Harms

Disclaimer

Please note that this page is a

DRAFT WORKING DOCUMENT

At this stage it has not been reviewed or approved by the Working Group. Indeed, it is very likely to change substantially before being included in any report.

Background

The working group is compiling a list of possible harms that in some way bear on the issue of Vertical Integration. We are developing an approach to analyze the harms on this list in a variety of ways with the hope that we can uncover new ideas that can become part of consensus recommendations. At this stage, the list of harms is an unfiltered compilation of every harm that anybody has raised -- we are planning to substantially refine and flesh out the list before using any of it as a basis for recommendations.

In short -- it's a very early draft and work in progress.

Analysis approaches

Mikey's first try:

One approach to analyzing the harms would be to answer a series of questions about each (this list of questions is preliminary and only intended as a starting point for discussion):

What are examples of the harm having happened (court cases, compliance activity, sanctions, newspaper accounts, etc.)?
Who was involved and how did they do it?
Who was harmed and what was the financial and/or operational impact?
How long, and how often, has it happened?
Who's proposing that it's a harm and why?
Did this harm happen only in a vertically-integrated environment?
To what extent would vertical separation be an effective means to prevent the harm in the future?

Possible components of analysis

How relevant is vertical integration/separation to the harm
How attractive/interesting is the harm to the prospective bad-guy
Impact of the harm (scope and scale)
Categories of harm
Which harms relate to which proposal
Issues that result
Cost of mitigating the harm
Cost of not mitigating the harm

Analysis Approach

Go out to the community for a list of harms (wiki, poll)
Assign harms to WG members for analysis
Binary forms of questions, conduct a poll, identify areas of agreement and disagreement
Risk assessment (one time or continuing)

Implementation approach and issues

Ongoing risk assessment?
Audit/compliance?
Other?

Possible Harms Arising From Vertical Integration/Co-Ownership of Registries and Registrars

Competition Harms

Decreased competition in Registrar space – reduced cost for a VI registrar will allow lower prices and increase in marketing from other registrars, forcing competitors out of market
Decreased competition in Registry space - Integrated Registry Service Providers (RSP+Rr) could demand equity stake or harsh demands from prospective Registry thus decreasing competition in Registries
Decreased number of Registrars able to offer VI TLD – extremely complex requirements from VI entity for their TLD could keep out all Registrars except the integrated Registrar Registrars have 10 year head start acquiring customer base - If a Registrar were to acquire a Registry and a TLD this would be unfair to current stand-alone Registries
Registrar could easily exert control over market channels for new gTLDs - by requiring costly fees to promote a TLD, unless Registry contracted with Registrars integrated RSP Reduced industry competition - due to increased discrimination against non-affiliated Registries and Registrars
100% vertical integration - or anything goes - negates the justification for registrar accreditation and for consensus policy. Only minimal technical requirements on DNS provisioning and resolution services would be needed.

Availability/Pricing Harms

Almost Free Domain Tasting – A Registry could register domains with an integrated Registrar in that TLD. Once domain is tasted then deleted, the overstock/penalty fee Registrar would pay would go to integrated Registry. Only cost would be ICANN fee lowering costs by >90% Domain Front Running – A Registry can see DNS traffic for all non-existent domains (not registered). With access to this data the integrated entity could identify potentially high value names and monetize them through auctions, traffic sites or secondary market sales. Harms that could result include:
- Unavailability of names
- Higher prices
- Benefits flow to the registrar rather than the registrant
Post Tasting Pricing – A Registry could register the domains with their integrated Registrar and taste the domains, realizing the traffic and ppc value for domain, then price accordingly to the a value assigned based on these factors.
Higher prices - Each gTLD is a monopoly of that name space, competition within that name space has been provided by registrars. Allowing a gTLD to vertically integrate, operate both the TLD and the channel, relieves pressure on the gTLD operator to keep prices low that typically come from competing registrars.
- This will also lead to a lower level of stability, security and service for same reasons as noted above

Data Harms

Deleted domains drop data – A Registry could share the algorithm/schedule to their integrated Registrar of when deleted domains become available, giving Registrar an advantage in picking up domains in the integrated TLD
EPP command data sharing – A Registry could share the check command data coming in from all Registrars (check to see if domain is available) with their integrated Registrar and Registrar could use the data in that TLD for nefarious purposes (warehouse, pricing)

Other Harms (harms that were mentioned on list but could not find details)

Creation of complex structures and relationships will be difficult or impossible to enforce. ICANN will have several new compliance issues to deal with regarding dozens and likely hundreds of new gTLDs - IPv6, DNSSEC, new IP protection mechanisms/tools, and possibly other new rules regarding malicious conduct. Compliance is not merely a matter of money, there is a practical limit to what ICANN the organization or community can optimally keep up with.
Lack of innovation - vertical integration or high levels of co-ownership only further entrench the incumbent registries and registrars, leaving little incentive for new service providers (back end, registrars, etc.) to be created.
Collusion between two parties Unfair marketing relationships between two vertically integrated entities
- “Shelf Space”
- Easier cybersquatting
- Account Lock-ins

Possible Harms Arising From Vertical Separation and Limits on Cross-Ownership

Harms to consumers

Increased pricing to consumers and prevention of price reductions.

A vertically-integrated registry/registrar could pass on lower prices to consumers. Vertical separation rules will therefore increase prices to consumers.
A Chinese-Wall arrangement will mean duplicate staffing, rent, etc. costs for a “separated” registry-registrar, resulting in higher prices.
Registries and registrars will be prevented from merging, thereby preventing marketplace efficiencies and keeping prices artificially high for consumers.

Consumer confusion/difficulty

Because many new TLDs will not be carried by particular registrars, the inability to purchase domains at a vertically- integrated registry-registrar may prevent sales of domain names to people who want to buy them.
Dispersal of responsibility – it will be more difficult to pursue bad actors (e.g., cybersquatters) because there are multiple loci of responsibility in a vertically-separated registry-registrar function.

Harms to businesses

Businesses, either registries, registrars, or those providing registry services (e.g., DNS providers) will be prevented from pursuing business mergers or shareholding arrangements with each other.
Decreased competition in Registry space due to prohibition of registrar service providers in favor of incumbent providers.
Existing businesses, whether registries or registrars, will be restricted from pursuing otherwise legitimate business opportunities.
Furthermore, revisions of strict separation at a later date may negatively impact the ability of restricted parties from competing in the registry market while facing additional competition in their original market.
Some new gTLD registries may not be able to find appropriate registrars for their new gTLD, and, if prevented from selling gTLDs directly to the public, may find either no distribution channel, or a substandard one.
New gTLD registries who do not wish to sell to the public may be forced to work with registrars in order to register names on their own behalf – an obvious absurdity.
Single Registrant, Single User gTLD applicants currently trusting a single provider with their domain business may be forced to find an additional provider for the technical operation of their TLD.
Collusion between two parties. Unfair marketing relationships between two vertically separated entities.
Higher service costs for registry services due to reduced competition.
Prevention of potential applicants business models may prevent their proposed applications altogether.

ICANN-related harms and impacts on ICANN

Cost to ICANN (and trickle-down costs)

Vertical separation and cross-ownership restrictions will need to be enforced. This will increase costs at ICANN. Eventually these costs may find their way to the consumer.
Harder to enforce unknown collusion instead of known cooperations. Increased cost due to higher requirement of investigation.

Legal Liability for ICANN

Restrictions on vertical integration may be held to be restraint of trade in some jurisdictions, and ICANN may be sued on these grounds.

Impacts on ICANN

There is "harm" to ICANN's representation value in allowing presently contracted parties to determine the prevailing modes of service offered by future contracted parties. The harm is to the regional diversity interest, and also to the diversity of services interest.
There is "harm" to ICANN's competition value in allowing beneficiaries of the legacy monopoly contract, and the beneficiaries of the 2001 and 2004 rounds of registry contract expansion, perpetual beneficial interest in every subsequent round of registry contract expansion.
The harm of vertical integration is that it captures the first, and perhaps only, 201x round of registry contract expansion. While it is incomplete, in the sense that it does not prevent ICANN from contracting for abstract registry operations with independent applicants, it does restrain ICANN from contracting for registry operations likely to involve more than marginal market share with any but the current contracted parties, whether contracted as registry operators, or as registrars.

Difficulty, Complexity and Cost of Monitoring: Compliance and Enforcement Overview

A significant number of VI WG Members, in Brussels and online discussions, expressed their belief that compliance and enforcement is the most important concern for ICANN and the ICANN Community on this issue.
Ease of detection, monitoring and enforcement are deemed by many to be priorities -- not involving ICANN in new levels of compliance/enforcement responsibilities which are extremely complex, time-consuming, and expensive.
VI/Co-Ownership are seen as adding compliance and enforcement ICANN's existing duties which are complex, time-consuming and expensive. Such duties, according to VI WG work, would likely include:
- Data Audits, both cutting-edge in their field, and global in their scope and reach (as the same rules, and enforcement, apply to all);
- Record review of the considerable paperwork to be generated by Registries and Registrars as part of the much more detailed compliance process;
- Advice line, running a resource for companies and actors wishing to share information with ICANN;
- Other forms of compliance and monitoring; and
- Penalties, including the formulation and implementation of penalties for a range of breaches and violations.
For VI/Co-Owned companies, new responsibilities are seen by the VI WG to include:
- Complex recordkeeping;
- Chinese walls;
- Clear accountability and responsibility of senior management for compliance;
- Training of all employees;
- Active screening/sampling for potential problems; and
- Internal Audits.
Possible harms:
- Increased costs to registrants (presuming that these higher ICANN, registry and registrar costs are passed along)
- Unable to provide assurances to the ICANN community (because of complexity of delivering compliance)
Possible benefits:
- Reduced fraud
- Lower overall costs
- Reduced damage to the industry
Critical success factors
- A clear strategy from ICANN regarding the role, priority and resources available to the compliance/enforcement function
- ICANN ability to efficiently and effectively deliver the required compliance and enforcement capabilities

Annex -- Original drafts and references

Original drafts

Antony Van Couvering 16 August draft of some harms that will be caused by VI/CO restrictions: VI WG AVC 16 Aug FIRST DRAFT HARMS CAUSED BY VI-CO.pdf

Scott Austin 6 August memo to the VI WG on Potential Harms: VI WG S Austin 6 Aug Memo_to_VIWG_re_Potential_Harms.pdf

Jeff Eckhaus 9 August Summary of Potential Harms from WG participants: VI WG J Eckhaus 9 Aug Harms Projectv1.pdf

Harms from Register.com comments

Affiliated entities have a reduced true cost for each registration because the registry fee is not fully realized. This allows the affiliated registrar to sustain lower prices, or spend more on bundled services or marketing than other registrars, potentially forcing competitors out of the market.
An affiliated registrar would have de facto better access to registry systems, deleted names and operation support. Similar to the above this would allow the affiliated registrar to receive additional benefits that are not available to competitors which the affiliated registrar could pass on to its customers.
A registry could institute registrar requirements which on their face do not discriminate, but which in execution make it difficult for all but the affiliated registrars from obtaining approval. For example overly complex API requirements could have the impact of either raising costs to unaffiliated registrars or, in the extreme, keeping other registrars out of the market.

Harms from Lubsen letter to PDT (Afilias)

If allowed to go forward, this proposed deregulation will facilitate “insider trading” that will open the door to abusive domain registration practices and higher domain name prices for some registrants. It will provide the affiliated registrar access to sensitive registry data that includes the entire universe of data for potential and existing domain names from all registrars that sell the TLD. A registry has the unique power to see DNS traffic in its domain; with access to this data, an affiliated registrar would be in a unique position to identify potentially high value names and monetize them through auctions, traffic sites or secondary market sales
Domain name tasting and front running are just two recent examples of the type of abusive practices that will result if a vertically integrated registry can sell its TLD through its affiliated registrar. Both practices could result in registrars withholding valuable names from average registrants and have the effect of raising prices for the average consumer who seeks to register names in what is supposed to be a first-come-first-serve system. ICANN’s proposal will make it possible for a combined registry-registrar entity to eliminate nearly the entire financial penalty on tasting. A vertically integrated registry registrar eliminates 94% of the current cost imposed on a registrar that engages in tasting

Harms from Raad letter to PDT (PIR)

A large Registrar could easily exert control over market channels for new gTLDs by requiring costly fees to promote a TLD, unless the registry contracted with the Registrar’s affiliated back end service provider. The provider could even demand an equity stake or other interest in the Registry. Registry operators may feel compelled to accept harsh contractual terms to ensure market exposure and a likelihood of success. Rather than increasing competition, cross-ownership, in this example, could significantly reduce competition by creating an oligopoly market consisting exclusively of large Registrar. In anticipation of the new gTLDs, Registrars have been actively luring prospective applicants for Registries to serve as their back-end Registry operator by promising to deliver an established customer base, prominence on the Registrar’s website and guaranteed marketing designed to reach ultimate end users. In other words, these Registrars can promise to deliver exactly what existing Registries cannot - a direct marketing relationship with the ultimate registrant
Allowing the existing Registry Operators to serve as Registrars and vice versa, immediately prior to a new round of TLDs being selected, is also fundamentally unfair in that the Registrars have had a ten year head start in acquiring large customer bases. This head start will take years to overcome by the existing Registry Operators

Sapington presentation

Reduced industry competition, due to increased discrimination against non-affiliated registries and registrars Limit the access of non-affiliated registries to prime shelf space

Links

Vertical Integration PDP Working Group (VI WG)