To: Frank A. Coluccio who wrote (8567 ) 1/3/2005 5:50:30 AM From: Sam Salomon Respond to of 46821 Thanks, Frank, for your detailed response.I recently spoke with a VC friend about the plausibility of doing a rollup of non-dominant players in the emerging fiber-to-the-x arena, along with some of the smaller cable operators and VoIP plays. Muni nets and independents also came into focus. If I understand your idea correctly, it would require two major phases, namely the roll-up of the players you mentioned (Phase I) and capacity purchase contracts between the roll-up and LVLT/WilTel (Phase II). I would guess that both LVLT and WilTel would be interested in direct consolidation of the interstate networks. IIRC we have about 8 major interstate networks in the US (ATT, MCI, Sprint, Qwest, WilTel, Broadwing, Global Crossing, Level 3) although Broadwing and Global Crossing may not be counted in the 1st tier in the US IIRC, at least for certain services. Still 6 is obviously too much, I would think that three to four would be the max for a healthy long haul industry. As you rightly remarked WilTel and Level 3 seem to have limited end user (consumer or business) operations, while ATT, MCI and Sprint have a strong end user focus. So the easiest consolidation would be that one of the end user focussed organizations would outsource some of their wholesale/long haul operations to the wholesale focussed carriers, and from your post I take it, it would be most coveniently done on the lambda level. If that trend would come into reasonable force within three years, that would IMO permit Level 3 and WilTel to survive profitably, free ATT, MCI and Sprint of some capex and probably even reduce their opex. I am, however, too much of an outsider, in order to have any gut feeling, what the thinking at ATT, MCI and Sprint is. It took IBM years of unprofitable PC business (and everybody knew for years or decades that their PC business was unprofitable) to sell it. I don't think that it makes sense for a Sprint or an MCI to compete with Level 3/WilTel on lambdas and IP backbones. However, if they behave like IBM, they will do it for another 10-20 years and won't mind the cost to their organizations and the industry. As neither IBM nor most other PC players minded the enormous overcapacity in the PC industry for decades.