[arin-ppml] IPv4 Depletion as an ARIN policy concern

[Eliot Lear]

Milton,

> A priori, the concept of "too much" is literally meaningless. If the market is functioning (i.e. sufficient levels of liquidity, appropriate ways to bring together buyers, sellers, etc.) the price passively reflects what a willing buyer is paying a willing seller. It could be $1 million per address and if transactions occur it means that someone thinks an address is worth that much and they will profit from having it. It also means that ARIN and the rest of society need to take seriously the objective reality of what is happening and ask themselves why an ip address has become so valuable. Covering that up may do more harm than the high price.
>
> A more serious measure of an "ineffective" market is one where no transactions take place at all. Or there are legitimate reasons for intervention when and if a price spike is part of a panic or some kind of temporary emergency shortage, etc. But in neither case could one specify a specific number.
>
> A decision that, say $10 (or $100, or $1000) is "too much" is simply to impose a form of non-price rationing. So if the actual market price is indeed $1000/address and ARIN or someone decides that anything over $10 is "too much" then you get all the well-known and well-document problems with price controls. i.e., shortages, non-transparent or political forms of rationing as a substitute, back and gray markets, etc.
>
> But even then to intervene properly one would need a very clear _economic_ understanding of what is driving the demand for and supply of addresses and what the impact of an intervention would be. It would be primarily an economic policy issue, not a pure engineering issue.
>
> Just be aware of the complexity and interdependence with which one is dealing.
> _
>    

Your message raises as many questions as your previous.  Here are a few:

1.  What is a reasonable transaction volume to expect, absent any 
regulation?
2.  How will we distinguish that volume from the "initial" adjustment to 
market equilibrium?
3.  What policy implications are there for new entrants to the SP 
market, and for enterprises?

This may sound like  we're covering old ground (we are), but as you 
point out, while perhaps these are simple questions, I doubt there are 
simple answers.

Eliot