What is the most important factor influencing the price of IPV4 blocks? Nobody really knows the answer to this question, there is no single correct answer here. In fact, the average price that sellers have in mind could be considered to be the price. This is regardless of whether buyers are willing to pay this price. This is one way of looking at it. Another definition of price is looking at real transactions where seller and buyer have come to an agreement. This does not render the first definition invalid. If owners of rare classic vehicles want a certain multiple of millions of dollars for their cars and nobody is able to afford them this should still be considered the price. Therefore we can safely say that seller’s perception of price is more important than buyers. When trying to answer this question it is important to distinguish between historical prices and future prices. Personally I find future pricing more interesting to look at. Let’s first have a brief look at historical pricing as this teaches us a little on how to look at future pricing.
The first transaction recorded was between bankrupt Nortel and Microsoft at USD 11.25 per IP address in 2011. This transaction set the standard for many transactions after this point in time. Note that this was so called legacy IP space which is generally considered more valuable than RIR allocated IP’s. We know that these IP’s were brokered by Addrex who ended up with seven interested parties who submitted bids on all or some of the IP’s. Microsoft having submitted the largest bid walked away with the valuable prize. Assuming these bids were not auction style, one bidder did not know the other bidders bids. That means each bidder had to come up with whatever they perceived to be a feasible price. Microsoft valued these IP’s the highest and we can only assume this happened because Microsoft had future plans for the IP addresses. Microsoft must have already made plans to become a cloud provider, Azure, Office 365, and made sizeable investments into these new ventures. The 7.5 million dollars Microsoft invested in the address space was, in hindsight, probably a small percentage of the total that was invested in becoming a large cloud provider. Interestingly, in this scenario it was the buyer that ultimately decided on price. And price was largely created based on future plans and a larger investment plan.
Since 2011 prices for IP blocks have gradually been declining. This is strange as IP space is becoming scarcer every day. Since the ARIN runout or depletion this year there has been an increase in pricing again.
Let’s have a closer look at which forces are at play.
Mostly webhosting companies, cloud providers and similar companies are in need of IPV4 as they need to run parallel networks with IPV6 as IPV6 is not backward compatible.
Access providers, cable companies with end users that have deployed IPV6 can now sell their old IPV4 stock. This will obviously depress prices.
The Regional Internet Registries have adopted a policy whereby each member can only request for a last /22 (1024). ARIN members can’t even get any IP numbers in this way as ARIN has completely run out. This last /22 can only be transferred after 24 months.
Transferred address space can only be transferred or sold again after 24 months. This ‘locks’ the market and increases price. Each block sold will be off the market for some time.
There are millions of IP addresses unassigned. Most of these are said to be legacy space. Some say this influences price but it does nothing for a small host that needs a /24 at short notice. Will the unassigned space be unloaded on to the market anytime soon? I doubt it. With IP space becoming scarce anyone who might have a future use for it will hold on to it.
IPV6 deployment is slowly increasing but what does this mean for future IPV4 requirements? How is this measured? How quickly will deployment grow? It seems that hosting companies will need to continue running parallel networks for years to come.
In the RIPE region alone there are approximately 500.000 IP addresses offered through the listing service. But the same service advertises a whapping 27 million IP addresses requested. How can this not drive prices up?
There is a huge pool of reserved space. What will happen if this space is made available?
The recent ARIN depletion, which can be seen as a landmark in internet history, has caused a change in perception all over the world with regards to IPV4 scarcity. It has also caused a real change for fast growing US and Canadian companies with a shortage of IPV4.
Since September 2015 RIPE has allowed for inter RIR transfers. It seems however that this option has not in fact been used yet. There is a lot more paperwork and fees involved for anyone using this option. Does having this option influence price? It looks like it has, it gives people a false idea that there is an abundance of IP space on its way. Really?
What else influences IPV4 pricing?
The USD EURO exchange rate. When it comes to the internet, America has always been leading. Whatever you come up with to disprove this, is an exception to the rule. This is also true for IP space. The first transaction was in the US, the first broker was US, the first RIR to runout is ARIN. The internet was conceived in the US so no wonder this is the case. Therefore the US dollar is leading. A stronger dollar, ceteris paribus, makes IP’s quoted in EURO in the European region more expensive.
Block size. This is a very important determinant of price. Currently smaller blocks fetch a higher price per IP. It is said that this is due to the handling involved in transferring a block relative to its size. In other words selling a /16 comes with the same paperwork as selling a /23. Personally I don’t agree with this theory as selling a/16 comes with a lot more responsibility. If you’re selling a /16 you better have your ducks in a row. The demand for NDA’s, contracts and good escrow increases when block size increases. The real reason smaller blocks fetch a better price is because they are more in demand. There are a larger number of smaller players with small budgets competing to buy the same lots. Larger blocks are intrinsically more valuable as from a technical perspective they are easier to manage compared to a whole collection of smaller blocks. It is even said that all these fractionated small blocks in the routing tables will impair the efficiency of the internet.
There are a lot of brokers active on the IPV4 market and they don’t really care whether IP’s are sold at USD 8 or twelve. They just want to see as many IP’s changing hands as possible as this will increase their commission based revenue. They of course should care about price because 20 percent of 12 is more than 20 percent of 8 but in reality they seem to prefer a lower price as they believe this is good for total turnover. This has a real influence on price as they tend to talk sellers into lowering their price. These lower prices in turn creates perception in the buyer’s minds thus making it very real. Most brokers don’t deal in blocks smaller than a /20. Could this be another reason why small blocks are relatively more expensive?
I haven’t mentioned the quality of the IP’s yet as this only influences individual blocks and not price in general. Now it’s mentioned!
Now what can we do with all this information?
Everyone to whom price is a factor of importance needs to decide which factors weigh the most. Personally I believe budgeting and future plans should determine how much you are willing to pay for IP space and then buy as much as you can and need at or below that price. If that price is below anyone is willing to sell you their IP’s you need to readjust your plans! The second strongest indicator is the huge difference in the number of IP’s offered and requested. This is the most real indicator out there and speaks volumes about where this is going. If you are a seller you may want to think about how quickly you want to sell your IP’s, how much risk you are willing to take (IP’s can theoretically drop in price) and come up with a price accordingly.
If you believe, like I do, that IP’s will become more expensive in the near future. Buy as many as you think you’ll need for the next 10 years. Now! If you can afford it that is.