ARIN IP6 policy for those with legacy IP4 Space

John Curran jcurran at
Fri Apr 9 07:13:48 CDT 2010

On Apr 8, 2010, at 4:35 PM, Joe Greco wrote:
> The problem, as I've heard it, is that ARIN's fees are steep in order to
> pay for various costs.  Since there isn't the economy of scale of hundreds
> of millions of domain names, and instead you just have ... what?  Probably 
> less than a hundred thousand objects that are revenue-generating?  If you
> charge $1/yr for each registered object, that means your organizational 
> budget is sufficient for one full time person, maybe two.  At $100/yr, you
> have enough funding for some office space, some gear, and a small staff.

Joe - Your financial breakdown is heading the right direction, but let
help out with some more information (FYI - ARIN's 2009 Budget is available 
at, and the 2010 one
should be there sometime next week.)

ARIN runs about a $15M annual operating expense.  As you noted below, it
can be hard to separate into distinct "products', and in fact, in some 
cases it is not appropriate to separate since one function (e.g. support 
for public policy development) might actually be a prerequisite for another 
( address allocations).  I am actually working to get more service-
oriented cost information going forward, but this is non-trivial to make 

In terms of fees, we have about 3500 ISPs (whose registration subscription
service fees cover the bulk of ARIN's expenses, i.e. an average of several 
thousand dollars per ISP per year) In other fees, we have over 1000 end-user 
organization and presently about 800 legacy RSA holders which pay $100/year 
for maintenance. This doesn't really cover much expense, and that is quite
appropriate since handling registration services requests (and the supporting
public policy process) does dominant the expenses of ARIN, at least today.

The question is how that evolves over time, particularly if the level of 
registration services requests in an post-IPv6 world is very modest.  At 
that point, ARIN's expenses will be predominantly registry systems support, 
and whatever public policy process the community wishes us to maintain.  
These costs will need to be predominantly covered by the maintenance fees, 
and will support the objects in the database, which includes the resource 
records of 3500 ISPs, 1000+ enduser organizations, the signed LRSA holders, 
and estimated 15000 legacy resource holders who have not signed an LRSA...  
At the end of the day, the Board of Trustees will determine the best fee
schedule to provide for cost-recovery of whatever functions are needed for
the mission at that time.

> So when you run into expensive stuff, like litigation, the best course of
> action is to avoid it unless you absolutely can't.


> Further, if you've suffered mission creep and are funding other things
> such as IPv6 educational outreach, that's going to run up your costs as
> well.

Presently, IPv6 outreach is not considered "mission creep", as it has
been an overwhelming request of the community both online and in the
public policy meetings.

> An established entity like ARIN typically has a very rough time going on
> any sort of diet.  Further, companies typically do not segregate their
> "products" well:  if IPv4 policy enforcement runs into legal wrangling
> and lawsuits, ARIN as a whole gets sued, and it is tempting to spread
> the resulting expenses over all their products.  Segregation into two
> (or more!) entities is a trivial way to fix that, though it also brings
> about other challenges.

Absolutely correct.  I think it is possible to understand those costs
better, but in some cases they can't be put into separate organizations
without some changes to structural assumptions about ARIN's mission.

> I have my doubts, based on a ~decade of observation.  I don't think ARIN
> is deliberately evil, but I think there are some bits that'd be hard to
> fix.

Joe - If you want to improve ARIN policy, jump right in.  If you want to
propose policy for the sake of changing the nature of the organization,
that's also fine, if you contact me I'll assist in providing estimates of 
cost savings and structural changes that can result from your proposals.
At the end of the day, it will be the community's discussion of your 
proposal, and the AC & Boards consideration of the discussion which will
decide the matter.


John Curran
President and CEO 

More information about the NANOG mailing list