Death of T1 and Rise of EFM

In November I blogged about T1’s Expiration Date, but predicting the death of T1s and the rise of the new age of Ethernet in the First Mile over copper without a strong economic case would be neither credible nor complete. The economic case for the end customer and carriers must be factored into the discussion.

For end customers – enterprises in this case – value is derived from the applications that rely on multiple factors, including service availability and the available bandwidth, not the underlying network alone. The underlying elements, which enable delivery of these applications, become a component of operational cost. A carrier’s inherent need to increase – and even maintain – customer penetration while focusing on the immediate bottom line has highlighted the compelling economic case for selecting the most cost-effective method of delivering increased bandwidth. In most cases that involves replacing legacy T1s, or the equivalent E1s outside North America, with the most cost-effective alternative.

While the actual numbers vary between and even within countries, the drivers that deliver the savings are similar. One of the drivers of opex is protocol translation. Technology that leverages native Ethernet as its underlying protocol (the same protocol that is used in LANs) eliminates the need for protocol translation between LAN and WAN interfaces. Ethernet is widely understood, well proven, and has simple management and configuration tools available. Ethernet has plug ‘n’ play capabilities requiring no specific training on the part of the enterprise, which reduces IT staff costs and, therefore, opex.

From a carrier’s perspective, the drivers of operational costs are cost per bit (cost of a unit of bandwidth) of the bandwidth delivered, and the cost associated with managing the circuits delivering bandwidth. Bandwidth cost is directly proportional to speed. The complexity of management is associated with the number of circuits required to deliver the bandwidth and protocols used to deliver the bandwidth.

The cost per mbps of bandwidth delivered by T1 (or E1) for a 3mbps or 6mbps services is about 45 percent higher than available copper alternatives. For speeds above 8mbps, T1 and E1 circuits are seldom an option because of limitations on the number of circuits that can be cost effectively bonded, driven by the availability of copper pairs and the cost of network hardware required to support such configurations.

It would be problematic to suggest that, while T1s and E1s have disadvantages, the only alternative is fiber, especially when the existing copper can exceed the growing bandwidth demands. In fact, only a small fraction of businesses (less than 20 percent) have access to fiber. When compared to T1s and E1s, EFM over copper delivers to 20 times more bandwidth per copper pair, eliminating construction costs associated with laying fiber.

Let us proceed with the assumption that carrier pricing for higher bandwidth service is based on the number of loops used rather than the actual bandwidth. To illustrate further, if a carrier charged $350 for a 1.5mbps T1 service – which uses two pairs of copper – they would charge $350 for a 60mbps EFM over copper service, which uses the same number of pairs. When aggregated over the entire base of customers that currently use T1 services, the opex savings achieved can run into millions. This logic applies whether carriers are purchasing loops from incumbents or incumbents are utilizing their existing copper loops – using fewer loops to deliver equivalent service translate to lower monthly operational costs.

If one challenges the assumption – which would reflect reality – that customers are unwilling to pay more for higher bandwidth, then it is clear that the carriers will also grow their revenue. While there are opportunities to charge customers more for higher bandwidth, it is also clear that customers will not pay 20 times the price of a T1 for 20 times higher bandwidth. It is more likely that customers will pay a smaller multiple of the T1 price, which translates to much better value to the end customer when measured as dollars paid per mbps of bandwidth. Even if the carriers charged a 10 percent premium over T1 services for 20 times higher bandwidth, that would result in 10 percent revenue growth without adding a single new customer! Add to this the prospect of growing the customer base and the numbers become even more appealing.

There are two more factors that must be considered: the cost of delivering the additional bandwidth and the capital expense incurred in deploying EFM over copper services. The incremental costs for additional network bandwidth are a very small part of the overall cost of the service. With the magnitude of the savings and revenue increases, that cost becomes insignificant. With the types of savings described above, the payback period for the upfront capex, for the EFM equipment, is going to be aligned with the current economic climate – and that’s months not years.

So, to the T1s and E1s, we thank you for your service and rise to toast your legacy!

Eric Vallone, vice president of marketing at Actelis Networks, is responsible for setting the direction of the company’s product portfolio, as well introducing its products to the market. Prior to Actelis, he was with AT&T, ADTRAN, and Paradyne Networks. His areas of focus included DSLAMs, mobile backhaul, xDSL, VoIP and VoATM products, and network management solutions.