Cogent is a very unique company. Founded by Dave Schaffer back in the late 90's, he had private equity capital backing him with around $600mm and picked up the pieces after the dot.com fallout. He went on a buying spree, diluted his ownership level in his stock, and then proceded on a 18 year binge of growing organically until his core businesses had 25% of global internet traffic going through his network. He essentially stuck a-load of assets togtether, kept things simple, undercut the competition in price and time, and wala, he compounded top line at 10%, ebitda around 11-12%, and bought back 20% of his shares and has grown the dividend for around 48 consecutive quarters.
A few years ago CCOI did there first deal in 18 years when they bought the old Sprint network from TMUS in a deal where they handed over just $1 and TMUS was to provide staggered payments over the course of a few years of $700mm. This was because the assets lost money hand over fist and contracts would need be wound down. The income statement reads like a dogs breakfast, but Dave is telling us the old Sprint contracts are peeling away one by one, and we should see top line revenue growth on an annualised basis from now. He anticipates 5-7% top line growth + 6-8% ebitda growth from here.
But this is where the story gets interesting, as with the Sprint deal they got a $20bn sunk cost fiber network (built in 80s and early 90s), data centers and IPV4 addresses. They are repurposing the sprint network to enter the $2bn US wavelength market where they anticipate taking 25% market share by 2028 ($500mm rev at 95% gross margins). Essentially the same people buying network transit and VPN from Cogent will also be buying wavelengths, so Dave has the salesforce (albeit with a high turnover) and the existing customer relationsips. Its just selling to them, hooking them up, and proving themselves to be worthy operators. So far they have been behind schedule in getting these waves sold, but Dave hasnt changed his forward guidance and anticipates 25% by 2028 time. Time will tell. Dave says they will take market share the same way they did with transit. By undercutting and being the low cost provider, the quickest provisioning times, and also offering a higher quality service. Reddit posts have said that Cogent is the budget brand for transit. It seems Zayo and Lumen may be higher quality, but also charge more due to higher cost structures. Dave having a cost of $1 for his network has a serious competitive advantage over the incumbents. He also has a much more streamlined and focused business.
Cogent also has 40-50mm IPV4 addresses which they are already monetising via secured lending. Daves thoughts on this business is that he build up the revenue or sell large blocks to willing buyers. He thinks this is a $1bn asset eventually. Given the financing so far, its hard to argue with.
Also with the Sprint deal they picked up so many data centers which are currently out to market and they are signing LOIs with buyers. Dave thinks the data centers could be a 100mm ebitda business or fetch 700-1000mm in a sale. Zero cost basis with this, so any sale will be subject to tax.
And with the fiber network they also have a ton of dark fiber which Dave thinks could fetch 1bn. Unsure on this detail and how many willing buyers there are today and its a point which doesnt get talked about alot. The main bits are currently; transit/vpn, wavelength, data center and IPV4.
So essentially, you have a 1.8bn business which could have run-rate ebitda of >700mm, around 120mm in sus-capex, and maybe a further 2-3bn in asset sales. The cash of which will most likely be returned via special dividends and/or buybacks.
I plan on buying a decent amount of CCOI after TNZ re-rates from its inevitable next deal it does