[arin-ppml] Fwd: ARIN-prop-165 Eliminate Needs-Based Justification

[Tom Vest]

On Feb 24, 2012, at 1:02 AM, George Bonser wrote:

>> On Behalf Of Tom Vest
>> Subject: Re: [arin-ppml] Fwd: ARIN-prop-165 Eliminate Needs-Based
>> Justification
>> 
>> Consider how other de facto non-duplicable "bottleneck" inputs that are
>> critical to entering some market tend to be priced (absent regulatory
>> intervention) when the only source(s) for those inputs are incumbent
>> market participants.
>> 
>> There is a straightforward explanation for this behavior, but it seems
>> to get very little attention, straightforward or otherwise, in
>> conventional economics.
> 
> I don't believe it is within ARIN's mandate to "create markets" of any sort.

I couldn't agree more, but I'm not sure how this has any bearing on the question of what considerations might motivate certain operators, under certain conditions, to exploit the increasing scarcity of globally interoperable (IPv4) addresses to strengthen their competitive positions, aka to exercise market power.  

> And this would be particularly so out of creating "real estate" out of the IP address space at the precise point in time where things are going to already be bad enough in that space.  Maybe after v4 becomes operationally obsolete, no big deal, but I still don't see how this gives all players in North America any operational advantage.

If you don't see how this would be advantageous to all market players in North America, that only means that you're paying attention ;-)

> The only advantage I could see is that some legacy holder might transfer some unused resources to someone with deep pockets but someone with deep pockets is likely to be the most capable of migrating to v6 and might be attempting to obtain v4 addresses for no other reason than to create a barrier of entry to competition.

Your hypothetical case is a classic example of the "allocative efficiency" argument -- which is not really an empirical argument/claim so much as it is a convenient axiom, sort of like the "transience" of private monopolies. As in the case of private monopolies, if one accepts that the highest bidder always yields the greatest allocative efficiency,  then "creating a barrier to competition" just might be one of the manifestations of that "efficiency."

>  There might be other ways of accomplishing the voluntary relinquishing of unused legacy space.  One of those ways might be simple peer pressure.  Simply publishing the top ten holders of unused legacy space, for example, in a very public way might be enough to generate some movement over time.  The thing is that any market (or more accurately, any "killing" to be made by brokering scarce resources) will probably have a very short window of time.  I also don't believe it is in the community interest to do things that would foster the continued growth of IPv4.  


According to the Census Bureau, appx. 10% (13m) of all of the single-family dwellings in the US are currently unoccupied. Millions of those have been vacant for one or more years, and quite a few have never been occupied at all. Even in the face of a steadily growing national population and persistent and increasing unfulfilled demand (aka the homeless), the existence of a very liberal/open housing market doesn't prompt millions of frustrated residential developers and would-be landlords to rush all of those idle/unproductive (but nevertheless "efficiently allocated") properties to market; instead they tend to wait until they get the price that they want, and/or to adjust their expectations downward only *very* slowly and reluctantly, over a period of months-years. I decline to speculate what market boosters who predict that IPv4 markets would yield significantly different outcomes might be thinking, but the burden should be on them to make a persuasive and empirically (not just theoretically) grounded case as to why things would turn out differently in this case. 

TV