Nick, I am detecting a contradiction in your argument: you say excessive costs associated with fees and registrations WILL stop people from hoarding under the current system, which underprices address blocks. But at the same time you are arguing that if we remove needs assessment, droves of "hoarders" will pay excessive and rising costs to buy up large quantities of number blocks they don't really need. That argument just doesn't make sense. The most reasonable way to get out of this contradiction is simply to admit that hoarding will have a very limited impact on the remnants of the IPv4 address space due to existing market conditions. The price of v4 blocks will rise as scarcity increases. People who pay that money have strong incentives to move the blocks to people who can use them to generate revenue. Possession of number blocks is widely distributed now, so no one can corner the market. The real hoarders are the legacy holders who are not releasing their addresses as widely as possible now. Why? Because needs assessment prevents them from transaction with many willing buyers. If you were really concerned about hoarding, you wouldn't be talking about some hypothetical future state, you'd be talking about the situation we are in right now. 30% of the v4 address space is hoarded. You can see a pretty good challenge to the hoary hoarding argument from Louis Sterchi here: http://blog.kalorama.com/archives/1239 Hoarding is an economic phenomenon, driven by economic incentives. If you want to argue it will take place, don't just assert it, provide a coherent economic analysis.
-----Original Message----- From: address-policy-wg-bounces@ripe.net [mailto:address-policy-wg- bounces@ripe.net] On Behalf Of Nick Hilliard Sent: Thursday, July 25, 2013 1:39 PM To: Tore Anderson Cc: David Conrad; address-policy-wg@ripe.net Subject: Re: [address-policy-wg] 2013-03 New Draft Document and Impact Analysis Published (No Need - Post-Depletion Reality Adjustment and Clean up)
On 25/07/2013 16:31, Tore Anderson wrote:
For what it's worth, the motivation behind 2013-03 has nothing to do with "the market" whatsoever.
perhaps it's not the intention, but the consequence of the proposal is full market deregulation.
What this Evil Greedy Speculator needs to do is to register a few thousand LIRs, since each new LIR is entitled a single /22. Today, each of those LIRs must demonstrate "need", but this is merely a formality - if you have an intention of assigning one (1) IPv4 address to some End User within 12 months, you qualify. I seriously doubt that this formality is what has prevented the Evil Greedy Speculators from having done this already.
the current ripe ncc pricing structure means that each IP address has a nominal capex value of €3.70 + third party costs. So the nominal residual value of the remaining .88 /8 is about €55m + a real headache when you realise that you also need to convince some company registration office that they should be ok about 15000 new company registrations if they wouldn't mind thankyouverymuch.
Microsoft paid $11 per integer in 2011 for ~660k addresses. So right now, opening up bazillions of LIRs doesn't look like a particularly good option.
[that said, maybe we should put forward a policy to ask RIPE to create a restful api for opening new LIRs, and to plump up Axel's medical insurance policy for when he needs to sign his name 15000 times. Just thinking out loud here...]
Nick