It's an obvious enough question. Each area code can hold, in theory, almost 8 million telephone numbers. And yet area codes split and then split again, in many cases into regions with populations well under 1 million people.
The reason most often given for the explosion of new area codes is the explosion in the use of new technologies. Every pager, cellular telephone, fax machine, modem, credit card verification box, and ATM machine has to have a telephone number. Also, where in the past a company would have a main telephone number that reached the receptionist, who would then connect the caller to the desired extension, now each telephone in the building has its own direct-dial number. The increasing popularity of home offices, as well as the perpetual clamor of teenagers to have their own phone lines, also contribute to the demand for telephone numbers.
Those factors all contribute to the exhaustion of the supply of numbers in an area code, but they're not the biggest culprit.
The telephone numbering plan in North America is designed around certain specific assumptions. The telephone number consists of three parts: the area code (3 digits), the prefix (3 digits), and the line number (4 digits). For example, (311) 000-9999 is area code 311, prefix 000, line number 9999. Each area code is divided geographically into rate centers, geographic areas that are billed as if they were a single location. Each rate center contains one or more wire centers, the switching equipment that serves a local area. Ordinarily, each prefix is assigned to one specific wire center. (There are only very limited exceptions to this rule.) That means that numbers are assigned to wire centers in blocks of 10,000.
Until the advent of local telephone competition, one company — perhaps one of the "Baby Bells" or perhaps an independent company such as GTE or Centel — had a monopoly on each rate center. In small towns with populations of only a few dozen, an entire block of 10,000 numbers would still be assigned, although most of the numbers were not needed. However, this inefficiency was entirely accounted for at the very beginning of the process: very few new small towns appeared on the map to gobble up additional prefixes. Where the sudden new demand, and the sudden increase in inefficient allocation, occurred, was in the cities.
When local telephone service opens to competition in a city, each new competing carrier must have numbers available in each rate center it wishes to serve. (Competing carriers generally honor the rate center boundaries, since they affect call pricing. However, they are not bound by wire center boundaries, since those are purely a technical feature of the ex-monopoly carrier's network.) Under the current system, that means it must be assigned an entire block of 10,000 numbers for each rate center in the territory it serves. Let's say that a particular metropolitan area has 30 rate centers, including different sections of the city and various suburbs. Now let's say that there are 15 companies that want to compete against the dominant carrier for local service. Each of those 15 companies requires a prefix in each of those 30 rate areas, for a total of 450 new prefixes. That's a total of 4.5 million out of the fewer than 8 million numbers possible per area code, EVEN IF the new carriers have only signed up a few thousand customers! (The example is not at all unrealistic, by the way.)
The reason is that the prefix designates the local switching equipment. When you place a call to 311-000-9999, the telephone network examines the area code and prefix in order to route you to the correct switch. The far-end switch then selects the correct line based on the line number. That means that each prefix must designate both the wire center and the carrier, since the carriers can't be forced to share a switch. (Even in cases where the "competing carrier" is only a middleman reselling the dominant carrier's service, a separate prefix is assigned to designate the customers of the competitor.)
Reading the paragraph above, you might think that the new competing carriers are to blame for the sudden demand for new area codes. That's not really a fair assessment, though. The problem isn't that these companies want to compete for your business, it is that the numbering structure wasn't designed to accommodate that competition. About 70% of the numbers in an average prefix served by a "Baby Bell" in an urban area, are in use at any given time. The other 30% are numbers that have recently been taken out of service, or unused numbers available as spare capacity. If you push the utilization rate much higher, you run the risk of spot shortages, because there is a long lead time from deciding to add a new prefix to having it in active service. The system worked pretty well for the environment for which it was designed, but rapidly shows the strain of a major shift in that environment.
In other words, no one is to blame for the problem. The current system was an efficient design for the requirements of its day, but it doesn't work as well now. We need to change the way that telephone numbers are allocated.
Local Number Portability (LNP) is a process that unlinks a prefix from a specific carrier. The prefix is still assigned to a rate center, but any carrier can assign numbers from that prefix within that rate center. This is accomplished by having each call to that prefix look up in a database which carrier and switch serves that number — the call is routed on the full 10-digit number instead of only the area code and prefix. If you're calling long distance across the country, the area code and prefix can still route your call to the general area, where the database will determine the final route to complete the call. Since the prefix is still tied to a rate center, customer billing software will not need to be modified — the area code and prefix will still determine the charge for the call.
Of course, there are both technical and political issues to work out in implementing LNP. On the technical level, the lookup must be done quickly enough to make the process transparent to the user. If it takes even an extra second or two to complete a local call, users will complain. The system must be robust enough to handle an enormous volume of concurrent lookups. On the political level, there is the question of who is responsible for administering and maintaining the database. If the Baby Bell administers the database, the competitors may worry that their updates may not be posted in a timely fashion. Whoever administers the database, there is the question of how to distribute the costs in an equitable manner.
Once LNP is up and running, though, the frenzied pace of area code exhaustion will slow dramatically. Growth in cellular usage, fax machines, and regular lines, will continue to consume numbers, but at a much more manageable pace, since the numbers will be allocated much more efficiently. Tiny Rural Town, U.S.A., will still have its block of 10,000 numbers, but Medium Sized City of 600,000 people won't need 16 (or even 24) million numbers.
Historically, wire center boundaries have been used in drawing area code boundaries, even if that meant splitting a rate center between two area codes. For instance, in the 1997 split of area code 415 in California, the "San Francisco 3" rate center (covering the southern part of San Francisco and portions of some adjacent suburbs) was split between 415 and 650. In the days before competition, this would have been a reasonable solution, keeping as much of the city as possible in 415, while moving as much of the suburban area as possible into the new area code.
Under present-day conditions, with local competition but with LNP in its infancy, a split that divides a rate center actually aggravates the problem of prefix exhaustion, because each competing carrier now needs a prefix in each area code of a rate center, in order to serve customers on both sides of the area code line. Even once LNP is fully implemented, it will still be a bad idea to split a rate center between area codes, because it will require effectively creating two rate centers.
The bottom line: all geographic area code splits must strictly follow rate center boundaries.
Return to top
Return to LincMad's Telephone Area Code page
©1998 Linc Madison, a.k.a. LincMad Consulting, all rights reserved. URL: <www.LincMad.com> Comments to contact at LincMad d0t c0m. (Please note: mail sent to "webmaster" will be deleted unread.)
Last updated 1998-03-23.
LincMad.com received over a million visits from 1997 until the counter broke.
Check out my political blog at LincMad.blogspot.com.