Value Through Scarcity
Picture this: you stride into a bank, slap a sticky note that says “10 Downing St.” on the manager’s desk, and say, “I’d like a loan—my collateral is pure postcode prestige.” Absurd? Maybe, but 10 Downing Street, London, UK, is the residence of British Prime Ministers and has a storied history, as it has been around since the 16th century. If the address without any property were put up for sale, someone might pay good money to be able to say they live at 10 Downing Street, a bit like buying the rights to be called Warren Buffett from Warren Buffett. It’s a bit tacky, but a story for trivia nights. What makes this address and others of this type valuable is their scarcity, as not many addresses are special. However, addresses as such are plentiful and thus not monetarily valuable, as more can always be added. John Smith would probably not fetch much cash (unless, of course, there’s a collectors’ market for aggressively average names).
Speaking of addresses now, all devices connected to the Internet, roughly speaking, have an address quite similar to a house address. Whether it’s a smartphone, a website, or a smart device, once it connects to the Internet it has an address. Just, instead of cozy digits like '123', they're more like 142.250.191.46—Google's address at the time of writing. These addresses consist of four numbers in the range from 0 to 255, which are separated by dots. Now, you might be thinking, "But I just type google.com and get there, right?" True, but Google.com is essentially an alias, a user-friendly nickname for that string of numbers. Because forcing billions of people to memorize four random numbers is the fastest way to make the Internet the world’s largest Sudoku puzzle—and watch everyone rage-quit. But the actual address used to reach google.com is 142.250.191.46.
How about getting a loan by giving as collateral a block of Internet addresses? In 2024, the Internet Service Provider Cogent announced that they were looking to borrow money, which is all well and good, but they also announced that the loan would be backed by a block of Internet addresses. They are, in a very real sense, pledging a list of numbers to back a loan. It turns out that Internet addresses are scarce in general, and when there is scarcity, value can be extracted from this scarcity. The Internet address blocks not only have value but also can be owned. This whole situation feels weird. After all, they're just numbers. If you're running out of numbers, can't you just... make them bigger? Why should there be any scarcity at all?
How Can Numbers be Scarce?
The limitation on Internet addresses is kind of like the stubbornness of postal addresses—you can't just scribble whatever you want and expect your mail to arrive. Back in the early '80s, someone thought four numbers between 0 and 255 were plenty, giving us about 4 billion addresses. The same optimism that gave us keytar and shoulder pads, so you know, it was solid. Fast forward to 2025, and we've got over 7 billion mobile phones, not to mention all the other gadgets demanding their own slice of the address pie. There have been technological tricks employed to keep getting by with only 4 billion addresses, but clearly, that is too few. This has been recognized for a long time, so in the mid-90s a new proposal for addressing format was made to allow for a much, much, much larger range of addresses. This proposal is known as “Internet Protocol Version 6” or in short, IPv6, while the currently prevailing address standard is known as “Internet Protocol Version 4” or in short, IPv4.
If a solution has been known for almost 30 years, why are we still here? Well, it’s one of those classic coordination problems. Imagine the postal system in your hometown—let’s say San Francisco—wants to adopt a flashy new address formatting system. It’s clearly better, and in some ideal universe, they’d just switch. But here’s the catch: until every other city also jumps on board, San Francisco’s postal service has to support both the old and new formats. That means extra work, extra cost, extra headaches, and for now, zero real benefit—unless everyone else joins in. So, naturally, San Francisco waits. But every other city is also waiting, peering over their neighbour’s fence to see if anyone will blink first. The rational move is to keep waiting, which means everyone keeps using the old system. And that, basically, is why the Internet’s addressing still hasn’t switched to IPv6: everyone agrees it would be great, but everyone wants to be the last to do so.
Where There Is Value, There Is Owning and Trading
So, the Internet has been stuck on IPv4 addresses, which are scarce. And as always, where there’s scarcity, markets appear. In 2011 as reported by BBC, Microsoft spent $7.5 million to acquire 666,624 IPv4 addresses when assets of a bankrupt telecoms firm, Nortel, were put up for sale, in effect spending $11.25 per address. While picking up address blocks at “everything must go” auctions has its merits, a more regular market is even better. There are institutions with a hoard of addresses, and others scraping by; if they could find each other, they could do a win-win trade. To facilitate this trade, specialized brokers have sprung up where IPv4 address blocks can be put up for sale, and buyers can bid for them, quite like eBay. At the time of writing, the addresses are reported to go for one of the brokers for about $20 to $30 per address.
Now, Microsoft bought those address blocks from Nortel, but where did Nortel get them? And, really, what does it mean to “own” a collection of Internet addresses, anyway? Ownership of Internet addresses is quite similar to ownership of a home—there is a list somewhere saying that you own the home. For things like homes, the list is kept by the government, and if there is disagreement between people about the ownership of the home, the government list is the truth. Well, it is the truth in the sense that when all is said and done, the government will side with the person who is recorded as the owner on the list. But with Internet addresses, it’s not the government keeping score. Instead, it is a non-profit called Internet Assigned Numbers Authority (IANA), based in Los Angeles, with each world region having its own non-profit under IANA known as Regional Internet Registry (RIR). There are five RIRs covering the world. For example, ARIN, roughly speaking, covers North America; AFRINIC covers Africa, etc. Basically, five neighborhood watch groups for IP addresses, swapping address lists instead of gossip. The owner of the IPv4 address block is, in effect, the institution as specified on a list managed by a RIR.
Modern society has figured out that there is value to be made by creating and maintaining lists, and RIRs are the guardians of the IPv4 address block ownership list. When, as noted above, an address block is sold or purchased, what makes the trade in some sense real is the change in ownership in the ledger at the RIR. Another way to acquire an address block is to request one from an RIR for a legitimate business need. As a non-profit, RIRs objective is not to maximize its revenue, so while applicants need to pay some money to RIR for the ownership of the block, the value of addresses is larger than the ownership fee. One can run a legitimate business and get an address block, but running a legitimate business is hard, so there might be temptations to skip that step. In the late 2010s, according to the U.S. Attorney's Office, District of South Carolina, a company “represented itself as providing hosting services and providing customers with technologies and services needed for a website or webpage to be viewed on the internet”, and, as part of this, acquired a block of IPv4 addresses from ARIN. This turned out not to be entirely true, and “made over $3 million selling rights to those IPv4 addresses on the secondary market before his fraudulent scheme was caught.” Some engineering folk felt 4 billion numbers for addressing should be plenty enough, which 40 years later in a chain of events, led to a chap going to prison for wire fraud.
Legacy Windfalls
Back in the early days of the Internet, which for our purposes meant “before anyone had heard of Yahoo,” you could just ask for gigantic blocks of IPv4 addresses and you would get one. Like 16 million address blocks, for free, no paperwork, if you sent a note to IANA and they said “sure, here you go” because at the time the Internet was small enough to fit on a corkboard. Reflecting their informal origins, ARIN reported in 2016 about 25,000 legacy network records, of which only about half had validated points of contact.
One of the legacy owners of addresses has been MIT, and, for example, in 2017, MIT announced plans to sell a block of 8 million addresses. Which is nice, that somehow the scarcity of numbers finished up funding some labs and university research. But this also might cause some unease: is there something odd, or maybe even unfair, about the fact that some early Internet pioneers scored a windfall for being around at the right time and place? That’s one way to see it. But you could also say networks only become valuable when people use them, so having incentives for early participation makes sense. MIT’s participation made the primordial Internet more useful, and now it’s reaping an unexpected and belated, though not unwelcome, research grant from the invisible hand. Somewhere, a grad student’s robot is now laser-cutting latte art in appreciation.
A Consensus Is Valuable
The notion of owning Internet addresses is really just a big consensus game—everyone sort of agrees it’s in their collective interest to let non-profit RIRs keep the official lists of who owns which address blocks, because the network is more valuable if it’s big and everyone plays along. Lots of activity has sprung from the charmingly naïve technical decision, way back when, that four billion numbers would definitely be plenty for the Internet, because how many computers could there possibly be, right? Fast-forward to today: those numbers are fetching about $20 to $30 each, and the market is now alive with the usual parade of human schemes and creative shenanigans, all built on the surprisingly robust foundation of “we all agree this list matters.”