[council] Re: [registrars] Regarding transition to market forces

[Ross Rader]

Bruce Tonkin wrote:
> Hello All,
> 
> I find this question and answer interesting:
> 
> 
> "Q: 1.2 Why is it appropriate for VeriSign to be allowed to raise its
> prices by up to 7% per year?
> 
> A: 1.2 In order to provide for a transition to allowing market forces to
> determine prices, ICANN and VeriSign agreed to relax the current price
> cap, which has remained unchanged since ICANN came into existence, on a
> graduated basis"
> 
> I would argue that if market forces were being applied then the price
> should drop by at least 7% a year.
> 
> Thus I think it would be reasonable as part of a "transition" to market
> forces that the maximum price should be DROPPING 7% per year.
> 
> I am not aware of market forces leading to an increase.  I certainly
> haven't seen that amongst registrars where market forces do apply.

Market forces do not apply when there is only a sole source provider. 
There are no incentives for Verisign to drop its prices, improve its 
services, or otherwise behave as they would in a competitive market.

The few times when circumstances have forced competition in the registry 
market (rebids, new TLD bids, etc.) the community has seen great benefit 
- EPP was introduced, 5 minute DNS propogation became the norm, .net 
prices were decreased.

I am reminded of a conversation that I had with NSI back in 1999 just 
after the accreditation test-bed program was announced and prior to the 
announcement of the test-bed participants - back when they were also the 
Registry, before Verisign stepped into the picture.

At the time Tucows was an extremely small domain reseller for Network 
Solutions. We were number 85 or so in terms of domains under management 
- which at the time meant low tens of thousands. I called up our sales 
rep at NSI and let him know that we had applied for accreditation but 
were uncomfortable making the huge technical investment needed to become 
a registrar. In fact, the only thing that was forcing us down that path 
were the exorbitant prices that NSI was charging their resellers. I 
explained to him that we would be happy to remain a reseller if we could 
just get a better price from them. Somewhere around the $12 mark would 
have kept us happy at the time (this was when registrars were 
anticipating paying $18 for a two year reg and NSI was still charging 
its resellers something close to $35 per year.)

The net of his rather candid response was that because NSI controlled 
the registry and the registrar, and that the registry was the only place 
to get .com names, that there was no way that we were going to get a 
discount because the registry was going to get our money no matter 
whether we went through the registry or the registrar - and in fact, it 
was better for them if customers like us became accredited because then 
we would cost less to support!

ICANN's competitive march has been a boon for the registry. They have 
been allowed to strip off the expensive part of their business (the 
registrar that had to deal with the messy business of customer service 
to end-users, et al) and keep the cost-efficient part of the business. 
Even if they had kept the more costly aspect of their business (and I'm 
not advocating that they should have been allowed to) they have also 
benefitted greatly in the reduction of technology costs over the same 
period of time...

Here are some interesting "cost of technology" facts:

  - On a cost per MIPS basis, mainframe personnel costs are declining by 
10 percent a year, driven by productivity increases. At 25 percent 
annually, the declining unit cost of hardware (CPUs and storage) per 
MIPS is the greatest contributor to overall lower costs. Most data 
centers are reducing their unit cost of software per MIPS at a rate of 
12 percent to 15 percent annually...

- Unix costs are decreasing by 18 percent annually, and are declining by 
30 percent a year if a measure of processing speed is applied. This is a 
direct reflection of continuing hardware performance improvements at 
similar price points. Measured by logical server, costs are declining by 
only 15 percent per year. Breaking down these trends, personnel costs 
are declining by 20 percent, directly reflecting productivity 
improvements in support services.

- Costs per utilized GB are declining at an annual rate of 40% per 
utilized GB - which is a lower rate than costs per installed GB. This is 
due to increasing redundancy being built into the storage farm and 
(partially as a result) relatively low utilization rates.

All of these trends can be leveraged through the implementation of 
technologies like server virtualization and storage consolidation which 
could drive these costs down even further (up to an additional 25 - 90% 
by some measures).

And despite this, Verisign and the ICANN staff are trying to sell a 
price increase while at the same time eliminating whatever chance we had 
left from seeing competitive behavior from the registry operator.

There is no way that the market for .com, or any gTLD for that matter, 
can be expected to behave in a normal competitive fashion. We need to 
move beyond this expectation and start managing these resources in a way 
that balances the need for stability with the requirement for innovation 
and competition. This doesn't mean reassignment or rebid every four 
years, it means increasing the number of registry operators, ensuring 
that ICANN can cancel the contracts or put them up for rebid in the 
event of a worst-case outcome and it means maintaining the involvement 
of the consensus policy development process in the management of the 
registry. This means preserving the precious checks and balances that 
are ratified in the current structure and not abandoning them because of 
litigation pressures from a self-interested commercial provider.

The registries are ultimately suppliers to the community, which implies 
that we are customers of ICANN. A little more thought about these 
accountabilities would go a long way towards rationalizing some of the 
bizarre an d harmful clauses that are being written into this proposed 
settlement

...I wonder if any of this will show up in the FAQ?

-ross



http://www.networkworld.com/supp/2004/ndc5/082304costs.html
http://whitepapers.itcinstitute.com/detail/RES/1058187154_291.html&src=TRM_TOPN
http://www2.cio.com/analyst/report3908.html
http://www.neweconomyindex.org/section1_page12.html
http://www.findarticles.com/p/articles/mi_m0SMG/is_4_20/ai_65951791