The following text is copyright 2002 by Network World, permission is hearby given for reproduction, as long as attribution is given and this notice is included.

The end of a bad idea?

By Scott Bradner

It was not all that long ago that Nortel's $19 billion paper loss in just 3 months set a standard for fiscal prowess that other businesses could only wonder at. Now Enron one upped Nortel by losing about $23 billion in the last quarter of 2001 on top of a loss of $68 billion for the rest of 2001. And this was not the paper "good will" types of losses like Nortel's were, this was the actual value of shares on the New York Stock Exchange. Enron topped off an impressive 2001 by declaring bankruptcy, the largest in history, a few weeks ago. I wonder who will try to top this achievement.

One of Enron's big things was acting as a broker. They inserted themselves between the producers and consumers of some commodity, such as electricity. Apparently both the producers and consumers liked the arrangement since Enron was very successful at this -- accounting, for example, for as much as one quarter of the energy trading in the U.S. even though it never produced all that much energy on its own. Enron was a very successful company -- at its peak claimed annual revenues in excess of $100 billion.

The Enron web site (http://www.enron.com) lists 32 product areas from power to fertilizers. (There must be a lesson in that last category but I won’t go there.) This column only concerns one of Enron's product areas -- "bandwidth."

At first glance brokering and selling bandwidth seems like a reasonable idea but I'm not sure that its time has come, if it ever will. Enron's on-line slide show explaining their idea of bandwidth trading (http://www.enrononline.com/jsp/marketing/Markets/Bandwidth/US/index.jsp) starts with a quote from Machiavelli which talks about the lukewarm support new ideas get from the establishment, what it does not mention is that some ideas deserve lukewarm or no support because they are bad ideas.

Enron's model of bandwidth trading was to permit companies to get network bandwidth between aggregation points when it was needed and for the length of time that they needed it. This bandwidth could be used for special events such as televising sporting events or to augment an existing network. They felt there was an opportunity in this area because of the inflexibility of the traditional players in the field. Historically, one purchased bandwidth in fixed sized chunks for extended, often multi-year, periods, which took months to negotiate and provision.

There are a few problems with Enron's idea. First, much of the problem with bandwidth these days is that it does not exist just where it is needed (no fibers), and second, the traditional model is dying. IP-based connectivity is easy to get and can satisfy most requirements. You do not need dedicated connections for all but the most demanding applications. It is very hard to see how one can make a viable business model out of satisfying just these niche cases.

The bandwidth trading concept was bankrupt long before its champion.

disclaimer: Bankruptcy is not a common topic at Harvard (maybe at the Law School) so the above is my own opinion