How convergence could cost you money

By David Rohde
Network World, 06/14/99

Most people who are skeptical about converged voice and data networks say they think shipping voice over packet networks might not work very well. But even for them, it's often an article of faith that if convergence works, it will save users money.

Is that true? Even that much is not totally clear, according to a few acute observers of the carrier scene.

Certainly if you take today's transport costs and start eliminating some toll charges, you'll save money.

But no one believes transport costs are going to stay where they are forever. On the broadband side, carriers are installing 1.5M bit/sec digital subscriber line circuits that are a fraction of the cost of dedicated T-1s; optical networks that eliminate expensive electronics; circuits that split bands of light into separate paths for four, eight, 36 or whatever times efficiency.

On the narrowband, voice-grade side, telephone tolls keep dropping. On July 1, local exchange carriers have to lower their access charges to longdistance carriers again. Is your carrier passing along the savings? You might want to bring that factoid to the negotiating table.

Take note of the converged voice/ data network offers coming from carriers. I think you'll see that most of them so far, and almost certainly most in the near future, will be managed nets. That means there will be many charges in any contract, including not just the usual access, port and circuit charges, but also equipment leasing fees and so on.

Telecom analyst Steve Sazegari also points out another factor in the convergence stakes. Among the legacy carriers, convergence has become a fancy way of saying "ATM" without going through the heartache of selling that long-maligned technology. Usually through a carrier-managed ATM access box, the carrier can aggregate voice, video, fax and data traffic onto a single circuit.

But how much simpler it would be, Sazegari notes, to take native IP and throw it directly onto those screamingly fast optical networks that carriers are now examining. If nothing else, this would eliminate ATM's overhead and the need to fool with its multiple adaptation layers. Sure, without ATM maybe you couldn't yet be sure that tollquality telephone calls could ride over the same network. But at some point - as corporate voice drops below 6, 5, then 4 cents per minute and perhaps even less - the returns from doing so grow less and less.

Asked about convergence cost savings, many carriers like to talk about "total cost of ownership," as in, "Our service will save you 15 to 25% off the total cost of ownership of managing dual networks." What they don't want to talk about is the actual savings. The reality is that much of convergence today is about account control - the ability for carriers to lock you in as they face growing competition.

So the jury is still out on the real question: Will convergence save you money, or is it an excuse not to lower prices dramatically on existing broadband data services right now?