[CPWG] Communication to ICANN GC

Sat Aug 26 22:41:53 UTC 2023

Evan,

I am keenly aware of the remit of ALAC under Section 12.2(d)(i) of the ICANN bylaws.  However, I do not view the interests of Internet end users and Registrants as being mutually exclusive.  I have multiple domain names and email addresses, some of which are purely for business purposes and others purely personal.  I have enjoyed participating in CPWG calls for years because I believe they give some of the best objective reviews of the issues being discussed within the community. Do I always agree with them, no, but there is much less spin than other stakeholder groups within the ICANN community.

I FULLY agree with your comments about ICANN funding, in fact, I raised some of these concerns in connection with the Verisign .NET registry agreement renewal. During my time on the ICANN Board (2003-2006) ICANN staff (Paul Twomey, John Jeffrey and Kurt Pritz) were instrumental in changing the original ICANN funding model.  I had significant concerns then, and I believe those concerns have been validated in hindsight based on current trends within the industry.  For those non-dinosaurs on this mailing list, allow me to provide an “extensive” historical context.

ICANN currently operates on what can roughly be described as a tax and spend model. ICANN imposes a mix of fixed and variable fees on Registrars and Registries. In connection with Registrars, they pay an annual $4,000 accreditation fee and a quarterly variable fee ($120/$370 depending upon the domain names under management) and a per domain name ICANN transaction fee of approximately $0.18, see https://static.sched.com/hosted_files/icann76/e0/ICANN76%20Being%20an%20ICANN%20Accredited%20Registrar.pdf. On the Registry side, the vast majority (but not all) of Registries pay ICANN a $25,000 annual fixed fee, and then a $0.25 per domain name transaction fee for domain names transactions in excess of 50,000.  There are a couple of exceptions to this rule of thumb, the most glaring being Verisign which pays ICANN approximately $11 million annually above and beyond the baseline requirements in the standard Registry Agreement in connection with .COM and .NET. So back of the napkin math, the vast majority of ICANN’s 140 million dollar budget comes from these ICANN fees imposed on contracting parties.

Prior to implementing this transactional tax and spend model, ICANN operated on what I would characterize as a cost recovery model.  ICANN would collect fixed accreditation fees from Registrars and Registries and then impose a per domain name variable fee on Registrars to pay for the previous year’s approved budget. There was a tiered pricing structure for Registries based on the number of domain names under management, see Paragraph 3.14 of the baseline non-sponsored registry agreement, see https://www.icann.org/en/registry-agreements/multiple/proposed-unsponsored-tld-agreement-11-5-2001-en. My recollection from the 2003-2005 time frame was that Verisign and Afilias were each paying ICANN approximately $127,000 annually for the operation of the .COM, .NET and .INFO TLDs because of the fee caps provision in Paragraph 3.14.4. On a quarterly basis, ICANN would then impose a per domain name fee on Registrars based upon their domain names under management to collect the remainder of the annual budget. As the number of domain names was regularly increasing during this time, this variable fee fluctuated but generally trended down.

Now there were a couple of problems with this original funding model, the biggest being that ICANN was beholden to getting the annual approval of 2/3 of the accredited registrars (based on market share) for the fees ICANN charged them. During this time Registrar fees were the vast majority of the fees collected by ICANN, as Verisign was only paying ICANN approximately a quarter of a million dollars per year as the Registry Operator of the .COM and .NET TLDs. If ICANN did not get the annual approval from Registrars, ICANN would quickly become insolvent as an organization. While I have had my fair share of disagreements with Kurt Pritz over the years, it generally fell upon his shoulders during this time to corral the registrars into approving the budget/fees. To this day, that probably remained one of the most difficult tasks ICANN staff EVER had to undertake and for that I do respect Kurt.

For those wondering, why didn’t ICANN just collect the registrar's fees from the Registries. Simply put it did not have the contractual authority to do so. To ICANN’s credit, ICANN has amended the baseline registry agreement so that Paragraph 6.3(a) now provides ICANN the ability to collect a variable registry fee via the Registry Operators if the Registrars do not approve the budget, see https://itp.cdn.icann.org/en/files/registry-agreements/base-registry-agreement-30-04-2023-en.html.

Given this conundrum, ICANN was in a very precarious position at a time that it needed to raise its annual budget from 8.3 million to 15.8 million. I would encourage anyone interested to read this article from 2004 that gives an excellent historical analysis, see https://www.computerworld.com/article/2566345/icann-ends-malaysia-meeting-with-budget-approval.html

At this time, many people per the ComputerWorld article thought that ICANN was going to seek additional funds from the ccTLDs. However, ICANN staff had decided a different plan.  As part of its negotiations with the Registrars, ICANN raised the per domain name fee to $0.25 annually with the promise that the fee would be fixed annually. This was of economic interest to the Registrars as they could now pass this fee directly onto Registrants. Because the per domain name fee under the old model was variable, Registrars could not pass this “fixed” fee onto Registrants.  Most Registrars today now include in their invoice to Registrants a per domain name ICANN fee. After ICANN began collecting substantial revenue from the Registries, ICANN lowered the Registrar variable fee to the current $0.18.

During these negotiations, ICANN represented that they would be getting the additional fees associated with the doubling of ICANN’s annual budget from 7 to 15 million from gTLD registries, under the auspices of getting gTLD registries to pay their fair share.  What ICANN staff did not share with the Registrars during these negotiations is what they would be giving up to get Registries to agree to these increased fees. As noted above the baseline non-sponsored registry agreement had fee caps on the prices Registries could charge Registrars, and the potential for a competitive rebid process in connection with each Registry term, see Paragraph 5.2 of the original non-sponsored registry agreement, https://www.icann.org/en/registry-agreements/multiple/proposed-unsponsored-tld-agreement-11-5-2001-en. As evidenced by the Registry Agreements from 2005 onward, ICANN not only granted the ability for Registry Operators to increase their prices, but it also made every gTLD Registry Agreement a de facto monopoly in perpetuity - unless the Registry Operator is really stupid, see Paragraph 4.3 of the current baseline registry agreement, https://newgtlds.icann.org/sites/default/files/agreements/agreement-approved-31jul17-en.html. The other major concession that Registry Operators granted ICANN was the removal of the ICANN indemnification provision, see Section 4.6 of the original baseline non-sponsored Registry Agreement https://www.icann.org/en/registry-agreements/multiple/proposed-unsponsored-tld-agreement-11-5-2001-en

As a result of these ICANN funding and Registry Agreement changes, Verisign went from paying ICANN approximately $250,000 per year in connection with the operation of .COM and .NET, to Verisign paying ICANN over $55 million to operate the .COM and .NET registries.  Today Verisign is paying ICANN in excess of 41 million annually to operate .COM, and in excess of $10 million annually to operate .NET. When you throw in the $4 million dollars annually of the Binding Letter of Intent that equates to 55 million dollars annually and that does not include the Registry fees (a couple of hundred thousand) associated with .NAME, .VERISIGN, and Verisign’s portfolio or defensive IDN TLDs.

Now does anyone in the ICANN community find it odd that Verisign went from paying ICANN a couple of hundred thousand dollars annually in the early 2000s to over 55 million dollars annually today?  Does anyone find it concerning that over 1/3 of ICANN’s total revenue is attributed to one company?  Is anyone in the ICANN community concerned about how this financial dynamic could potentially impact the health and operation of the ICANN multistakeholder model?

Now for the past 20-plus years, this has not been a problem because the zone files during this time have consistently increased. However, with a growing maturity in the domain name market, the industry as a whole has experienced a much lower growth rate. Most problematic for ICANN is the potential for the domain names under management to contract. While this is unlikely in the near foreseeable future, the trend long term does not bode well for ICANN especially as its costs continue to grow steadily.

The removal of price caps from the Registry Agreement, with no potential for ICANN to periodically rebid the TLD Registry Agreement removes any downward pressure on that Registry Operator to constrain pricing.  This is one of the reasons I so strenuously advocated in my .NET public comment period for ICANN to include Section 2.15 of the baseline registry agreement into the .NET Registry Agreement. This section states in relevant part that “[i]f ICANN initiates or commissions an economic study on the impact or functioning of new generic top-level domains on the Internet, the DNS or related matters, Registry Operator shall reasonably cooperate with such study, including by delivering to ICANN or its designee conducting such study all data related to the operation of the TLD reasonably necessary for the purposes of such study requested by ICANN or its designee.”

This provision is missing from both the .NET and .COM registry agreements.  Again I ask anyone “still” reading this email, do they find it a bit odd that a single entity controlling approximately 80% of the gTLD market share AND contributing over 1/3 of ICANN’s annual budget is one of the only gTLD Registry Operators not required to participate in an ICANN economic study?

After reading the NameCheap IRP, I began accumulating publicly available pricing information from gTLDs and ccTLDs and I noticed a very disturbing trend. Specifically, the pricing on gTLDs appears to be rising substantially higher than ccTLDs. I found this interesting for a couple of reasons.  ccTLDs and gTLDs provide the same technical functionality, and in several cases, ccTLDs are actually ran on the same backend infrastructure as gTLDs. While there are different policies that may impact pricing, the following graphic from Neustar’s .US bid package really hit home the divergent pricing on ccTLD and gTLDs, see https://ntia.gov/files/ntia/publications/technical_proposal_volume_3.pdf

[A graph with different colored lines  Description automatically generated]

It is noteworthy that while .US domains I still believe are priced at $6.50, .COM is now $9.59, an almost $2 dollar increase in the approximately 5 years since this chart was published.

So Evan here is my proposal to you. I am a big believer in wanting to be part of the solution as opposed to be part of the problem. Now if we both want to be part of the problem we can engage in a tit for tat email exchange what will likely get us nowhere. Alternatively, I would really like to tap your old ZDNET skillset to help expand my research into the growing disparity between ccTLD and gTLD pricing. As noted above, as a result of changes made back in the early 2000’s to the baseline registry agreements, ICANN’s current tax and spend model is heading for trouble as increasing gTLD prices will likely have an adverse impact on the number of domain names under management. While Registry Operators have the unfettered ability to raise their prices, ICANN is locked into what it can charge gTLD registries.

Let me be VERY clear, I want to work with the ICANN community to help bring this information to the attention of the current ICANN Board. I believe we are confronted with a unique opportunity in the history of ICANN for this Board to critically analyze a number of issues which some ICANN staff have conveniently brushed under the rug in the past.

Hopefully this is something that you and I can reach consensus on and collaborate together within ALAC.

Best regards,

Michael

From: Evan Leibovitch <evan at telly.org>
Sent: Friday, August 25, 2023 12:36 AM
To: mike palage.com <mike at palage.com>
Cc: Steinar Grøtterød <steinar at recito.no>; CPWG <cpwg at icann.org>; John Jeffrey <john.jeffrey at icann.org>
Subject: Re: [CPWG] Communication to ICANN GC

On Thu, Aug 24, 2023 at 3:17 PM mike palage.com<http://palage.com> via CPWG <cpwg at icann.org<mailto:cpwg at icann.org>> wrote:

So we are in agreement that protecting registrants should be a focal point of our work,

Some of "us", that is . Others see such a corrupt consensus as mission creep of the highest order.

As usual.

ALAC has no mandate and thus no standing to protect the interest of registrants. Anyone checking the bylaws will notice that fully half of the GNSO is dedicated to representing the interests of various kinds of registrants. They don't need ALAC as a booster, ALAC has its own unique mandate and community.

With so much energy spent on the bylaw-infringing protection of registrants, no wonder so little gets done on behalf of non-registrant end-users. This disease has, does and will continue to inhibit ALAC from putting all energies and resources into serving its bylaw mandate. If ALAC were doing its job, anyone with a conflict of interest -- making money in the creation, buying or selling of domains -- ought to be disqualified from At-Large decision making. There is already so much self-dealing within ICANN; ALAC could be an oasis of policy ethics but actively chooses not to.
Let alone focal point, prices of Internet domains are no business of ALAC's mandated constituency. I have yet to hear ANY evidence, let alone compelling evidence, that domain pricing is an ALAC issue. Arguably, ALAC's only concern should be that ICANN fees must be sufficient to allow it to understand the needs of end users, enact complementary policies, provide education and enforce protections. And if that means fees that are higher than registrants want, well that's too bad, this becomes an area where end-user interests conflict with those of registrants and the side for which ALAC must advocate is quite explicit.

Pity the diversion of appropriate focus is so predictable. At least it's spoken out loud with no attempt to hide.

- Evan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mm.icann.org/pipermail/cpwg/attachments/20230826/d2f5a982/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 31701 bytes
Desc: image001.png
URL: <https://mm.icann.org/pipermail/cpwg/attachments/20230826/d2f5a982/image001-0001.png>