[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [wg-c] Switching costs: a proposal
> Behalf Of Kent Crispin: Monday, August 30, 1999 7:42 AM
> On Mon, Aug 30, 1999 at 02:33:11AM -0700, Roeland M.J. Meyer wrote:
> > someplace to switch to. The only relation cost has to do
> > with it is if
> > the cost of switching becomes greater than some revenue fraction.
> Uh. This is precisely the point.
> > This
> > indicates the degree of "pain" felt as a result of the
> switch. Typical
> > costs that are less than 1% of revenue are not felt much, when they
> > become greater than 10% they are noticed and anything above 25% of
> > revenue generally can't get approved without going to the
> > The fact that 1% may be 10,000 times the price of a new domain is
> > irrelevant.
> You know, if you only wrote a few lines more you would have completely
> made my case for me -- you were doing an excellent job. Just let me
> finish it for you:
> You are precisely correct -- the point is not switching costs per se;
> the point is that domain names are so cheap relative to their
> switching costs. This is the opportunity they create for monopoly
> profits. I tried to express this clumsily by pointing to the ratio
> between switching costs and price; you have fleshed out a much more
> detailed explanation. Thank you.
I hate to say this, but this would indicate that the item-cost was very
much below the marketable price. It does not indicate monopoly
conditions. What indicates monopoly conditions is the fact that there is
no place to move to, at any price, short of new construction
(self-registry). This moves the business from the rental market to the
> As you point out, if your rent is doubled, you will go to the BOD
> because it is a significant expense. In fact, if your rent was
> doubled, you might very well *have* to move, because you simply
> couldn't afford to stay. If your rent were raised by 20% you might
> start considering a move.
I really hate carrying the analogy further but, if the rent doubled and
I was already paying much below competitive market rates, I wouldn't
necessarily move. However, not having a place to move to, I would be
strongly motivated to build my own so that I could control the price
directly and not have to deal with such capricious land-lords.But, even
that depends on the cost of build vs. rent.
To cover the cost of the build I would have to over-build in order to
become a land-lord myself. This has its own plusses and minuses. Initial
build-out, for a new registry, is quite expensive, however, and as we've
covered quite adequately here, once this becomes a sunk-cost additional
TLDs add very little incremental cost. Ergo, I can easily over-build.
The biggest cost-factor is the database maintenance and RDBMS licenses
(Gawd, Oracle is getting expensive).
> But if someone doubles the price of your domain name, you will
> simply pay it. A domain registry can double or quadruple its
> income by exploiting switching costs; a landlord can't.
Oh yes they can. Retail business leases can and do go up. They usualy go
up by an amount just under 40% of a typical move cost. One friend of
mine, who owns substantial retail space in Colorado Springs, asseses the
cost of a tennets move, deducts 20%, amortizes it out over the life of
the lease, and adds it to the existing lease cost as the new lease
amount. When they come back to negotiate, he shows them the analysis.
Usually, they renew the lease unless they need more space than he has
available. He typically does 5-year retail and office leases. He has a
reputation of being extremely fair.