[arin-ppml] ARIN Draft Policy 2017-1: Clarify Slow Start for Transfers proposed updates

[Mike Burns]

Is there any interest in moving towards a slow-end of the needs based tests; where the minimum 12 months (in 8.3, 8.4, and 8.5.5) becomes 6, after a review period the 6 becomes 3 and later the 3 becomes none as the transfer market is monitored throughout the slow-end to ensure this isn't enabling the negative scenario(s) it's designed to prevent?

 

-Austin

 

Hi Austin,

I would support a phase out of the needs test, but I am not sure how you would monitor the transfer market to detect the enabling of negative scenarios.

I would be curious to know what those scenarios are and how they would be detected.

If the scenario is market manipulation/speculation, we have evidence of years of needs-free transfers in RIPE from which to draw conclusions.  And analysis of those transfers does not reveal speculation, instead it shows the most dynamic market of any registry.

But I like your idea to trend away from needs tests. I believe that the price of the addresses in the transfer market provides the conservation that the needs test was designed to provide in the free pool era.

Likewise the holding periods which were intended to prevent flipping of addresses.  I myself authored that waiting period for transfers in ARIN, but  I am not even sure why flipping is a problem at all.  APNIC has never had a holding period. We have many sellers whom we have to advise to hold on to their un-needed addresses until their holding period elapses.  This takes addresses out of inventory and prevents their timely use by those who express their real need for the addresses through payment of money.

I know that we would be an entity that would like to purchase address blocks and then resell them in smaller pieces. Looking at allocation history, many /16s were allocated, but looking at transfer history, most transfers are for smaller blocks.  Many /16 holders are not as competent or desirous as we are in processing many sales, and prefer just one transaction.

So we would like to buy a /16 in a single transaction (good for the seller), use our experience to ease the transactions for the many small block buyers out there (good for these buyers), and make money on the price differential between large and small blocks (good for us).  (Bad for routing table size, but that’s inevitable. Blocks, like Humpty Dumpty, can only be broken down.)

I believe that address stewards, fearful of market manipulation that could be harmful to the Internet, have adopted policies whose intents are benign, but whose effects are negative.  It’s time to acknowledge the information from APNIC and RIPE related to these market-stifling policies, and consider adjusting our policies to reflect that information.

I had an economics professor, Paul Samuelson, who had a famous quote which is relevant to stewards like us:

“When my information changes, I alter my conclusions. What do you do, sir?”

Regards,

Mike

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20170505/29074748/attachment.htm>