Level 3 trades at a huge discount to replacement cost and this gap is likely to close within a few years. If it does, Level 3 will be a good long-term investment due to its unique and irreplaceable assets.
In essence, Level 3 (like Peter Kiewit, its Omaha based former parent) provides vital infrastructure. In 1985, Walter Scott and James Crowe decided it was a good idea to lay conduits (= pipes/ducts) along the railway tracks connecting all major US cities.
This is something you want to do well once. Scott and Crowe's idea of doing it well meant they burried 10-16 pipes along the tracks of Burlington and other railway companies. They filled one conduit with some fiber optic cable. This network is the single best Internet backbone by far. 92% of all the worlds ISPs pay Level 3 for the priviledge of connecting to this backbone. http://as-rank.caida.org/
In return, data originating on Level 3's networks, gets a free ride on every single IP network on the planet ! Companies that do not pay Level 3 to connect (true peers) are Qwest, Verizon Business Sprint, TeliaSonera, NTT Communications, Tinet, Global Crossing, Savvis, AT&T and Tata Communications. The list is dynamic; read the article http://en.wikipedia.org/wiki/Tier_1_network
Remember, that is one of a dozen conduits filled. No one has a national network of conduits that comes close. There are many global IP networks and they typically use Level 3's routes.
Others have caught on to the idea of laying down fiber along railroad tracks. Not on a national scale mind you and no extra conduits... What they do is mount a plough on a railroad car and connect two points. The good news is that this provides us with an estimate of the replacement value of Level 3's assets. http://www.telecomramblings.com/2010/05/allied-fiber-splashes-into-the-fiber-pool/ http://www.heraldtribune.com/article/20100909/ARTICLE/9091064?p=1&tc=pg
$100k per long-haul mile...... x 54k miles = 5.5 B
In the case of Level 3 we are talking about a national network and not just a single route that is useless for a company like say..... hulu.com
IMHO Level 3 could sell 2 of its 12 nation wide conduits to a knowledgeable buyer and repay its long term debt. This is probably why some gurus were willing to buy all that debt. I can also see why mr. Crowe would rather cut off his right arm than sell one of his conduits.
Level3 also owns some 27k miles of Metropolitan fiber. The replacement cost of these assets is…… prohibitive. But let’s use $ 150k per mile. Estimates you find that ammount to less probably don’t include the RoW (Rights of Way) or may be based on the cost of installing the stuff in an existing conduit. $150k per metro mile x 27k miles = 4 B
150k makes sense. Doing this in a city is probably more expensive than doing a deal with a railway company and sending a railcar mounted plow down the line. Of course, Level 3 has lit many miles of fiber with expensive network gear (transceivers) and owns the best trans-atlantic routes. They also own some stuff in Asia and Europe..... but this article is too short. 3 B
Level 3 claims the current replacement cost of its assets is about 23B. Well, I'm a pessimist.
Level 3 is selling the network capacity from that single network on one of its conduits to wireless customers, cable customers, internet companies and traditional carriers. They offer these customers both long haul and metro connectivity. The fact that the data can then travel anywhere on the planet at no cost gives Level 3 a cost advantage over a number of competitors (Q, GLBC, XO). These companies have to buy either metro connectivity or Long-haul. I believe this business will grow at double digits over my lifetime.
Let's kill this thesis
1) I don't buy replacement value arguments. The excess conduit laid down during the great silliness should last 100 years.
That’s right, at least in Level 3’s case. You do it right, once. It’s a good thing, as long as you can eventually earn a reasonable return on capital. That day will come when demand catches up with supply.
2) Many similar companies have been sunk by their debt; wiping out the equity.
Yes, but not Level 3. These guys have been at the helm for two decades and have, again and again, managed to secure backing from poeple with deep pockets (Warren Buffett, Francis Chou, Prem Watsa, Mason Hawkins). These guys bought all the (convertible) debt they could. Some are now buying the equity directly.
3) This thesis has brought nothing but grief to investors since 2002.
No. The company never sold for less than my pessimistic estimate of replacement value.
4) Why Level 3 ? XO, Q etc. etc. compete but earn profits.
Good point; but most of these have been restructured. Equity was wiped and debt reduced. PP&E (the cost of those cables) written down to nothing. Now all they have to do so the accountant can report profits is buy some transceivers and depreciate/maintain those. They no longer deal with repaying the debt and depreciating the cost of the network. They understate Capex !
Level 3 is losing money because capex reflects economic reality. Also, they still carry that debt. The good news is, Level 3 now has big NOLs. Some of these guys used to work at Leucadia. The NOLs are one reason Hawkins and Watsa would not want to trigger a change of ownership.
Why now ?
1) Burying fiber has become fashionable once again. Demand has caught up.
2) Level 3 has increased its sales force bigtime. They finally start selling now that they believe there is cash to be earned; no use selling stuff at a loss…. http://seekingalpha.com/article/187901-level-3-communications-inc-q4-2009-earnings-call-transcript
3) Google is buying dark (=unlit) fiber. Probably not because they think bandwidth will forever get cheaper. http://www.internetoutsider.com/2006/04/how_much_dark_f.html http://gigaom.com/2008/02/25/googlenet-update-google-buys-a-piece-of-transpacific-cable/ http://news.cnet.com/Google-wants-dark-fiber/2100-1034_3-5537392.html
4) Insiders are buying. http://www.gurufocus.com/InsiderBuy.php?symbol=LVLT
Is there something to worry about ?
- We are wrong about LVLT’s cost advantage
- A 4th competitor (besides LVLT, T and VZ) is able to create an end-to-end network.
- Slower than expected demand growth due to a reduced rate of broadband adoption, etc
Is there something I should let others worry about ?
- Google will kill them all….. No. Google looking for bandwidth is a good thing. They see a problem brewing.
- Wireless is the future, cables are yesterday….. LOL you need to carry all that local wireless data across the nation. Wireless solutions are not going to do that. Never, really ! Laws of nature prevent this.
And before I forget..... Sanborn Maps recently became a client of Level 3. Their data now runs along the track of Burlington Northern (remember those conduits). Sanborn used to ship their stuff to clients on disk drives. http://seekingalpha.com/article/187901-level-3-communications-inc-q4-2009-earnings-call-transcript
Any and all comments and questions welcome as usual.