Would Level 3 Really Sell Its Data Centers?

In their earnings call last week, Level 3 Communications (NYSE:LVLT, news, filings) mentioned that they have been re-evaluating their colocation business.  Specifically, in response to an analyst question about their plans for their colocation footprint, CEO James Crowe made the following comment:

“This is a matter of some strategic importance.  We’re doing a review that is largely completed.  We will make some decisions here in the coming months.  Our data center business would be one of the larger colocation/data center businesses in the industry if it were a standalone.  It’s a big opportunity for us.  We want to make sure we take advantage of it properly.”

Since then, I have received several questions from readers regarding whether Level 3 is likely to sell this segment of its business.  I didn’t post on the subject immediately, though, because I felt it required some thought.

Now, on one level it’s easy to see why this might be attractive.  Level 3 has a huge colocation footprint that it has always operated as carrier neutral, in that they encourage other providers to offer access in their facilities.  However, because they are (obviously) a carrier, their facilities are still not generally viewed as the same kind of ‘carrier neutral’ as those of a pure colocation provider like Equinix (NASDAQ:EQIX, news, filings) – it’s somewhere in between.  So while the colocation business is good and growing for Level 3, it is not valued by the markets at the same level as it might be if the same facilities were operated wholly separately.  Therefore, the idea of separating it and selling it off has come up before and will surely come up again.  From a spreadsheet point of view with some broad strokes, it certainly makes a lot of sense.

I do think that Level 3 is facing some decisions here.  To put it bluntly, a colocation business requires capex for new space and, especially, power to grow.  That is why we have been seeing Equinix, Dupont Fabros, Digital Realty, QTS, Telecity, and others raising and spending major dollars.  Level 3, however, must always manage its debt and cash flow very carefully in economic times such as these, and therefore is not likely to start spending piles of cash on colocation projects the way things are currently set up.  So to keep it growing they will need to build out more of their space, but the capex for that may not be there.  Selling the colocation business so that others can fund its expansion would perhaps solve many of their cash problems.  But I really don’t think such a sale is going to happen nevertheless.

Why?  For one thing, it would be very difficult to separate.  Level 3 uses much of these facilities for its own operations and would of course still need to do so.  They are deeply integrated with the rest of the company’s business, and there would be substantial integration expenses required (or is that disintegration expenses?) just to make it ready.  But more important is perhaps the fact that only a fraction of their footprint is really of the type that competes with Equinix in the main colo markets and most of that is largely full.  The rest is often in less active colocation markets or has space or power limitations that other providers solve by simply building giant new standalone facilities.  Despite how it might look on paper, separating this segment would be more costly and painful than one might think, would bring in less money than one might project, and would very much distract Level 3 from solving its real problem of finding a path to real growth in the world of bandwidth.

HOWEVER, that doesn’t mean that Level 3 can’t do anything here.  What I saw in Crowe’s comment was not a sale, but an indication of a possible organizational shift akin to its local markets initiative that decentralized power to metro fiber markets.  If they have sufficient reason, and the necessary cojones as well, they just might follow the lead of Zayo and find a way to horizontally disaggregate the colocation business.  They could functionally separate it, such that it really is carrier neutral and (hopefully) viewed as such by the rest of the industry.  Their carrier business would be the flagship customer of course, but it would have to negotiate for good rates just like anyone else, while other carriers willing to pay more could outbid them.  Thus the colocation business might find its way to the 20%+ growth rates its independent peers have found, while the bandwidth business might untether itself further from that footprint and thus find more opportunities outside it.

It not easy to do though, let alone to convince the marketplace of.  But it might let both the bandwidth and colocation segments of Level 3 focus better on their respective strengths, and therefore help them to turn their huge asset base into more value for investors.  Just a thought…

If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!

Categories: Datacenter · Internet Backbones