IP Telephony Goes Mainstream
March 23, 1999 - 9:30 AM
By Roderick Beck
A year ago, many pundits were predicting the death of Internet telephony due to rumored changes in FCC regulations. As usual, they were wrong.
Not only is Internet telephony thriving in the international telecommunications markets, but many of the largest U.S. carriers, like AT&T and GTE, are planning or building IP (Internet Protocol) networks to handle virtually all their domestic traffic: voice, data, fax, videoconferencing, Internet and Intranet.
Let's now take a look at some of the key events of the past twelve months and discuss the future direction of Internet telephony.
The Access Charge Battle
There are two key facts to understanding Internet telephony. The first point is that most Internet telephony runs not over the public Internet, but rather over better-managed private networks that use Internet technology.
The second key fact is that long distance companies pay for the right to use the local phone company network to originate and complete toll calls. These 'access' charges, which are used to subsidize local residential service, currently total 3.71 cents per minute, the single largest cost in the long distance business.
In 1997, long distance providers paid in excess of $20 billion in access payments. The appeal of Internet phone calls is that current FCC rules exempt traffic carried by a network based on Internet technology from paying these access charges. Without the rules, local phone companies could levy per minute charges for anyone dialing into the Net, a highly unpopular idea.
In April of 1998, reports surfaced that the FCC was planning to eliminate IP telephony's exemption as a result of Congressional prodding. The fear was that exploitation of the FCC's loophole by long distance carriers would deprive the local phone companies of revenues and force them to raise local service rates.
In an early Raging Bull article, I correctly suggested that the FCC would not upset the status quo given that this would ignite a political storm. Subsequently, the FCC fudged by declaring that the issue should be decided on a case-by-case basis.
Frustrated by their inability to get the regulations changed, several months later US West and BellSouth declared they would unilaterally impose charges on IP voice calls. Their argument is that the original intent behind the access charge exemption was to protect data traffic, not voice. This is quite true. At the time the exemption was put in place, placing phone calls over data networks was not possible.
In classic tit-for-tat, Qwest and IDT responded that they would not pay them, and one VOIP provider, ICG NETCOM, plans to take BellSouth to court. The most likely scenario is that a drawn-out legal battle will ensue in which the courts, lacking the expertise, will refer to the FCC for a definitive ruling.
As a result, domestic Internet calls appears safe for at least another year, probably two. Even though VOIP appears safe for the time being, many IP voice providers are hedging their bets by looking for alternative ways to reach their customers without using the local phone networks. AT&T plans to use cable networks, and CLECs will leverage their own local networks. In neither case will access charges apply.
Interestingly enough, the predictions of access charge melt down have not come true. The reason is simple. The long distance giants are afraid of Internet telephony. They are wracked by internal battles over the merits of the technology with the older generation 'BellHead' engineers championing the traditional circuit-switched technology and younger 'NetHeads' proselytizing on behalf of the emerging Internet technology.
As a first hand witness to this debate, it has many of the trappings of a religious war. Another factor holding back the largest carriers is that they cannot do Internet telephony on a large scale without triggering FCC regulation, which would destroy the value of their investments in the required infrastructure. Indeed, the dominant carriers, having lost over 20% of long distance revenues to smaller players, fear the impact of Internet telephony on prices.
This contrasts sharply with the competitive fringe of smaller long distance carriers. Many small carriers are leasing very expensive long distance circuits and exploiting the fact that 5 or 6 IP calls can be squeezed into a single phone line. This offsets the high costs of leasing circuits, lowering per unit (minute costs) and reducing the heavy capital requirements that discourage entry in the long distance business.
However, the large carriers own fiber, meaning their transmission costs are already very low. Hence, IP telephony is a great equalizer in that it helps smaller carriers considerably more than it benefits larger carriers. Given that the cost of adding network capacity to handle an extra minute is less than half a cent for the largest carriers, residential rates of 9 or 10 cents have fat profit margins.
Small carriers are beginning to take aim at these profits. IDT's Internet phone service offers consumers 5 cpm for all long distance and regional toll calls without monthly fees or minimums. For the small carriers, IP telephony is a price weapon.
Large carriers like AT&T and MCI view IP telephony very differently. They want IP networks in order to simplify their networks and provide advanced services that are difficult and expensive using traditional technology. Currently phone companies have many distinct networks to provide a variety of different services.
A videoconference requires coordinating the voice network with a streaming video network and requires advanced scheduling. More generally, any service that mixes voice and data can be handled with greater success by using an IP network. The dream of young engineers are 'Convergence': a single IP network to carry all traffic.
Such a network would drastically reduce overhead. It would also create economies of scale, because several smaller networks would be replaced with one larger network. Another key advantage is introducing new services and upgrading them. The Internet Protocol is an open standard.
This means that it is ideal for offering new services and upgrading them. For example, when AT&T decided to improve the sound quality of its voice network a few years ago, it discovered that it was very difficult to do so without replacing hardware. However, an IP network can easily do so by change the software code used to process sound signals.
Traditional networks are based on hard-wired proprietary technology. By contrast, IP networks use software-based open standards. The result is that what is easy to do on the IP network becomes virtually impossible on a traditional network where technology is embedded in the hardware.
- Roderick Beck's Message Board:. |
