Ever since the news earlier this week that Sprint Nextel (NYSE:S, news, filings) might outsource the operation and maintenance of its cellular network to Ericsson (NASDAQ:ERIC, news, filings), a question has been tugging at me. Why is it that outsourcing of wireless networks is so hot, yet outsourcing of fiber networks is not? In this particular case, why does Sprint not choose to outsource its longhaul network instead? After all, they have been running the wireline business for cash rather than growth for years now. Much of the fiber that makes up that network is some of the older that is still in operation, some of it is even direct-buried rather than running in a conduit. One would think that as a candidate for cost savings, turning over the transport network to someone like a Qwest or a Level 3 would free up those strategic resources while still allowing them to work on their growing IP services business. But instead it is the more modern wireless business whose underlying network is apparently non-core.
It's not just Sprint of course, there have been several wireless operators internationally that have outsourced existing networks in the last year. But the concept of outsourcing a network is not new, it's just that the technology is different. The 'inevitable' outsourcing of fiber networks was a dream that helped drive the fiber boom in the late 90s; the economics seemed to favor everything being rolled onto the newest fiber and wholesale was going to be huge. But the only major deal of this type that ever actually happened (I think) was the SBC/Williams deal, and as we all know SBC bought AT&T and brought it all back in-house. It turned out that network operators did not want to yield control over their networks. Even when they were willing, the pricing just didn't work - there was no room in the margins for middlemen.
Now, I'm not saying that they were wrong back then or even now. I'm just puzzled by what actually makes wireless networks into better candidates for such an outsourcing. Both require skilled techs, wide geographical reach, standard equipment, and lots of communication - is there an essential bit of wireline networks that makes it unsuitable? Or, perhaps there is no substantive difference and this trend will also be short lived. Either the equipment manufacturers may find the costs to be higher than they thought or the wireless carriers may find that the savings aren't worth the extra hassle and lack of direct control. My gut tells me these deals look a lot better on a spreadsheet than in the field, but that may just be my cynical streak showing.Internet Backbones · Wireless