Cogent Communications Group's CEO Presents at Deutsche Bank's DbAccess 21st Annual Media and Telecom Conference (Transcript)

Mar. 4.13 | About: Cogent Communications (CCOI)

Cogent Communications Group, Inc. (NASDAQ:CCOI)

Deutsche Bank's DbAccess 21st Annual Media and Telecom Conference

March 04, 2013 10:50 am ET

Executives

David Schaeffer - Founder, Chairman, Chief Executive Officer and President

Unknown Analyst

Yes, everybody, will -- please take your seat. We're going to get started here with our next session of the morning.

Welcoming back to the conference this year is David Schaeffer. He is the CEO of Cogent Communications. What we're going to do today is Dave's going to start off with just a brief overview of the company for those of you who are new to the story, and then we're going to go back into our usual Q&A format.

Dave, they should be in here. Go, go ahead.

David Schaeffer

Okay. Thank you, Brett [ph]. I'd like to thank Deutsche Bank for hosting us and all of you for taking the time to hear a little about the company. I see a lot of familiar faces in the room, so I'm going to try to keep this brief and just give you a quick overview. Safe Harbor language.

Let me describe what we do at Cogent. We are a provider of dedicated Internet access service. We carry about 18% of the world's traffic across our network. It's a physical network that goes from Kharkiv in the Ukraine, actually all the way around world to Tokyo, from Helsinki, Finland to Mexico City. Approximately 56,000 miles of terrestrial fiber, long-haul connected to an additional 26,000 miles of metropolitan fiber. We serve just under 35,000 customers on that network.

Our customers are divided into 2 groups. 51% of our revenues come from corporate customers. These are companies that are typically in North America, located in large multitenant office buildings. We have fiber into the building, up the riser all the way to the customer suite. And that customer will typically buy a 100-megabit non-block, non-oversubscribed symmetric Internet connection for a flat price of $700 a month. 49% of our revenues come from connecting to 650 data centers, and in those data centers selling to about 16,000 NetCentric or service provider customers. In that market segment, what we are selling is bulk Internet connectivity. It is sold on a per-megabit basis and we are the price leader in the market. We have a network where we're utilizing about 20% of the lit capacity in our network. We understand that we are a single-product company. We sell nothing but dedicated Internet access services. And we've been able to organically grow the business top line at about 15% for 31 out of 32 quarters consecutively, organically. In the most recent quarter, we grew sequentially 3.7% quarter-over-quarter, and with this growth comes substantial margin expansion.

We have been able to deliver approximately 220 basis points a year of EBITDA margin expansion. On a going-forward basis, we expect to continue to grow the company between 10% and 20% top line. Each of our 2 business units, corporate and NetCentric, will grow at about the same pace. We operate at 36 countries, 180 markets. We operate the world's largest fiber network, and we are the second largest carrier of Internet traffic in the world. We are the network that is the most interconnected. The Internet is a network of networks, more networks connect directly to Cogent than any other network around world. And the result of this is we've been able to generate free cash flow consistently from operations and use that free cash flow to buy back about 15% of our outstanding flow, 55% of our convertible debt, and we've implemented a dividend program and have been able to raise our dividend each quarter for the past 3 quarters.

With that, let me go sit next to Brett [ph] and see if I can get some tough questions here.

Question-and-Answer Session

Unknown Analyst

Great, and thanks for that overview. So let's start up with some of the guidance that you've given. You have sort of a general objective that you reiterated for a period of time, 10% to 20% annual revenue growth along with the targeted expansion in your margins. I want to try to understand what makes you so confident that 10% to 20% is the right goal because you've articulated not just as something you think you're going to accomplish in '13. It's been kind of the running objective, and I think you've sort of alluded to the fact that for the foreseeable future you feel like that's the right zip code for investors to have expectations. You have 2 businesses, you have corporate, you have NetCentric. They each make a contribution. Can you walk us through what you're seeing in terms of the demand for the services you sell that gives us confidence that that's the right objective and an achievable objective for Cogent?

David Schaeffer

Fair enough, Brett [ph]. So first of all, we're little different than your business. Past performance is a good indicator of future performance. And in fact, we have been able to demonstrate about 14.5% top line growth now for 8 years as a public company. There's clearly some volatility based on product mix, customer mix and some foreign exchange, but that's been the long-term trend. Let me now disaggregate our addressable markets and talk about penetration in each of those markets and how we increase our penetration. On the corporate on-net portion of our business, which is approximately 1/3 of our total revenue, we today have 1,303 multi-tenant office buildings connected to our network. These buildings total 712 million square feet or about 11% of all rentable office space in North America. Within those buildings, we have 10.5 customers per building out of an addressable market of 51 customers on average. We've been adding approximately 1.8 additional customers per building per year and adding about 60 or 70 buildings to the network. We expect that corporate footprint expansion to actually slow, but expect our penetration to continue to increase at, at least add 1.8 additional customers per building per year. What is pushing those customers to use us as opposed to the copper-based service is the quality of our service. So a common misconception of Cogent is that we are purely a price leader. On the corporate on-net portion of our business, the price per connection is identical to what you would pay AT&T or Verizon, who are typically our competitors, but you're getting a fiber not copper network, ring-protected, not point-to-point, Ethernet, not TDM, and you're getting 65x as much bandwidth for the same price. We also have a corporate off-net business. The addressable market there is virtually infinite in that there are over 5 million potential businesses that we can serve in North America. But the challenge is, unless it's tied to our on-net business, the cost of revenue acquisition is too high. We typically sell to someone an on-net circuit. And then if they have a branch office, where we will sell off-net. Our off-net business has actually experienced about a 17% top line growth over the past 8 years, and that slightly higher growth rate has come from the fact that much of our TDM off-net base has switched to Ethernet. So the most common product off-net is actually a 10 MB Ethernet circuit for actually about $2,000 a month. So 3x the cost, 1/10 the bandwidth. But that's the cost of doing business in one of the suburban campus buildings where we cannot generate an adequate return on capital to bring that building on-net. Now the third portion of our business, which is the biggest single chunk of revenue, is the NetCentric on-net business. It is 49% of our total revenues. It is a global business. 40% of that revenue is outside the U.S., 60% is North America, U.S. and Canada. In that business, we are selling a bulk commodity in a location that is effectively a supermarket for bandwidth. So what we are selling is Internet connectivity to other service providers, whether they be content companies who are pushing applications or content to the Internet or regional access networks pulling. In that market segment, the market is dominated by unit volume growth and price declines. We've actually been at the forefront of leading those price declines. We are the price leader. We actually guarantee potential customers that we will undercut their current provider by 50%. We are in fact selling there purely on price. We're selling a commodity in a location that's effectively a supermarket for bandwidth. Today, we connect to 650 data centers around the world. Those data centers are located in about 565 unique buildings. Within those facilities, we generally have 24 customers per data center. We have approximately 20% of the bandwidth consumed and about 12% of the revenue spent on Internet connectivity. For Cogent, we have been able to demonstrate we can grow our unit volume that is megabits sold per month at twice the rate of the market. In the most recent quarter, we actually saw our rate of market share gain accelerate. We grew our bit volume 27% sequentially in the quarter. Our rate of price decline last year was 16.6%. That's below the industry average of 40%. Our long-term rate of price decline over the past 5 years at Cogent has been about 22% per year. The net result of these volume increases and price declines resulted in our revenue in the NetCentric portion of our business in Q4 sequentially over Q3, growing at 6.3%. Some of that was foreign exchange benefit. If we netted that out, we still grew that business at 5.1% sequentially. We feel very comfortable that the end-user demand for bits continues to grow. Our NetCentric business is really a derivative of residential consumers. While Cogent itself has no consumer business, all of our NetCentric customers, in fact, are selling to consumers. So on the content side our customers include Apple and eBay, Yahoo! and Microsoft, Netflix or Google. On the access network side, our customers include ComCast or Charter or Cablevision or Cox. It also includes international phone companies like Bell Canada, BT, all the way down to in the news this week, Telecom the Vatican down to Telekom Malaysia, Telkom South Africa, Telecom Vietnam. We sell to international cable companies, state and local governments that distribute our bandwidth. All of those users, both on the content side and the access side, are being driven by the adoption of over-the-top video. Today, about 75% of global Internet traffic is video, about 82% or 83% of Cogent's traffic is video and virtually 95% of the unit volume growth is over-the-top video. That's what gives us confidence that our growth will continue.

Unknown Analyst

So let's talk about the NetCentric side of it. We hear launched coverage on a series of data center operators over the last year, and one of the trends that we analyzed in that process was the growth in global IP traffic which, to your point, is very much driven by residential video. You talked about these supermarkets that you are in, these are data centers, right? That's what you're NetCentric business is effectively attaching into our data centers.

David Schaeffer

Correct.

Unknown Analyst

So in a sense, you're sort of saying that underlying this is all the same trend. Data centers are being developed for the purpose of helping to serve some of these needs, and you guys are finding connectivity into these facilities. So there's a connection to all this, I guess?

David Schaeffer

Yes. So first of all, the data center exists so people can put equipment, servers in that facility by space and power in order to access the public Internet. The average data center will have 12 global networks connected to it. We're one of those 12 networks. We, however, are the network that is the most interconnected network. We have equal or better reliability to our competitors. But most importantly, we're selling a megabit of connectivity to the public Internet at half or more of a discount to our competitors. The challenge for our industry has been that the dollar value of that aggregate bandwidth purchases within those supermarkets for bandwidth has been flat at $1.5 billion for about a dozen years. So as capital flows in and more data centers are built, we're seeing that $1.5 billion market be diffused across more locations. We at Cogent, as part of our long-term maintenance CapEx, expect to always build to those facilities. Our revenue growth in that NetCentric business is coming solely from market share gain. We are not projecting a growth in the dollar value of the market even though the market in unit volumes is growing at just under 30% per year.

Unknown Analyst

So a couple of follow-up questions on that. First of all, do you just automatically go into any data center that's being built, or do you wait until you have some signed business?

David Schaeffer

So in evaluating a new data center, we look at 3 things. First and foremost is it truly carrier-neutral? Will they allow us to sell our bandwidth? So we don't go into a Sprint or an AT&T data center because they really don't want us there. Two, we look at the size and diversity of the customer mix either in the center or, if it's brand-new center, how large it is. And we do look at the reputation of the operator. We are fairly indiscriminate about data centers and feel that it is almost necessary that we're in virtually all centers where there is addressable market. So out of our footprint globally in those 36 countries, we're in about 650 data centers. There's about 1,100 data centers in total, many of the difference, about 400 of the differential, is proprietary data centers where they are not carrier-neutral. They could either be an operator who then connects to a carrier-neutral. An example there would be Google or Facebook that build around data center, and then they build fiber back to a major aggregation point and then we then sell bandwidth to them at those types of locations, or it could be a data center that is just too small or the operator is too financially unstable. But for the most part, we make sure that we're relevant to almost the entire addressable market.

Unknown Analyst

And why is it that you have been able to sustain a price advantage over your competitors? You acknowledge it's a commodity, why are you able to deliver service at half the retail price that they can at these facilities?

David Schaeffer

Well, a common misconception about Cogent was that our advantage came only from our balance sheet. We did purchase a large number of distressed assets at a deep discount to the market price when the telecom market melted down. But if that was our only advantage, once those assets depreciate, we would have no further advantage. Secondly, if it was only an asset advantage, we would not generate EBITDA margins that are at least 1,000 basis points better than our competitors, and our competitors, on aggregate revenue, are anywhere from 20x to 100x our aggregate size. So we clearly have a different operating cost structure. Our advantages are multifaceted. One, it's the way we purchase the fiber, both in a distressed situation plus under IRU, meaning we have just a pair of fibers, we're not carrying excess inventory. Two, the optronics that we have placed on our fiber allow us a very different cost structure than someone who is superimposing the Internet on a telephone network. So our network is running IP directly over DWDM, protected at layer 3, terminating only in the highest traffic locations and using Ethernet. Everything about Cogent was designed to produce the lowest-cost interface routed bit mile. If you think about it, that's really what we sell. It's called Internet access. We then have a very simple product suite. We only have approximately 8 products compared to thousands that our competitors have. So that makes our back office, our billing, our provisioning, our customer care systems much simpler. We also are more efficient in where we deploy incremental capital going after only these very high traffic locations. We then have a very different sales model. We understand that what we sell is a commodity. We don't try to protect legacy services and products. If you look at telecommunications broadly, the whole industry, every company at this conference, they're under tremendous top line pressure due to the deflation of the Internet. The Internet is cannibalizing all other goods and services. We don't have that cannibalization problem, so we can lead on price. Cogent's cost of revenue acquisition, its sales and marketing costs, are 1/6 the industry average. We acquire $1 of revenue for 1/6 of what the industry acquires it at. That becomes then a virtuous circle that gives us greater volume which then further drives down our average cost because marginal cost is always below average cost. We have built a model that works for a $1.5 billion addressable market. The vertically integrated telephone companies we compete with have cost structures that don't compete and have products that are susceptible to this internal cannibalization. Those differences allow us to make money when those companies continue to struggle.

Unknown Analyst

All right, so then why you lose business?

David Schaeffer

Our biggest competitor is actually inertia. It's just, "I don't care. I've got a connection, it's fine today." And let's take each of our customer segments. We go on to a big high-rise building. Today we have 85 of the 100 largest law firms in the world, but we go to 1 of those 15 that we don't have, and they say, "Look, I've got AT&T or Verizon, it works well enough. I'm not going to get fired for having them. I've got a contract. I don't have to think about it. Yes, your product may be better, but why should I switch? Off-net. If I'm not tied to an on-net relationship, nothing's special about Cogent. You're just renting a loop from an incumbent. You're adding the IP layer. Interesting but a 10% cost savings, not compelling for me to switch." Now let's move to the NetCentric portion of the business. In the NetCentric portion of the business, "I just don't care. I use so little bandwidth for my application. It just doesn't matter to me." So upon someone running an eCommerce site and just putting up some static web pages, just as you use a lot of bandwidth. Maybe I'm selling diamonds and my aggregate turnover might be $500 million a year, my bandwidth costs are $100,000 a year. It's just irrelevant. Now ironically, we have the #1 diamond merchant on the Internet as a customer. But many eCommerce companies just don't use that much bandwidth. The second reason is there's still a dozen other competitors out there, and they're all large, vertically integrated household names who spend a lot of money on advertising. But we have come from 0% market share to the second largest carrier of Internet traffic in the world carrying 20% of the world's traffic with much less capital invested than any other company. And it's because we delivered the best value. Probably the most telling number is the number of networks that buy connectivity from us. So on the Internet, there have been 42,000 ASs or autonomous system numbers issued. Of those, about 6,000 of them account for 98% of the Internet's traffic; 4,400 of them connect to Cogent; 4,350 of them pay us for connectivity. So more buy than don't, but inertia there is a problem. And then overcoming the incumbency problem of, "I buy from Level 3, I buy from AT&T, I buy from these very large companies that have a long history." Here we are as an upstart, but we continue to grow and get more business than anyone.

Unknown Analyst

Right. So let's talk about the growth drivers on the corporate side of the business. And you've mentioned that you connect into 11% of the rentable office space in the U.S. Not every office building would be appealing to you, how many of the addressable buildings do you even feel you're connected into at this point in time? You've talked about CapEx coming down, so it seems that you're getting too closer to being where you want to be as opposed to further.

David Schaeffer

We are getting near saturation in that market. So in North America, there's approximately 1 million commercial office buildings. There's almost 5 million commercial structures, but a lot of those are gas stations and carwashes and restaurants. But I'm talking about office buildings, there's about 1 million, and there's about 1.4 million in Europe. We're only in 1,303. But the average building we connect to looks very different than the average building -- office building, in the Western world. Our average building is 41 storey’s tall, is 553,000 square feet and has 51 tenants. The average building has 1.2 tenants and is 6,400 square feet. So it's about 190th the size of what we connect to. We think there's probably about another 60 or 70 buildings that make ultimate success for Cogent to build into. We will continue to spend capital to get to those buildings and then stop. Again, this is an industry where investors have been burned by companies chasing growth at all costs, ignoring returns on capital. We have proven that we can be very disciplined. And what we need to then do is go after buildings where the $120,000 that we spend in the building gives us a rate of return once we layer in our sales and marketing cost that's substantially above our cost of capital. So we generally look at about a 20% return on capital hurdle rate. But then you have to look at the lease up costs or the fill rate time to get that building on-net. And once the building falls below 0.5 million square feet, where the tenant count goes much below 40, you need to dig a lot deeper to figure out whether it makes sense. So you are correct, Brett [ph], that we will slow our footprint expansion, not because we don't have capital. We have $250 million on our balance sheet. We generated excess free cash above our dividend, above our interest payments, and our cash flow is growing. But we're very disciplined about making sure that capital is efficiently deployed.

Unknown Analyst

I see. Let's talk about that. I mean, if we're a couple of years away from seeing the CapEx really diminish quite a bit and just based on what you outlined, and it's already in the process of becoming a less-capital intensive business. You feel like you've got visibility on 10% to 20% top line growth, 200 basis points incremental margins. It sounds like there's going to be a lot of cash generation in the business, if that's what your anticipation would be. What do you expect to do with that cash? Because you already have lot of cash on the balance sheet as well.

David Schaeffer

So ultimately, I am a firm believer that the purpose of any business is to produce cash for its stakeholders. In our case, most of our stakeholders are our equity holders. So we do anticipate a combination of top line growth, margin expansion and capital reduction. In fact, we've demonstrated now, not only as a percentage of revenue but in absolute terms, the ability to reduce our CapEx, including prepaid capital leases for each year for the past 4 years, and we expect that to continue going forward. We believe that with that excess cash, we will be able to both grow our dividend and, at the same time, opportunistically take advantage of market dislocations and buy back our equity. Since the company's inception, we spent about $180 million buying equity. In hindsight, we should have spent more because our average purchase price was at about $14 a share. We bought back 55% of our convert when that seemed to make the most sense. We also have implemented a dividend to allow investors to redeploy that capital as they see fit. But we anticipate substantial growth on our cash flow. If you look just at the last year, we generated enough cash to be able to pay $9.5 million in dividend. We generated an excess $9.5 million on the balance sheet. We paid $36 million of interest, of which $31 million was cash interest, and we bought about $1.5 million worth of stock. That was last year's snapshot. Obviously, 2013, based on our guidance, provided we hit those targets, will be substantially more free cash flow. And we are committed to returning that capital. Oftentimes, I get a question from investors, should we put more leverage on the business? And the answer is possibly to return capital at a more accelerated rate. Again, that's all relative to market opportunities. We have looked at opportunities to invest capital internally. We believe we're investing as fast as we prudently can in the business. We've also looked at external or kind of inorganic ways of buying businesses or revenues and, quite honestly, have not found things in our adjacent space even though we look broadly, that are cash flow accretive.

Unknown Analyst

I'm going to stop and see if we have any questions in the audience. And we actually have one here in the front. Just wait 1 second for the microphone.

Unknown Analyst

I have 3 quick ones. Why did -- why was the acceleration at 26% growth rate in the fourth quarter? What was the genesis of that acceleration? It was pretty dramatic. Second, 2012 was a year that you cycle through all the mega-disruption. And you said in your fourth quarter call that you've become the exclusive data provider for the new Mega. So maybe you could give us an update there. And then there's -- you didn't mention anything with regard to the CDN arrangement with Netflix, so maybe you could just give us a quick update on that.

David Schaeffer

Okay. Thanks for your questions, Eric. First of all, Q3 to Q4 and Q4 to Q1 seasonally tend to be good traffic growth quarters. However, this year, off of the largest base in the company's history, our Q3 to Q4 growth was the greatest in percentage terms sequentially, 27%. I think that was driven almost exclusively by over-the-top video. We think those trends will continue. And our ability to benefit from that is actually on both sides, both as the producers of that video publish it to the Internet and as the access networks consume it. In the case where the access network is also a Cogent customer, we get revenues on both sides. If they are a peer, we obviously only get it on one side. But I think the key driver far more than cloud, far more than enterprise, far more than any kind of legacy application is video and I think that will continue going forward. To your comment and question around Megaupload, we were their sole provider. They were a global operator and distributor of shared file video around the world. Independent third parties pegged Mega as accounting for 4% of global Internet traffic. They were exclusively a Cogent customer. On January 19, 2012, the U.S. government shut their site down, alleging copyright infringements. That resulted in a 5.5% loss in revenue, and in fact the only negative revenue quarter sequentially and growth in the company's history for Cogent. They relaunched on January 19, 2013, as a customer and have publicly stated that they once again are exclusively using Cogent. It is a new corporate entity. It is much smaller than it was previously, but we obviously hope they, as well as all of our customers succeed, providing what they do is within the bounds of laws in the countries in which they operate. We think that will continue going forward. To your third question around Netflix, they also have announced that they are relying on Cogent as a major provider of bandwidth. They are migrating away from their mail-order business to more of a streaming model. Previously, they had used third-party CDNs. They are building their own CDN, and we are the primary bandwidth supplier to that CDN. Again, just like with any of our customers, we wish them the greatest success, and we hope their model works. Right now, they're doing well. We hope that continues.

Unknown Analyst

Great. Well, we've just about running out of time, so thanks a lot for doing this.

David Schaeffer

Thanks for having me, Brett [ph].

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!