ICSA 855 – Telecommunications Policy and Standards
In the early 1980’s, when the Internet was just starting, the federal government decided that, in order to allow the new network to freely grow, they would exempt Internet Service Providers (ISP’s) from paying local exchange carriers (LEC’s) access charges for the interconnection that long-distance companies were and are still required to pay (1). However, now that Internet growth has exploded, and the number of users that dial into the network is very large and continuing to grow at a tremendous rate, the telephone companies are clamoring for a reexamination of this policy. They state that the telephone network may collapse if they are not allowed to charge for Internet connections, because without the additional revenue, they will not be able to expand their networks at the growth rate necessary.
This paper will first provide an overview of the issue, then explore the views of the local telephone companies, the Internet Service Providers (ISP’s), the consumers, and the federal government. Also, the views of the Internet Access Coalition, a group formed to keep Internet access costs low, will be discussed. Finally, possible soltuions to the problem, both short and long term, will be discussed.
Synopsis of the Problem
“Under guidelines established in the early 1980’s, Internet Service Providers are exempted from paying the additional fees that long-distance companies pay local providers for calls that are transferred to another network (1).” These regulations were designed when the Internet was young and needed as much assistance as possible in order to enable its easy growth. Local phone companies argue that these regulations should no longer apply now that the Internet has grown and is being utilized in many commercial ways. The FCC is expected to review these regulations by mid 1997 (1), but any changes may significantly alter the current economics of the Internet, where users can now connect for the price of a local call and a low monthly access fee.
Changes in Local Phone Call Dynamics caused by the Internet
The phone network was designed with certain assumptions in mind, such as the fact that a typical phone call lasts just a few minutes, and that the majority of calls would occur during certain time frames (4). However, the growth in Internet phone calls has changed the usage of the phone network dramatically. “The root of the telephone network capacity problem lies in the fact that Internet calls are far longer than voice calls that the Network was designed to handle. Pacific Telesis found that an average Internet call was [approximately] 20 minutes long, compared with four minutes for an average phone call. Ten percent of Internet calls were six hours or longer (1).” Also, “some people stay connected to the Net 24 hours a day via a 2nd phone line (6).”
And finally, as telecommuting is promoted by the federal government for various environmental reasons, and by business for cost savings reason (less office space to lease, etc.), the number of calls needed to keep employees connected to their company’s networks or o the Internet for the entire business day is increasing dramatically. All of these calls require the local telephone company’s network resources.
Some Internet users are now using the network to make “free” long distance calls (5). They dial into their local ISP, and then use software that digitizes their voice, and sends it to another Internet user. Although the quality is not as good as a normal telephone call, it is improving, and more importantly, it is free. As more people begin to use the Internet for long distance phone calls, the dynamics of the phone network will be altered even more. This will only mean that the length of the ‘local’ call (to the ISP), gets longer and longer.
Besides the fact that the length of the average phone call has changed so dramatically, the peak hours for phone system usage has changed to the 7-11 p.m. time period, putting additional burdens on networks originally designed around weekday calling peaks (1).
All of these factors are contributing to the fact that local phone call durations are dramtically increasing and that normal usage patterns are changing. These issues mean that additional equipment must be placed in the phone switching offices, or some local phone calls may not be completed at peak hours.
The Local Exchange Carriers want to charge Internet Service Providers access charges on a per minute per call basis, so that the additional revenue can be used to expand the local network to ensure that missed calls are kept to a minimum. Opponents of the LEC’s claim that the LEC’s studies are flawed, and that they are in reality claiming there is a problem when there really is none. At this point, the Federal Communications Commision (FCC) is reviewing the issue, and asking for input from all involved parties. The rest of this paper will examine each party’s view, and then offer possible solutions to the issue.
The Local Exchange Carrier’s Side
The Regional Bell Operating Companies (RBOC’s) and Local Exchange Carriers (LEC’s) claim that the nation’s telephone network is on the verge of collapsing due to the growth of the Internet and the local phone calls to ISP’s necessary to connect to it(5). Several studies by various BOC’s and LEC’s have been completed that attempt to prove this point.
For instance, one Pacific Telesis study found that “16 percent of call attempts failed during peak evening hours because of Internet traffic (5),” whereas past statistics show that less than 1% of phone calls [did] not connect (6).
Although NYNEX has not experienced failed calls as PacTel has, a study they completed shows that about 200 lines in one service area are used for Internet calls, and that this number grows by approximately 10% per month (7). Because of this growth, if they do not upgrade their network to handle the additional capacity, they too will soon experience excessive failed calls.
The two studies by PacTel and NYNEX are representitive of all the RBOC’s studies. In order to solve the current problems and/or potential future problems, the RBOC’s feel that they must upgrade their networks to handle more telephone calls. However, they feel that they are not making enough money on Internet calls to cover the costs of adding additional equipment (4). PacTel estimates it will cost between $25 million and $100 million in 1997 alone to avoid the ‘brownouts’ that have occurred (1).
In order for the RBOC’s to expand their local telephone networks, they “favor a fee of about one-third penny per minute (3) charged to the ISP’s.” Therefore, a 60 minute call to an ISP would cost the ISP approximately 20 cents. The RBOC’s contend that this is a small amount so the ISP’s should be only minutely affected, yet that it is enough to fund the network upgrades needed to handle the Internet calls.
Another aspect of this apparent problem is that many Internet users purchased second telephone lines dedicated solely to Internet calls. Opponents of the access charges claim that these second phone lines bring in additional revenue to the local telephone companies, and that this revenue should be used to upgrade the network instead of access charges. However, Bell Atlantic states that not all second phone lines are used for Internet Traffic (4), and this view is shared by other local providers.
The Internet Access Coalition
The Internet Access Coalition is “a group dedicated to maintaining the affordability of Internet access over telephone lines and accelerating the availability of inexpensive digital telephone network connections to the Internet (9).” The IAC is composed of several companies and groups, some of which include: America Online Incorporated, Apple Computer, Inc., Compaq Computer Corporation, CompuServe Incorporated, Dell Computer Corporation, Eastman Kodak Company, GE Information Services, IBM Corporation, Intel Corporation, Internet Service Providers and Users Assn., Microsoft Corporation, Netscape Communications Corporation, Novell, Inc., Oracle Corporation, Software Publishers Association, Sun Microsystems, Inc., Voice on the Net (VON) Coalition.
There are other groups and companies that are also a part of the IAC. The reason this paper includes such a long list is so that the reader can see that all of the major Internet, computer, and consumer groups are involved. Though the ISP’s and consumer’s views will be expressed below in their own sections, it is important to see that they are binding together to try to influence the FCC in the fight against local exchange carriers’ wishes to charge access fees.
The IAC sponsored a study (7) that will be used often below in both the Internet Service Provider’s Side and in The Consumer’s Side sections of this paper. In addition, the IAC found several flaws with the RBOC’s studies and offers other points of interest as outlined here:
First, the IAC study claims that “the BOC and Bellcore studies present a distorted picture of the actual impact of data communications traffic by limiting their examinations of the ‘problem’ solely to certain central offices and switching entities that happen to serve ISP’s (7).” In fact, the IAC study claims that “the four BOC studies rely upon isolated, worst-case situations.” For example the PacTel study was done in Silicon Valley, which has perhaps the nations highest percentage of Internet dial users.
Second, the IAC study (7) contends that “there is no way to ensure that revenues generated from a hypothetical access charge would be used to invest in a network that can accommodate data traffic,” as the RBOC’s claim they would do.
And finally, the IAC claims that “the growth of ISP’s has created significant revenue streams for the BOC’s directly attributable to data traffic on the PSTN (7).” This is in the form of second telephone lines that many Internet users purchase as their time connected to the Internet increases and they find one phone line is not enough.
The IAC is using their study, along with their Web pages and promotional efforts, to bring attention to the issue of Internet access charges. From the information they provide, it is clear they are heavily against the LEC’s wishes to charge such access fees.
The Internet Service Provider’s Side
The current Internet Service Provider market dictates that, in order to be competitive, a basic plan for unlimited Internet access must be in the $20 range. Anything more, and end users will simply go to a provider that does charge in that price range. Jill Lesser, deputy director of law and public policy for America Online, states that “at the $19.95 rate, we’re operating at … small margins (4).” This is basically predatory pricing, as the ISP’s are trying to grab market share now while the industry is still young and growing at such tremendous rates. The battle is between many small ISP’s, and more recently large ISP’s such as IBM, AT&T, Sprynet, America Online, etc. This battle has resulted in price cuts, so that profit margins are indeed minimal. As the number of providers reduces (i.e., some will go bankrupt, some will merge with others, etc.), as it must since the profit margins are so low, the ISP’s that will eventually be able to charge more. Also, as Internet user’s bandwidth requirements become ever more intensive, the ISP’s must upgrade their networks, which will also raise the cost of Internet access.
At any rate, ISP’s across the board are urging the FCC to not allow local exchange carriers to charge access fees for Internet calls. The local exchange carriers desire to charge ISP’s one-third penny per minute for all Internet calls means an additional 20 cents per hour per call to each ISP. Since ISP’s are already operating at such small profit margins, the charges would certainly be passed along to the consumer. This increase could be enough to potentially stifle future growth of the Internet (3). The Internet Access Coalition also has the same view, and states that “allowing the Baby Bells to charge the access fees would ‘impede the continued growth and success of the Internet’ (7).”
In addition, ISP’s say local carriers have no right to the extra revenues that they would get by charging per minute fees to ISP’s, because growth in Internet access for dial users means many consumers are installing second phone lines in their homes. The IAC study states that “… nationally, from 1990 through 1995, the local exchange carriers have collected more than $3.5 billion in revenues from additional residential access lines for subscribers who use them solely or primarily for calling ISP’s (7).” Because of this increase in revenue just from second phone lines, ISP’s feel that the LEC’s should not be able to ‘double bill’ by also charging access fees to the ISP’s.
ISP’s contend that the solution should be to implement new technologies that would offer users faster access — which they would then pay for — rather than simply expanding the traditional voice system and asking them to pay more for the same service (1).” The ISP’s feel that the LEC’s should be installing devices that remove the Internet calls from the voice network, so that the potential bottlenecks of data calls would be eliminated. Also, since the voice switches of the PSTN are the limiting factor in potential data transmission speeds (due to their 3000 Hz bandwidth range), having data calls bypass the voice networks would mean faster data connections.
ISP’s also argue, “Why are the RBOC’s aggressively selling residential second phone lines which are often used for computer lines (4),” and why are the RBOC’s now selling Internet access themselves if it is hurting the PSTN (7) (i.e., hurting their own network)? The ISP’s feel that if there were truly a problem that the LEC’s would not be contributing to it as they appear to be.
Clearly, the ISP’s have several valid points, and if the LEC’s are allowed to charge access fees to the ISP’s, the ISP’s will have to pass the costs on to the end users of the Internet. According to the ISP’s and the IAC, the LEC’s should be spending the additional revenue generated from second phone lines dedicated to Internet access on installing new technologies to bypass the problems, instead of billing both the ISP’s for interconnections and the consumers for local carrier services.
The Consumer’s Side
Most consumers feel that their monthly subscription rate for local telephone service should cover all local telephone calls, which in their minds includes calls to ISP’s for Internet connections. Since the monthly local access charge has covered all local telephone calls in the past, why should a local call to an ISP be any different?
The FCC has requested comments on this issue from all interested parties, and they have received over 300,000 e-mail’s from Internet users (2). The sheer number of negative comments from consumers shows they have strong feelings against access charges.
The Internet Access Coalition created a Web Page (8) which outlines what consumers can do to prevent the FCC from allowing local carriers to charge access fees for Internet calls, with the following four points:
Consumers should express their support for the FCC’s tentative decision to not allow local telephone companies to assess new charges on Internet access providers.
Consumers should express their views to their state’s senators and their district’s representative.
Consumers should consider joining or supporting any of the consumer groups that favor the affordability of Internet access over telephone lines and the availability of inexpensive digital telephone network connections to the Internet. Such groups include “The Center for Democracy and Technology,” “Internet Service Providers and Users Association,” and “The Media Access Project.”
Consumers should tell other Internet users about this issue, as the more who hear about the local phone companies’ proposal to assess new access charges and the FCC’s proceeding on Internet access, the better chance consumers will have to prevent Internet access rate hikes.
Cleary, the IAC is strongly encouraging consumers to fight the LEC’s in their attempt to charge access fees. The fact that there are several consumers groups in the IAC, and the fact that consumers are reading the IAC’s views and acting on them (as seen by amount of email and discussions generated by the above points), shows that consumers do not want to pay additional charges for what they feel is normal local carrier service.
The Federal Communications Commision’s Side
The Federal Communications Commission (FCC), at this point, has not made a final decision on this issue of Internet access charges. On December 25, 1996, they did, however, make a temporary ruling. This ruling, called the combined Notice of Proposed Rulemaking (NPRM) (11), temporarily denied the request of local phone companies to impose access charges on the Internet and other enhanced service providers (2).
In the NPRM, the FCC states that such access charges could have “potentially detrimental effects on the growth of the still-evolving information services industry.” Clearly, the federal government does not want to stifle the growth of the Internet. As president Bill Clinton and vice president Al Gore are pushing for the construction of an Information Super Highway, they will do nothing that might in any way inhibit its growth, which the FCC feels includes access charges.
Despite this resistance, the FCC has not issued a final decision, as it is requesting comments from all parties that may be affected by such access charges, which includes local exchange carriers, Internet Service Providers, computer networking manufacturers (both hardware and software), and consumers. They will review all input and make a final decision on the issue. Although the final date has not been set, it is expected to be made by mid 1997.
In many ways, the ‘problems’ the local telephone companies are complaining about are short term issues. As the telephone and data networks mature, the bottlenecks on the voice network caused by data calls should be eliminated. New technologies and new network paradigms will be developed that will overcome the problem. These long term solutions will be discussed below. However, to get to that point, the LEC’s claim that they can not institute new technologies without additional income. This section of the paper will first address the short term issues, and then a few potential long term solutions.
Short Term Issues
The real issue is whether or not there is truly a problem. The RBOC studies claim there are bottlenecks, yet other groups, such as the IAC, show the weaknesses of the RBOC’s studies and contend that there is no real problem. Therefore, any potential short term solutions depend entirely on whether one group can prove their position is accurate, i.e., that there either is or is not a bottleneck in the PSTN caused by the tremendous growth in data calls.
The first solution is to leave things as is. Unless the RBOC’s can come up with better studies that do in fact show there is a real problem, unlike their current studies, and that the additional revenue they are gaining from the growth of second phone lines used for Internet calls is not enough to upgrade their networks, the FCC will most likely decide to not allow the LEC’s to charge access fees to the ISP’s.
However, if the RBOC’s can prove that there is a substantial problem, and that they truly are not generating enough revenue from the Internet calls to upgrade the network, then the solution will be the access fees the RBOC’s are fighting for. This will mean that the ISP’s will ultimately pass the charges on to the consumers, and the days of $19.95 unlimited Internet access will be over.
The only other possibility is that there really is a problem, but the FCC rules the LEC’s must upgrade their networks without charging access fees. The FCC must base their decision on the determination of if there is a problem or not, and if there is, who should pay for the network upgrades necessary to fix it.
Long Term Solutions
As the Information Highway continues to expand, eventually nearly every home and office will be connected to it. As that happens, the local telephone network as it is currently designed today clearly is not the best solution. It was designed to handle voice connections only, which means it is circuit switched (i.e., a connection from point A to point B is established at the start of the call to reserve the bandwidth just for that connection, and terminated when the call is complete). Data connections, on the other hand, are generally packet switched, which means no bandwidth is reserved in the network at the start of communications, and the packets are not restricted to a single path through the network. The fact that voice and data communications are handled so differently today means that there are two possible solutions for tomorrow. Either the voice network and data network can split and be totally separate, or they can merge and become one.
Split the Networks
Since data communications is inherently different from voice communications, many feel that the two networks should be totally separate. Today, the data network backbone is separate from the voice network backbone, but to get to the data network from a small home or office, an interconnection through the local voice network is necessary. However, the connection from the home to the data network can be kept completely off of the voice network, thus freeing the voice network of all data traffic.
This is the position of the IAC, whose study states that “the long-term solution for accommodating increased data traffic lies in the stimulation of competition and in the deployment of appropriate data-friendly network technologies…(7)” Also, the DATA Coalition, a group similar to the IAC, states that they are “trying to convince the FCC to create incentives for phone-company rivals to build alternate networks for Internet access (3) .”
Several companies are working on such solutions already. For instance, Bell Canada is developing ways to intercept local calls destined for Internet Service Providers so they can be routed around the traditional voice PSTN (5). Also, Northern Telecom is developing Internet Thruway (12), which allows telephone companies to move Internet users off their voice network and onto a more optimal data network (5).
Other new technologies, such as the xDSL suite, use the current local loop for both data and voice, but separate the data from the voice network before the local voice switch. Cable modems completely move data off the voice network, from the house out. Both of these technologies eliminate bottlenecks in the voice network caused by data calls.
Merge the Networks
Instead of splitting the voice and data networks, there is also the potential of merging them. Asynchronous Transfer Mode (ATM) is a new networking technology that attempts to do just that. ATM was jointly designed by both the telephony and data network communities, and was designed to handle any and all traffic in a single network architecture. Many in the industry feel that “…it is clear that ATM integrated-services networks are going to replace today’s single service networks, and become the networking standard for the telecommunications industry in the near future (10).”
If this is done, all traffic from the house out to the backbone network will be digitized at the house, and carried through the network as digital cells. The potential for bottlenecks, then, will not be because of the dissimilarities of voice and data and the fact that they are carried on two separate backbones, but because of sheer volume only. This, of course, can be eliminated with careful design and capacity planning.
The local exchange carriers (LEC’s) state there is a problem with data connections to the Internet and other data networks causing congestion in the voice network, and they want to charge for these data communications calls. Others, such as the Internet Service Providers (ISP’s) and the Internet Access Coalition (IAC), disagree with the LEC’s claims of such a problem and that the additional charges should not be allowed. It has yet to be proven if there is a problem or not, so any short term solutions must wait until that issue is resolved.
However, it is clear that with the tremendous growth in data communications, if there is not a problem today, there will be one tomorrow if changes are not implemented. The long term solutions involve splitting the voice and data networks completely from each other, or in developing a single network suited to handling both types of information (i.e., an ATM network).
In an ideal world, all parties would work together towards the ideal solution, but in today’s environment, it is each for his own. However, with the Telecommunications Act of 1996, and an age of new competition looming on the horizon, perhaps mergers between LEC’s and ISP’s will occur. Such a merger would mean voice and data access to the two networks would be handled by a single company, and that company would of course try to maximize profit by installing the network best suited for both tasks. Such a network would elimate any congestion caused by data traffic on a voice network.
(1) “The Internet is Jamming Phone Lines.” The Salt Lake Tribune, November 10, 1996.
Available at URL http://www.sltrib:80/96/NOV/10,tbx/21444028.
(2) “Pay Per Minute.” PC Magazine Online. 3/3/97. Available at URL
(3) “High-Tech Companies Fight Per-Minute Charge for Internet Access.” David E. Kalish, AP Business
Writer.” November 8, 1996. Available at URL http://www.sddt.com/files/librarywire/
(4) “Phone calls to the Internet could find up costing more.” Michael L. Rozansky, Inquirer Staff Writer.
Available at URL http://www2.phillynews.com:80/tech.life/ISP12.htm.
(5) “Internet gridlock threatens phone lines.” Paul Lima. January 1997. Availalbe at URL
(6) “Net use strains phone lines.” James Kim, USA Today. Available at URL http://www.rpf.com:80/
(7) “The Effects of Internet Use on The Nation’s Phone Network.” Selwyn, Lee L. and Laszlo, Joseph W.
Prepared for the Internet Coalition, January 22, 1997. Availalble at URL http://www2.itic.org/
(8) “What You Can Do.” The Internet Access Coaltion. March 2, 1997. Available at URL
(9) “The Internet Access Coalition.” Internet Access Coaltion. Information obtained from the IAC’s
Web page, availalbe at URL http://internetacess.org/.
(10) “Pricing of ATM Networks.” Wang, Qiong, et. al. from The Internet and Telecommunications
Policy, Selected Papers from the 1995 Telecommunications Policy Research Conference.
(11) “Notice of Proposed Rulemaking (NPRM).” Produced by the FCC. Dec. 1996. Available at
(12) “Nortel Internet Thruway.” Northen Telecom Web Page. 1997. Available at URL