ARIN records by date, source and recipient, its count fell from 98 to 96. ARIN’s recipient count remained unchanged at 92. The increase in recorded address volume was therefore much stronger than the change in these measures of participation.

RIPE’s growth involved more recipients

RIPE POLICY volume rose from 702,720 to 779,008 addresses, an increase of 10.9%. Recipients increased from 125 to 133, while groups by date, source and recipient rose from 139 to 153.

FREENET transferred 218,112 addresses to FARLEP-INVEST, RIPE’s largest September recipient. The registry record alone does not establish whether this was an independent market purchase. Excluding RIPE’s largest recipient from each month still leaves September volume 10.8% above August.

Other sizeable September transfers included:

Vodafone ONO’s total transferred volume increased from 8,960 addresses in August to 120,576 in September. More than half went to netcup, with another 32,768 addresses transferred to Anexia.

Large blocks supplied 70% of volume, while smaller blocks remained frequent

Across ARIN and RIPE, blocks of /18 and larger supplied 70% of September’s address volume outside M&A, up from 62% in August. The median volume per registry entry remained at 1,024 addresses, equivalent to a /22.

Smaller /24–/22 prefixes accounted for roughly two-thirds of transferred prefixes in both months. Their combined count rose from 245 to 299, but 55 of September’s prefixes belonged to the single Canada Post transfer. Excluding that transfer, the count was essentially unchanged at 244, and address volume in this range grew by just 0.5%.

RIPE showed a broader increase in this segment: /24–/22 prefixes rose from 159 to 191, while their address volume increased by 22.7%.

RIPE’s middle-sized /21–/19 segment moved in the opposite direction. Its volume fell by 34.6%, or 92,160 addresses, while volume in /18-and-larger blocks rose by 42.9%. RIPE’s overall growth therefore did not extend evenly across block sizes.

IPv4 lease prices: ARIN rose to €0.43 per IP per month

In InterLIR’s lease-price dataset, average monthly prices increased in September: ARIN rose 10.7% to €0.43 per IP, while RIPE edged up 0.6% to €0.42 per IP. ARIN reached its highest level in the April–September period, slightly overtaking RIPE.

April–September 2026 · EUR per IP per month

ARIN

RIPE

Scale starts at €0.00
Top of scale: €0.50

Apr
€0.40
€0.42

May
€0.41
€0.41

Jun
€0.37
€0.42

Jul
€0.41
€0.41

Aug
€0.39
€0.42

Sep
€0.43
€0.42

Source: InterLIR lease-price dataset. Under each month, ARIN is listed first and RIPE second. Labels are rounded to two decimal places; bar heights use the unrounded averages.

Average monthly lease prices in euros per IPv4 address. Percentage changes are calculated from unrounded values, so RIPE’s 0.6% increase is not visible in the rounded prices.

These figures describe InterLIR’s lease-price sample, not a market-wide index or a quote for a specific block. They are separate from the registry transfer data and do not measure IPv4 sale prices. Changes in the sample’s composition may also affect the monthly averages.

How to use September’s figures when comparing IPv4 options

The practical value of this report lies in separating the measures that affect a decision:

• For buyers: compare the required registry and block size. Growth in total transferred addresses does not establish how readily a particular block is available or what it should cost.
• For sellers: assess volume alongside recipient counts. Large transfers can lift monthly totals without a similar increase in the number of potential counterparties.
• For tenants: use the lease averages as context, then compare actual offers and their terms. The dataset does not establish that the same blocks became more expensive.
• For market comparisons: apply the same M&A treatment and recipient exclusions to both months. Removing a large recipient from only one month can change the apparent direction of growth.

September recorded higher transfer volume outside explicitly classified M&A, with growth remaining after the largest recipients were removed from both months. RIPE also recorded more recipients and more small-block transfers. These findings show increased transfer activity, but do not establish rising sale prices or shorter time to sale.

Sources and methodology

This review covers ARIN and RIPE NCC transfer activity for August and September 2026. The underlying registry sources are ARIN’s public transfer log and RIPEstat transfer-history data. Monthly totals and recipient comparisons are InterLIR’s calculations from the compiled dataset.

The analysis separates records classified as RESOURCE_TRANSFER and POLICY from explicitly classified M&A. Corporate transfers may still appear outside the M&A categories. Address volume is calculated from the prefixes actually transferred, rather than the source holders’ original allocations.

The September dataset lists TalkTalk → OXIO, 80.43.64.0/18, on both September 9 and September 14. This review treats those entries as a duplicate and counts the block once.

For each concentration comparison, recipients are ranked separately within each month by received address volume. The largest one, three or ten recipients are then excluded from that month. Regional comparisons use rankings within the relevant registry.

Groups combine entries sharing a date, source and recipient; they are not a count of commercial contracts. Combined unique-recipient figures count recipients across both registries and should not be obtained by adding the regional recipient counts.

The lease-price comparison uses a separate monthly series covering April–September 2026. September changes are calculated as the September average divided by the August average, minus one. The series does not provide a constant-sample price index, transaction-level prices or a measure of market-wide lease activity.