[IP] a pricing model??

[David Farber <dave@farber.net>]

________________________________________
From: Gerry Faulhaber [gerry-faulhaber@mchsi.com]
Sent: Wednesday, June 18, 2008 9:48 PM
To: David P. Reed
Cc: David Farber
Subject: Re: [IP] Re:   a pricing model??

Trying to make it simple, David.  Pricing with operating costs, presumably
dependent upon volume, would use a price/Gb at every moment.  If the
operating cost depends neither on volume nor capacity, it is then a fixed
cost, and there are various methods for recouping such costs.  But my
purpose was not to write an exegesis of price theory; simply to sort out the
simplest stuff.

Well, most of economics deals with situations of imperfect competition (the
interesting case), and there is interesting connections with fully efficient
prices and monopoly prices; again, not my job to explain price theory, but
this is fully explored.  When gov't is in the game, things can change
significantly but only when the regulators are attempting to regulate price
structure, which is not the case here.

Not a problem modeling competition with wireless BB (that'll be the day!).
Shouldn't be much of a price theory challenge.  And of course uncertainty is
a way of life in markets.  Fortunately, agents can change their prices as
uncertainty is resolved; you don't have to choose prices that last forever.
Game theory and real options make this much more complicated than it really
needs to be.

I too would favor competition, and if I could build my own fiber from my
home to..well...somewhere and control it myself, I'd think that was
wonderful.  But here in semi-rural Delaware I see complaining loudly about
having to spend $1,000 to hook up to County sewer.  Frankly, I don't think
there's many folk like you and me who would hook up their fiber to
somewhere, especially far away at the CO.

Good to hear from you, David.

----- Original Message -----
From: "David P. Reed" <dpreed@reed.com>
To: <dave@farber.net>; "Gerry Faulhaber" <gerry-faulhaber@mchsi.com>
Cc: "ip" <ip@v2.listbox.com>
Sent: Wednesday, June 18, 2008 3:36 PM
Subject: Re: [IP] Re: a pricing model??


> Er, all the costs of the Internet are not mostly or all capacity costs,
> Gerry.   Some of the costs are, and even those are functions also of time
> of construction/deployment and Moore's Law.  Access network opex has very
> little to do with capacity, as one well understood example.  But there are
> many more, and my sense is that they far dominate any pure capacity costs.
>
> And of course, references to the concept of economic efficiency when the
> system is far from an economic equilibrium, as in the reality we actually
> face due to regulatory capture, monopoly misbehavior, etc., makes me
> wonder if you're just kidding.  :-)
>
> Finally, one needs to consider that there is enough uncertainty in the
> system that one has to include contingent value on the cost side.
> There's certainly a reasonable likelihood that wireless access networks
> could substitute/compete for wired systems in many applications.   The
> uncertainty around that substitution and the disruptions that would arise
> in pricing is substantial enough that I'd think an analysis based on "real
> options", game theory, and some significant level of uncertain
> externalities would be necessary to model the system in a believable way,
> if at all possible.
>
> Now if, for example, the end user owned and maintained his/her own raw
> fiber up to a peering point, then we'd move closer to a situation where
> economic efficiency might be a better *local* approximation (maybe).   I'd
> be happier in a liquid economic world where gov't-industry collusive
> behavior were replaced by the policing that comes from competition, as I
> am sure you would as well.   But in that architecture we'd not be dealing
> with congestion in the access network - only with congestion in the
> backbone.  And we wouldn't need net neutrality in the access network
> either - use your own fiber any d***ed way you like.
>
> In sum, I'm not convinced that a simplistic economic analysis is helpful
> here.
>
> David Farber wrote:
>>
>>> Whew!  Somebody finally got it right!  When all the costs are capacity
>>> costs
>>> (as in the Internet), then the economically most efficient pricing (in
>>> theory) is what Bohn calls "spot" pricing,
>>>
>>
>>




-------------------------------------------