[wg-c] Switching costs: a proposal

[Milton Mueller <mueller@syr.edu>]

Ross makes some valid points about the inherent inflexibility of a "price cap"
regime. But all of us seem to be forgetting one of the most important points
against it:

No one is arguing *against* the creation of registrar-owned, cost-controlled,
non-profit registries. That alternative will be available. No one will be
forced to patronize a for-profit registry.

I would also like to address the "rent control" argument raised by Roeland.
A local Korean grocery store in our area just changed its location. It occurred
to me that the "switching costs" faced by businesses in regard to real estate
rental are equal to if not greater than those in domain names. The rental
property proprietor has about as much "lock-in" power as any domain name
registry, if not more. Think of the costs of moving merchandise, temporarily
closing the business, advertising the change of address and phone number,
altering business cards, notifying suppliers, etc., etc. Yet local businesses
do change locations, and there is no clamour for transforming all business
property into rate-controlled, non-profit ownership.

Now for the proposal:

In the domain name arena, the problem of switching costs is better addressed by
contractual obligations that establish a code of conduct for how transfers of
clients from one registry to another might be handled.

E.g., (just off the top of my head) there might be a specification that once a
customer notified a registry of its intention to switch, the registry is
obligated to maintain its existing rate for at least one year, and perform
certain relay functions adequately.

Anyone care to work with me on this off-line?

Ross Wm. Rader wrote:

> Several scenarios immediately ran through my head would be "against the
> rules" in a price increased controlled situation...
>
> - new TLD comes online, registry offers free addresses for a fix period of
> time to increase adoption, followed by a price increase to a reasonable
> level sometime later....
>
> - registry finds that initial price determination was not realistic and a
> price increase is necessary to ensure minimal profitability on the
> registries behalf.
>
> - registry decides that his/her TLD is not a price sensitive commodity, but
> rather a value based service that commands a premium dollar
>
> - registry notices seasonal demand fluctuates and implements a market
> demand based price system that charges less during slow periods and more
> during busy periods
>
> - registry possesses an extremely unique TLD that holds very limited
> interested to the larger market, but an unusually high level of interest
> for a specific portion of the market and decides to implement an auction
> based process that allows the market to actually determine the price of
> sale...
>
> All of these fall outside of the typical pricing models that we have
> encountered thus far (x$ for y time period, plus renewals) and would like
> become foregone avenues of pursuit under your proposal...
>
> Obviously, this isn't a binary decision, but gven the option of fix-price
> vs. registry determined price, I would have to back a registry determined
> price and let the market sort out the dead models.
>
> -RWR