[arin-ppml] Revised -- Policy Proposal 2009-4: IPv4 Recovery Fund

[Leo Bicknell]

In a message written on Sat, Apr 11, 2009 at 01:57:07PM -0400, Martin Hannigan wrote:
>    My thought was more along the lines of why would "sellers" agree to an
>    ARIN sponsored program when they can instead contact their nearest
>    domain name broker and start negotiating at higher rates? For this to
>    succeed, ARIN would be forced to recover at those market rates, gray,
>    black, or green, and passing those costs on will be _more_ expensive
>    due to program costs and cost of capital even if reimbursed. If there
>    are no "sellers", there is no ARIN recovery program in the context of
>    the policy, FWIW.

ARIN would be paying market rates if and only if those who still
need IP space are willing to pay market rates.

I hesitate to call ARIN a "pass-through" in this scheme, but from
a financial perspective that is what is going on.  Those who need
IP space figure out how much they are willing to pay (via a bid to
ARIN), ARIN goes off and recovers the space at market rate not to
exceed the bids.

From that perspective there is no advantage, prompting some folks
to ask "why put a monkey in the middle".  There are several advantages
to having ARIN in the middle:

  - Fraud is harder.  You and I, and probably everyone on the ARIN PPML
    list and at the ARIN meeting know a thing or two about addressing.
    We're not going to be taking in by someone selling 10/8, or doing
    any number of other bad things.  However, there are a lot of folks
    who get space from ARIN who don't do it every day.  We know there
    are companies that get space once, and never come back.  These folks
    don't have experience.

    ARIN would never pay to get 10/8 back.  How many folks for which
    addressing is not their day to day job will fall for the sales
    pitch that since IP addresses are short 10/8 is now in use, and
    they can buy a chunk of it?

    These folks are going to show up at an ISP's door and demand their
    space be routed, after all they just paid 10's of thousands of
    dollars for it.  They will demand ARIN recognize the transfer.
    We'll all have to deal with the mess that's made.

  - Bad transactions affect only one party.  If you and I agree to
    transfer space, perhaps sign a letter of intent, and agree to
    show up on the same street corner where the neighborhood notary
    has agreed to be our witness and I forgot my checkbook what do
    you do?  I suspect you wait for me to run home and get.  Or,
    what if you show up and say you need 1 more week to be out of
    the space?  I suspect I wait, not wanting to try and find another
    block, agree on a price, etc.

    If ARIN is in the middle and one of us messes up that party and
    ARIN will wait on each other.  ARIN can either give the space to
    another bidder that didn't forget their checkbook, or can recover
    space from someone else who got out on time to keep the other
    transaction moving.

  - The market becomes transparent, which should lower costs.  When
    folks don't know the going rate for something they can be convinced to
    overpay.  In 2008-2, 2008-6, and 2009-1 prices can be kept secret.
    Transactions are allowed to be private.  Folks won't have good
    information, particularly early on, to know if they are paying a fair
    price or not.

    With everything running through ARIN all pricing information is
    made available.

  - Deaggregation is controlled by ARIN.

  - Aggregation is possible.

>    I agree that this is beneficial. But this akin to making ARIN a market
>    maker. This means that we are in agreement that there is a market(and
>    I'm not trying to twist your words).

There is a black market today, I agree.  I don't think any of these
policy proposals will change the size or scope of the black market.

If 2008-6, 2009-1, or this policy is implemented there /will be/ a
white market.

I feel the choice this policy offers is between an unregulated
market (2008-6, 2009-1) and a central market.  In the other policies
ARIN pushes the financial transaction outside of ARIN; this is good
for ARIN's liability but leaves that aspect of the transaction with
no regulation, no rules, no standards, and limited laws (basic
contract stuff).  ARIN has no regulatory powers, so we can't create
a regulated market in the classic sense (e.g. create the IP version
of the SEC).  Thus the only alternative is what is offered in this
policy, a centralized market with ARIN in the middle.  ARIN is now
a party to all the contracts and can enforce rules via contract
with both parties.  It's not regulation, but it is as close as we
can come.

And thus, we have the core question this policy asks.  Is it better
to have ARIN take on some additional risk to (in my opinion) greatly
reduce the risk to those who are going to trade IPv4 number resources
(ISP's and their customers); or is it better to get the absolute
lowest risk to ARIN by leaving the ISP and their customers to fend
for themselves?

I'm not going to try and paint a rosy picture.  I think either way
we are in for a mess over the next 3-5 years.  However, the reason
I offered this policy up is that I believe ARIN is well positioned
to take on that small amount of extra risk, and that in doing so
will dramatically reduce the risk to the rest of the entire industry.

-- 
       Leo Bicknell - bicknell at ufp.org - CCIE 3440
        PGP keys at http://www.ufp.org/~bicknell/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20090411/37cbf7bb/attachment.sig>