Cogent Communications Holdings, Inc. (CCOI) Earnings Call Transcript & Summary

Welcome, everybody, to the next session at our Media, Internet and Telecom conference. I'm Matt Niknam, infrastructure analyst here at Deutsche Bank, and we are very pleased to be joined by Cogent Communications’ CEO and Founder, Dave Schaeffer. Dave, welcome back. Great to have you.

Well, thanks, Matt. As always thank you and Deutsche Bank for a great venue. It's not Palm Beach this year, but it's still a good venue. We'll be back there next year. And I'd like to thank all the investors on the call for their time and interest.

We can't wait to be back in Palm Beach. So yes, definitely looking forward to it. Maybe just to start, Dave, you guys recently reported 4Q results a couple of weeks back. Maybe if you can talk about some of the highlights from 2020 and lay out the top priorities you have for Cogent in 2021?

Yes. Well, thanks, again, Matt. Cogent is a simple business. In many ways, it has not changed in 20 years. But the market around us changes and probably the impact of COVID in 2020 had both positives and negatives for Cogent. In our corporate business, we saw a significant slowdown in corporate customers adding secondary locations and purchasing virtual private network services from us. Partially offsetting that was the fact that at primary locations corporate customers accelerated their purchase of 1 gigabit connections over our traditional 100 meg product, making it actually the 1 gigabit product, our largest seller. On our NetCentric customer base, it was primarily a series of tailwinds that came from the pandemic, we saw an acceleration in traffic growth and an acceleration in revenue growth, actually above long-term trend line. And as we look at 2021, I'm hopeful that we're nearing the end of the impact of the pandemic. I think people are anxious to get back to a normal routine back into the center cities, and we expect our corporate growth to accelerate. I think a lot of the IT departments are beginning to prepare for a return to office environment. And as we continue to see vaccination rates go up, I think many companies are expecting by the end of summer to at least have some employees back in offices, which I view as a very positive thing. On our NetCentric business, I think the acceleration in streaming will continue post pandemic. We still have a long way to go as the majority of videos don't distribute linearly, but increasingly with more streaming choices, more packages and more internationalization. We expect that business to continue to grow at trend line or slightly or it's above where it has grown over the past few years. So what we're clearly hoping for 2021 is a reacceleration in top line growth. Our multiyear guidance is to have roughly 10% top line growth and about 200 basis points a year of margin expansion. I think as we look at 2021, we'll get much closer to that revenue growth rate than we were last year. We were at about half of that growth rate. And we will also, I think, do better on margin expansion, even though last year, with a slower top line growth, we still delivered roughly 150 basis points of margin. So all in all, I think 2021 is a year of recovery. And for Cogent, we want to be in a position to serve our corporate customers and to continue to serve our NetCentric.

That's great. Maybe we can follow-up on the revenue growth. So you alluded to it. I think last year, obviously, about half of the long-term target, but the growth did decelerate, I believe, over the course of the year. And the exit rate growth was closer to the low single digits. And so I'm wondering what gives you confidence that, that 10% percent growth is achievable? And I guess, more specifically, what needs to happen from both a macro perspective and more company-specific within Cogent to sort of reaccelerate that growth rate?

So first of all, the growth rate is what Cogent has historically accomplished over the 16.5 years as a public company where we've grown at a compounded organic growth rate of about 10.3% year-over-year. So it's not an aspirational target that we have not met, but rather a recognition of what we've done in the past. For us to achieve that growth rate, really 3 things have to occur. First, we need to return to a normalized corporate environment. We saw a similar downtick in corporate growth in '08 and '09 and a great recession. The difference was that downturn only lasted a couple of quarters. This time, the downturn is much broader and is probably going to end up lasting 5, 6 quarters before we return to normal. Cogent's corporate business has grown at a historic growth rate of about 11.5%, again, all organically. We think that will resume as companies refine their new office environment because our corporate product is sold on a fixed connection basis. Even if some of those employees are working remotely, they will still be Cogent customers as long as they have a physical office at that location. Secondly, we may get some additional benefit from corporate customers, who actually reduce their real estate requirements, allowing us to grow our addressable market in our footprint. And third, for our corporate customers, we will see a resumption of a VPN replacement cycle using either SD-WAN and/or VPLS to replace the MPLS legacy networks that are rigid, expensive and antiquate. The second major point I want to touch on is our NetCentric growth. There, it is much more a volume-based business. We sell to about 7,300 access networks around the world as well as selling to approximately 5,000 content publishers. In that market space, we are selling a fungible, undifferentiated commodity and winning business based on price. That business has averaged a 9% growth rate, but actually before the pandemic had actually slowed down to a growth rate of only about 2% year-over-year. We are now starting to see a reacceleration of that growth driven in part by the acceleration in people streaming versus consuming video linearly. We think those trends will continue, and that business should be able for the next decade to be able to deliver growth rates in line with historic average. So if our -- 2/3 of our revenue are growing at 11%, 1/3 of our revenue at 9%, we'll be right within our targeted growth rate. I think the third point is where our office is going to reopen for the corporate customer. And I think that is really a tale of 3 different fact pattern. Primary offices where our relationship begins, I think we'll remain intact and continue to stay in place even if a portion of the workforce works from home. The second type of office are remote offices within the same metropolitan market. I tend to think that small set of those remote offices may end up shrinking. We may see companies eliminate 2 to 3 offices in the same metropolitan market in order to allow people to telecommute partially and then come into the main office on a more limited basis. And then finally, for those of in remote cities, I think they will remain in place. There's no substitute for having a physical presence in a remote market. Now those offices will require a combination of dedicated Internet and VPN services. So when I look at the kind of changing landscape of work, I feel Cogent is very well positioned to benefit from those brands. And then on the NetCentric side, the trends that were accelerated in the pandemic have been long-term trends and will continue for a long time. So I'm not expecting us to grow better than we have in the past, but rather in line with historical average, about twice as well as we did last year.

Got it. And when we think about -- maybe let's drill down into the corporate segment. And you had mentioned growth had been north of 10%. And obviously, we've seen that moderate and slow as well. But when we think about the reopening and we think about some of the commentary, I think you conveyed on the last call, how should investors think about the trajectory for growth in the corporate segment, at least near term, given that this is, as you mentioned, going to take 5 to 6 quarters for us to sort of get back out of this slowdown? And I guess more specifically, what sort of visibility do you have towards the sequential improvements that you talked about in corporate segment revenue in the first quarter?

Yes. So I think the first point is we're 4 quarters into the pandemic. So I think we're 2/3 to 3/4 of the way growth. Secondly, we have seen a number of leading indicators improve. Those indicators are number of phone calls connected each day by the corporate sales force, number of emails, the number of new opportunities created, the number of opportunities moving through the sales funnel. All of those are leading indicators that will show that corporate activity probably troughed at the end of August or in September, and has been sequentially improving. On a kind of headline basis, you see that in part and the improvement in our sales force productivity, while still below long-term averages, better than we were doing earlier in the pandemic. And I think companies are reevaluating what their IT infrastructure will be and that's a net positive for us. There are really 3 windows when a customer is right to be sold, when they move into a building, when their current provider fails or when they change their IT infrastructure. And I think the pandemic has caused all IT departments to reevaluate their infrastructure and figure out what does that new IT infrastructure look like, and it probably will contain more cloud-based services, more SaaS and probably some portion of the employee base working from home. And if those things are true, those will be net positives for Cogent. They will require those companies to have more reliable and greater Internet connectivity. And for our Internet service, one of the key differentiators is that our service is completely symmetric, meaning up and down are equivalent. And that really gives us a leg up on our competition. So when I look at all of these very specific sales leading indicators, it appears our corporate business, which normally rose at about 2% plus a quarter and declined last quarter at about 2% sequentially, has troughed.

Got it. And so maybe I want to go into the sales cycle comments as well because I think over the last year, you have talked about lengthening sales cycles. Customers a little bit more hesitant. And so is it fair to assume based on that commentary that maybe that's changed or improved for the better at all in recent months?

Yes. I think 3 things there. One, at the very beginning of the pandemic, install cycles widened, but those have since collapsed and are back down at our historical norms, which is to deliver on-net services at an average of about 9 days to deliver off-net services depending on part of which loop vendor, but in about 60 days, both well within our service level agreement bargains. The second thing that's changed is companies at the beginning of the pandemic were really paralyzed. They did not know what they were going to do, what the end looked like and what their infrastructure was going to look like. The only thing that was happening early on in the pandemic was this massive shift from FastE internet access or 100 meg connections to gigabit. While that continues, there's now kind of a second thing that's going on, which is companies are really evaluating what their new infrastructure need is. And as a result of that, they're having more conversations and the sales cycles are shortening. I think the third factor is that as people physically go back to the office, they -- IT departments really want to make sure that bandwidth is not going to limit those employees effectively. And I think that's a additional contributor to accelerating sales cycles of -- we close about 40% of the proposals that we write. I actually think with this return to office trend, we're starting those close rates improve slightly.

And so you referenced also the sort of upward migration to the 1 gig products in recent quarters. And I believe now it's taken a greater share of connections and revenue. And so I'm wondering how meaningful of a tailwind can this be to top line growth? And do you sense this sort of upward migration and sort of some of the tailwinds we talked about from a macro perspective could offset some of the pressure you've seen and may continue to see from lower off-net sales?

I do think they're roughly able to offset one another. The off-net portion of our corporate business are the smaller, more diffuse, branch offices and those are probably the last offices that will be repopulated. So I think we'll see some drag in that corporate off-net footprint, coupled with the fact that off-net has a more significant ARPU to go on than on-net because our loop costs have actually been coming down or those off-net locations, coupled with contract terms lengthening. Of the $200 a month uptick from gigabit sales is more than sufficient to offset that, and it's why we feel that our corporate aggregate growth rate even with these branch office headwinds from off-net should be able to return to 10%, 11% growth. The mix shift that will come from a greater percentage of corporate revenues being on-net versus off, also has a tailwind effect on our ability to expand margin. And it's part of the reason why our gross margins even last year were so exemplatory, even though the revenue growth was about half of trend line.

So I'm going to pause here. I just should have mentioned upfront. If anybody on the webcast wants to ask a question, you can just type it into the portal. I'll see it. I'll relay it over on the live Q&A. So feel free to do so, if you'd like. And Dave, maybe while we're on the topic of corporate. One other one, I guess, bigger picture from the corporate business is, can you talk about the addressable market for that segment? Where are you today in terms of market share? And what gives you the confidence that there's still meaningful room ahead for growth and increased penetration and share gain from Cogent within that domain?

Yes. So if we look at our average corporate building, we have 1,792 buildings at the end of 2020, we had 976 million square feet, which represents about 11% of all office space in North America. There are 51 discrete tenants in those buildings. We have approximately 25% market share measured by tenants. However, there are also more incorporated businesses than there are tenants. Sometimes, there could be 2 companies or even 3 companies cohabitating in an office. By that measure, we're more like 15% penetrated. As companies continue to need more connectivity to support those SaaS, remote workers and cloud-based services, that becomes the catalyst for us to continue to sell. We think, it is very reasonable that over time, in our footprint, we can reach a roughly 50% penetration rate. So our corporate on-net business should more than double from where it is today by increasing penetration and modestly at a 2% to 3% per year expansion of the footprint. The buildings that we go into tend to be the most expensive in a given market. They tend to have the most credit worthy tenants, and they tend in times of economic stress to lower rents but remain heavily occupied and most of the vacancies end up in the suburban campus environment.

Got it. Let's pivot to the NetCentric business. So this business, I think it's now about 35% of your revenues. Obviously, a bigger beneficiary of recent events, accelerating traffic growth. As you mentioned, recently returned to double-digit top line growth from a revenue perspective. And so from a high level, can you talk about your expectations for the NetCentric business in '21? And how we should think about revenue growth trending alongside what we've seen in terms of a recent acceleration in terms of traffic growth?

Yes. So our NetCentric business is almost exclusively volume based. The key application that's driving volume growth remains streaming video. And we're seeing 3 things that benefit that business: One, more minutes a day being streamed; two, the number of streamers is diversified, helping us achieve a higher effective price per megabit; and third, an acceleration and internationalization of that streaming business. Like many tech trends, they begin in the United States and are then exported around the world. I think streaming is a great example of that. And we are seeing services like Disney+, Peacock, Hulu, Amazon Prime, all capturing user share. And while Netflix remains the dominant player in the market, there is more diversity and more choice, and many subscribers are electing to take more than one package. We view all of these as positives for us. I often get asked a question, how can I predict traffic? And the answer is, I really can't, in a sense that it's totally usage based. But over time, we have continued to garner market share because of our aggressive pricing model to capture a disproportionate share of bits, and therefore, just proportionate share of revenue growth. When we look at the underperformance of the NetCentric business, it was really attributable to 3 discrete causes: the loss of Megaupload; port congestion by companies violating net neutrality; and then finally, customer growth coming from a limited number of customers. Those 3 onetime events are behind us, and we think returning to NetCentric growth, barring some kind of exogenous tailwind or headwind, should allow us to grow at our historic average. And in fact, we overachieved that average last quarter, growing 15.3% year-over-year and FX-adjusted 10.9%.

And so when we sort of pull that apart and think about the growth, one of the interesting things you mentioned was the internationalization of the Internet. So I don't know if there's any way to do this, but can you help us think about traffic and revenue growth trends you're seeing between the U.S. and international markets and how those 2 compare in terms of driving this reacceleration we're seeing in terms of top line growth in NetCentric?

Yes, sure. So the Internet was very much a U.S. phenomenon when it got started. 85% of global Internet traffic was in the U.S. when Cogent got started in year 2000. Today, that's only about 30% of global traffic. So there are approximately 4.9 billion end users globally of the Internet, there are 7.4 billion people in the world. So there is still room for 50% more growth in subscribers. Now clearly, there'll be some people too old, too young, but for the most part, we've still got probably another 5 to 10 years of subscriber growth ahead of us. But then within that, what we are seeing is a change in how people use the Internet internationally. Most of the developed world, such U.S., Canada, Japan, Western Europe, has been predominantly fixed line. There are 900 million fixed line subscribers. But with the migration to either LTE or 5G in the developing world, those 4 billion wireless-only subscribers are seeing their unit volumes go up rapidly. And while they may not get the resolution to watch video on a large 75-inch flat screen, they can do a lot on their handheld device and most of the video consumption in the developing world is mobile. Now mobile has got a very long way to go. Globally, mobile only accounts for about 3% of global traffic, yet it accounts for 80% of the user. Mobile's growth rate is over double that of fixed line. So all of that is driving increased penetration of streaming video, and with Cogent having more access network in more countries as its customers than anyone else in the world, we're disproportionally a beneficiary of that. And related to that, in the past, most of our traffic was a one-sided customer payment. Two years ago, half of our traffic, we got paid by one side, and then we got paid on both sides in 50% of the cases and 50% of the cases we peered that second side of traffic with no revenue. Last year, in 2020, we were up to 67% of the cases where Cogent was getting paid by both a sender and a receiver. This is a net positive for our business and should continue going forward.

Okay. And so let's now pivot. You had mentioned sales force, and I want to dig into this a little bit because I noticed productivity per sales rep, which had been trending downward to start 2020 tick back up. We got back to about 4.2 in this past quarter. And so can you help us think about the drivers over here? And I guess, in particular, can you talk a little bit about sales force turnover as well? Because I did notice that picked back up in the fourth quarter. So maybe we can touch on that, and I've got a couple of follow-ups. Maybe we can start there.

Yes. So when the pandemic hit, Cogent had to quickly pivot from sales offices around the world where our 750 sales professionals, 600 quota-bearing reps went every day to work from home. That transition was challenging. We had to get everyone laptops. We had to get them cellphones, and they had to begin to do their outbound telemarketing in a remote environment. We did very well at that. The second thing we had to do is start hiring new reps because we do turn about 5% sales per month. This is an outbound telesales model which requires high activity and not everyone is successful at that. In the first part of the pandemic, we did a good job of hiring decent job of managing and a poor job of managing out underperformers. We became more aggressive in the third quarter and fourth quarter of managing out those underperforming reps. That will continue. We also think that our sales force will be able to return to an office environment hopefully by the end of summer, early fall, and there is a benefit to having managers and colleagues directly in the same office as opposed to working remotely, but also makes training easier. So we feel pretty good about, one, the underlining demand backdrop; and then two, our ability to have sales -- enough salespeople to go after that market; and three, those salespeople to increase their productivity. So we did get a -- we understood that it was in large part because we were carrying some underperforming reps, and we manage them out. So if we step back and look at 2020, we grew our entire sales force 4% in a year. We grew our full-time equivalents at 8%. Our stated long-term goal is to grow the sales force at between 7% and 10%. So we did reasonably well against that goal. Even though we had to hire remotely and train remotely as we can get back to a physical hiring and a physical training environment, we think that's an added benefit to getting us back to our long-term average productivity of about 5.3% or below that today per rep, per month of installed orders and to get our sales force turnover in line with long-term turnover rates of just over 5% per month.

And so what inning are we in right now in terms of this managing the underperforms process because I assume the way I understand that it is, there were maybe newer reps brought in, and I guess, tougher to sort of manage them out if they're underperforming. And so these are shorter-tenured reps, the way I understand it, that are being managed out. But I'm trying to better understand how -- I guess, what inning are we in, in terms of this process?

Yes. I think just like with the pandemic, we're probably in the seventh inning at this point. We have a pretty good process in place for hiring, training, managing and managing out underperformers, but it's still all remote. The real benefit comes when we can get those people back into the office.

Got it. Got it. Let's talk a little bit about margins. So I think you had mentioned, obviously, the longer-term target of about 200 bps year-on-year margin expansion. I think you did pretty close to that. But my numbers, it was about 130 basis points of expansion last year despite growth being less than half of the long-term target. And so as I think about an improving or inflexing trend for the corporate side, more of a mix towards on-net versus off-net, at least in the interim, and we think about the continued strength in NetCentric, which is primarily on-net as well. What is the sort of expectation, if any, in terms of margins? Because I would think that sets you up pretty positively to maybe accelerate the rate of improvements in adjusted EBITDA margins in '21. So I'm just curious to get your thoughts there.

Yes. Again, our guidance, both in terms of revenue growth and margin expansion is not meant to be this year or rather multiyear in nature. I think that we are in a good place. We saw our gross margins expand by 100-plus percent basis points. And really, we should get the SG&A uplift when those reps become more productive. We also have the tailwind of a greater percentage of sales being on-net versus off. So in our Corporate business, the mix is roughly 60% on, 40% off. And in NetCentric business, it's 93% on, 7% off. So we get the benefit of selling more NetCentric and more on-net Corporate, which carries much better contribution margin. Our contribution margins on-net are 100% gross margin contribution, $0.95 of EBITDA. And I think we are in a position to see revenue growth revert back to historic norms and with that achieve at least the type of margin expansion we've done, again, over the past 21 years where we have grown margins since going public 16.5 years ago at 210 basis points a year average.

And then -- so now let's pivot to capital allocation. I know it's a very popular topic with Cogent. And so maybe I'll start first with leverage. I'm just wondering, in the past, you've talked about targeting leverage between 2.5 to 3.5 turns. I believe last quarter, you were closer to the upper end of that at about 3.4 turns. And so I'm thinking with leverage where it is, with interest rates rising, how do you approach leverage today? And does that backdrop imply incremental shareholder returns could moderate until you see maybe more meaningful reacceleration in growth?

So a few points. One, most of our leverage increase came from the fact that the euro-denominated notes had to be marked to a higher value because of the appreciation of the euro against the dollar. That was a $19 million headwind in the quarter. In fact, when we refinanced our unsecured debt, we lowered the interest rate from 5.875 to 4.375, lowering our cash cost. Similarly, we have a callable [indiscernible] offering that's a 5.375 that we're evaluating refinancing and lowering the interest rate, which I think would have a meaningful impact. Secondly, we have excess cash on the balance sheet. We have far more cash than the company needs to operate, and our goal is to gradually disgorge that cash through a combination of buybacks and dividends. Last quarter, we did supplement our dividend with $4.2 million of buyback. In total since 2006, Cogent has returned $895 million to shareholders, roughly $230 million of that has been through the buyback program and about $665 million through our dividend program. We have 34 consecutive sequential quarters of growing the dividend sequentially. Our dividend growth rate last quarter was 14.4%. Historically, since starting the dividend in 2012, we have grown the dividend at a compounded rate of about 16%. We have also during that period grown free cash flow at a similar rate. We feel very comfortable that we're going to be able to return increasing amounts of capital to our equity hold.

And adjusting for that, you mentioned the euro-denominated notes had to be marked higher, and that drove about $19 million. So where would you say sort of normalized leverage is right now adjusting for that? And is there sort of a comfort level in terms of where you would like to be, particularly with the uptick in interest rates of late?

Yes. So 2 points. First of all, that was just a 1 quarter movement. There was movement to prior quarters as well. So in aggregate, we've had almost $80 million of movement in the euro against the dollar to impact our leverage. We had initially targeted a range of 2.5x. When we became close to that, we raised that range to 2.5 to 3.5x EBITDA. We know that we have a very durable, existing revenue stream and a consistent growth rate. Let's put this in perspective. We grew just under 5% in a world where GDP was contracting at probably 5% or 6%. So I think we have been very durable, and when compared to other wireline companies, even more durable. So I think we had the ability to take on more leverage, if needed. To your comment on interest rates, while the 10-year has moved up over the past 3 months, high yield spreads have actually compressed some. So the high-yield index remains near historic record. Still, in absolute dollars, if we did something, we'd be lowering interest rates. And it's not clear to me with the certain efforts globally in a synchronized manner of all the central banks that we are going to see any kind of permanent increase in interest rates in the short to medium term. I think that there is a lot of stimulus going into the economy. The economy is benefiting from the vaccination and the optimism of reopening, but there's still a lot of economic damage that was caused by the pandemic that has not yet been fully dealt with.

Understood. And if we tie this all together, this is going to be my last question. What do you think is most misunderstood by The Street about the Cogent story? And where do you see the potential for outside surprise over the next 6 to 12 months?

So having done this for a long time, I've become a bit dated because every business over the long run to be measured by a simple, consistent yardstick, which is DCF per share and the growth rate and that discounted free cash flow per share. I think Cogent is misunderstood because many investors look at us on EBITDA multiple, and may say, "Oh, Cogent, well, it's expensive." But when they look at us on a growth adjusted free cash flow multiple, we actually look very inexpensive. So Cogent is, I think, misunderstood because it's lumped in with a bunch of wireline telecom companies that sell legacy products, have negative revenue growth and negative operating. Again, throughout the pandemic, and even before, we continued extraordinary top line growth and exemplary margin expansion. So when you put that together, we're really not a wireline telecom company. We look at some of the other companies in your coverage in reverse more in the data center space or tower space, much more capital-intensive than Cogent and lower organic growth rates yet trade at substantially higher valuation. We just got lumped into a neighborhood that's not a good neighborhood. And the only thing we can do is run our business. Guys like you and your colleagues that pick what peer universe you compare us to.

Understood. And I think with that, we will end it there. We're just about out of time. So Dave, on behalf of myself, everyone at DB, thank you for joining us, and we're really looking forward to doing this in person in Palm Beach next year.

All right. Thanks a lot, Matt. Thanks, everyone, for taking time. We'll talk soon. Bye-bye.

Thanks. Bye.