I’ve been procrastinating on my VIC obligations. I’ve been sitting on this idea for several months now and unfortunately the stock’s run up quite a bit. Fortunately, there’s plenty of meat on the bone still and I forced myself to sit down and write.

LUMOS NETWORKS

There’s been quite a few other telco ideas on the boards in the last few months. LMOS touches upon some familiar topics. It is a hybrid – there is a fast growing fiber business glommed on top of a crappy RLEC/CLEC business. However, unlike some other telcos that share this characteristic, LMOS is actually growing both topline and EBITDA. The legacy businesses have declined to less than 50% of sales on a run rate basis, and will be less than 40% in a couple of years.

A Very Brief History:

LMOS is a spin off – it separated from NTELOS in late 2011. And unlike other spin offs from that year, it basically crated – in this case from the high teens into the high single digits. There was almost zero sellside coverage on it, and for a long time it was basically left for dead – on the surface it appeared to be just another terminal decline telco.

LMOS (and NTELOS) was cobbled together by Quadrangle Partners, which owns 26% of the equity and has three seats on the eight member board. Over the years, they acquired a grab bag of assets, mostly in the mid-Atlantic / mid-Appalachia regions, comprised of wireless telco (now NTELOS), RLEC/CLEC, and a ton of fiber. The fiber came pretty cheaply -- $163 million in 2010 for FiberOne and $27 million for certain assets belonging to Allegheny Energy in 2009. These totaled 3,000 route miles of fiber, and the acquisitions were made for a middish single digit multiple of EBITDA.

As you might have guessed, when the spin happened, NTELOS kept the wireless portions and LMOS got the RLEC/CLEC, and the fiber.

Business Strategy / Dynamics:

LMOS’ business is nearly exclusively in western Virginia and West Virginia. FiberOne had a few doglegs of fiber extending to Pittsburgh and Southwestern PA, but LMOS currently don’t have a lot of business in these places.

The Central Appalachians are not exactly a wealthy or fast growing part of the country. There are a few relatively large and somewhat prosperous townships (by large, I mean like 50,000 people). But in general, it’s a mountainous, relatively poor, slow growing area. I’m sure whoever built all that fiber back during the telco bubble never made a dime.

However, the geography and economic situation of this area has one major advantage – there isn’t nearly as much competition as the rest of the country. There is some competition of course, but LMOS probably has the most robust network in its regions.

Also, being a relatively underconnected area of the country, LMOS has been able to find a good amount of growth. Its “Strategic Data” segments have been growing mid-teens each year and will likely grow at a similar pace for the next several years.

At this point, I should also mention that LMOS is almost entirely an enterprise data business. It does very little fiber to the home (it did get some business from an ARRA grant a couple of years ago, but the contribution is pretty small). In this sense, LMOS is very much like an Abovenet. Sure, the service regions don’t have the same building density as New York, but competition is less as wel.

LMOS’ website has maps of its on-net buildings and fiber networks. I don’t think the maps have been updated in several quarters, but it gives you a pretty good idea of where they are and which buildings they are connected to, as well as the buildings they are passing but not connected to. You can corroborate the addresses listed with the management team’s statements in this regard. In short, they tend to be mostly medium/large businesses, government, healthcare facilities, and educational facilities.

I should mention another very important driver of growth – fiber to the cell. This is not a new story for these telcos. Management says there are 2000 “near net” locations that could be in play. I don’t think they will get to 2000, but the opportunity is very significant as LTE rollout continues for the next several years. They had a target of reaching 300 towers by YE2012, and ended up handily beating their own goal with 370 towers connected, up from 148 the previous year. In my conversations with the company, they said a more reasonable target of towers connected is 1000 in 3-4 years. Growth in 2013 should be pretty robust, but decelerate thereafter. I believe on a run rate basis this tower business is maybe 10% of the “Strategic Data” revenues. So it’s relatively small, but very high incremental margin and fast growth.

Growth Trends, Financial Performance, Modeling Assumptions

As mentioned before, the legacy RLEC/CLEC business had been shrinking. A mid-teens pace is about right, and should continue for the foreseeable future. Fortunately, the RLEC/CLEC business combined has a similar EBITDA margin to the data businesses, so at least LMOS wasn’t replacing very high margin revenues with low margin revenues as some other telcos are.

Anyway – here is why the stock has moved from $8 to $13 since the fall. In Q3, the revenue trends finally turned the corner. The company is actually growing again, albeit modestly. Q4 was even stronger, and they put out pretty bullish revenue and EBITDA guidance. In Q3, on a run rate basis, Data became the majority of revenues. Given the mid-teens growth in data, and similar shrinkage in Legacy, revenue growth could actually accelerate past 2013! The firm should exit 2013 with Legacy roughly 40% of revenues, and 2014 with Legacy at 30% of revenues.

The weird thing is that management had been pretty vocal about these trends. And telco is a pretty predictable business so it could have been easily modeled. But the company was so small that I guess nobody cared. Even after the very, very bullish Q4 release, which projected 2013 EBITDA at roughly $97 million, versus $88 the previous year, the stock barely moved and then proceeded to grind up every day for several weeks. The same thing happened post Q3 when revenue trends turned.

How to Think about Valuation

LMOS is not an expensive stock – it trades just over 6x 2013 EBITDA. We can take a stab at valuing this business – the data segment is frankly pretty damn good, with high margins, decent defensibility, and solid growth. It seems similar telcos are largely trading round 9x EBITDA – TWTC, Charter, etc. Abovenet was taken out by Zayo at a similar multiple.

The CLEC/RLEC obviously is pretty crappy. I probably wouldn’t pay more than 3-4x EBITDA for them, even though they are pretty cash rich and it’s possible that in several years the declines will moderate.

If you take a grab at the blended multiple, it ought to be around 7x to 7.5x, which implies a stock price in the high teens. As the legacy businesses continue to shrink, valuation should improve further. Keep in mind, the rich cash flows from these businesses are finding high incremental returns in the data biz – this is not a dynamic seen in many of these “terminal decline” telcos.

Fast forward another year or two, and put an 8x multiple on higher EBITDA (I’m personally projecting $102-$105MM in 2014, and $105-$110MM in 2015), it’s not hard to envision a stock in the $20s. Oh, you are getting paid a pretty nice dividend while you wait too.

Chances for a Sale?

It’s my opinion that this company will be sold in the next several years. Potential buyers include Zayo, CTL, WIN, Time Warner Cable, among others that have contiguous or partly overlapping networks.

First of all, the 2011 spin simplified the NTELOS story quite a bit. I interpret this has making both pieces more salable. Second, Quadrangle had been stuck here since maybe 2006 and I’d guess are looking to make an attractive exit at some point.

Also, they hired a new CEO (Tim Biltz) in Apr 2012. Tim’s previous experiences include iPCS, Spectrasite, and Vanguard Celluar. He was CoB of iPCS. Anyway, all three had been sold for pretty hefty multiples. They lured Tim out of retirement, and gave him 500K options on top of 165K of restricted stock that. I bet he wants to get paid.

I should also note here that if there's a sale, it will be post the 2-year period from the spin for tax purposes... I'm guessing it will mostly likely be in 2014. Make your own assumptions on synergies / control premium paid, but needless to say if I'm right, it really doesn't matter -- everyone will make a lot of money.