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

All right. Perfect. Thank you, everyone, for joining us for the Crédit Suisse Communications Conference. I'm Sami Badri with Crédit Suisse, focusing on communications infrastructure and telecom and network equipment. Right now, we have the CEO of Cogent Communications, Dave Schaeffer. Thank you, Dave, for joining us today.

Sami, thank you for hosting us. I'd like to thank investors for taking time out of their busy schedule in a volatile market to pay attention at Cogent. And as always, I'd like to thank Crédit Suisse for a great venue.

Absolutely. Thank you. So Dave, I wanted to kick it off with a question on your Q1 2022 results. And mainly trying to talk about the improvement that we saw and specifically the corporate segment. Revenues were flat, excluding some parts. But I kind of wanted to dig into what you're seeing on the corporate side and where you kind of see the inflection going from here, at least from a business perspective?

Yes. So I think we're still a couple to maybe a few quarters away from returning to our long-term average corporate growth rate sequentially of between 2% and 2.5%. In the pandemic, our corporate customers suffered the most. Our corporate growth rate fell from 2.5% positive to 2.5% negative. In the first quarter of '22, we returned to effectively a flat sequential growth rate, a significant improvement. We were down one tenth of 1% when adjusting for universal service fund fees. We are seeing sales activities to corporate customers improve. We're seeing the number of spoke to's continue to increase as they had in the previous quarters, number of proposals issued and the duration of proposal issuance to contract signing continuing to shrink. We think all of these factors, coupled with the anecdotal discussion with customers that many of them are ready to now make decisions about their corporate architectures that they've delayed for the past 2 years. These decisions include increasing the aggregation points for work-from-home employees oftentimes growing those ports and adding additional aggregation points to shrinking their office footprint and eliminating redundant or not required offices, shrinking the size of each office and supporting a workforce that is anticipated to be between 30% and 40% remote on any given day. These are all factors that have led to our reacceleration in our corporate business and should continue for multiple years and allow us to return to that kind of pre-pandemic growth rate.

Got you. One area specifically is there's been a focus on vacancies in some of the major markets, at least in North America -- sorry, that's the next question I wanted to ask. Actually, that's right. Sorry, my eyes are skipping just because I noted in between my questions. One thing I wanted to really touch on is what trends are you seeing from a vacancy badge swipe or even other metric perspective? Are you starting to see the Northeast come back as far as activity, leasing or the multi-tenant office space is anything really starting to inflect upward?

So absolutely, we're seeing a return to office across our entire footprint in the U.S. and Canada, although the pace of that return varies dramatically by geography. The vacancy rates in our nearly 1 billion square feet of office space increased from 6% pre-pandemic to 16% at peak. Those vacancy rates are beginning to decline. We're seeing net absorption of square footage and leases. We're seeing the average lease size of new lease assigned being about 20% below pre-pandemic levels, which makes sense when you believe that the typical office is anticipating 30% to 40% of its workforce on any given day be remote. In addition to that, we are measuring third-party data such as leasing reports, CoStar data and security badge swipes. Badge swipes are an actual indicator of number of employees entering the building per day. We have seen that number trough at only a couple of percent of pre-pandemic levels during the initial lockdowns. They kind of reverted to about a 20% badge swipe number for over a year. We've seen that number steadily improve over the past 6 or 7 months, even with the Omicron variant and the subsequent return to hybrid from that. The differences are fairly significant. I would say San Francisco is our worst-performing market followed by a couple of our Northeast cities, D.C., Boston, New York, Upper Midwest, Chicago, Toronto, a little better and then South and Southeast probably the best with cities like Miami returning to between 75% and 80% of pre-pandemic badge swipes. I think across the entire footprint, we're hovering just below 50% of pre-pandemic levels. But under that average looks a great deal of geographic differentiation. But remember, our corporate product is not sold on a metered basis but rather on a fixed connection basis and because companies have some employees in the office, they need the connection. And then secondly, because those companies are still at similar or even greater staffing levels as we return to pre-pandemic total employment numbers that are effectively flat and there are 10 million job openings in America. Many companies have increased their ad hoc VPN aggregation size connections and number of locations. That has been a positive to Cogent and has allowed us to offset any branch office closures.

Got it. I wanted to go back to your -- the way you characterize your total addressable market. You used to target buildings that have about 51 tenants on average in a building. Maybe you can tell us if there's an update to that number? But you'd usually serve about 14.5 of those 51 tenants or ballpark that's kind of the TAM we were all thinking. Now can you provide some perspective on what happens when tenants average lease sizes in those office spaces change? And I guess, in this case, they're largely being reduced. But when -- even though the leases are changed, what happens from a Cogent perspective, from a services perspective?

So 3 very different factors. First, the aggregate vacancy rate in the building needs to revert to kind of a more normalized level. That is going to happen primarily due to the fact that a lot of the inventory in lesser buildings is being taken off market through residential conversion. So many tenants that were in B and C buildings are actually migrating into A buildings. And in every other economic downturn since World War II we have seen the A buildings, big, tall, shiny skyscrapers recover the most quickly. Rents typically go down, but occupancy rates return to pre-pandemic levels. The second thing that will happen is with the average new lease and many existing leases shrinking by about 20% to accommodate less employees in the office every day. So in a pre-pandemic office design, an architect would assume that 3% of the workforce would be vacant either on vacation or sick on any given day. Today's design criteria is more like around 30% to 40% of the workforce will be remote. So as a result of this and an increase in common meeting areas, we've seen the average square footage shrink by 20%. Those 2 statistics, offices returning to pre-pandemic occupancy as a percentage of square footage and average lease size shrinking means our TAM will increase from 51 tenants to roughly 60 tenants per building. And with the 14.4% signed today that will give us both the ability to capture market share that hasn't signed and a bigger total market. The third point is that because companies are expecting employees to be remote at least part of the time, their dependence on VPN connectivity and the Internet have increased. While the Internet was important pre-pandemic, it's become indispensable post pandemic. That is a positive. Our biggest competitor is actually complacency. It's the inertia of doing nothing. And for Cogent, there are 3 windows when a corporate customer will consider switching: when they first move, when their current network fails or there's a change in their IT infrastructure. And that change in infrastructure is occurring by this transition to a hybrid workforce. So we view this combination of factors as positive and sufficient even if there is an economic downturn due to a recession resulting in our corporate growth continuing to grow in that 10%, 11% year-over-year rate.

Got it. Got it. Just from a pricing scheme perspective, could you just walk us through any of the changes you're seeing? So even if actual office foot space is being reduced, is the amount of speed or is the price paid for that speed is that changing at all or is it trending in the right direction from a Cogent perspective?

So the effective price per megabit for corporate customers is coming down because customers are using more of the bandwidth they purchased due to these changes in their business models and where their employees work. The positives for Cogent have been: one, an acceleration of a trend that was in place pre-pandemic of migrating from 100 meg connections to gigabit connections uplifting ARPU by about $200 and more than offsetting the natural kind of 3% per year price erosion, kind of a same-store sales price decline due to contract lengthening and competitive pressures. The second positive factor has been increasing, but still not statistically significant portion of the base migrating to 10-gigabit connections, which represents about an additional 4x increase in ARPUs. The additional factor of adding redundancy, that third factor of adding another location, so you don't have a single point of failure for your remote employees, has caused many corporate customers to buy connections in data centers for that second aggregation point. That is a positive. On the negative side, it's this closure of remote offices and it's the lack of office-to-office VPNs that have been more difficult post pandemic. Putting these 5 factors together, we think our ARPUs will remain relatively flat to increase in our corporate on-net business and our revenue will increase due to flat to increase in ARPUs and an increase in number of connections.

Okay, got it. Perfect. I want to shift gears into the NetCentric segment a little bit. So we've now had almost about 2 years of very robust growth in the NetCentric segment. But a lot of businesses that saw the massive pandemic-driven growth, including streaming companies and other OTT-type service providers have started to see a roll down of demand. So what are you seeing on your end as far as how these pandemic beneficiaries are performing now from a network perspective?

We're continuing to see robust corporate and NetCentric growth. We're continuing to see traffic grow at similar levels to last quarter. While we don't give specific quarterly guidance, we are doing, I think, pretty well on traffic growth. I know that some of our customers have not grown as quickly. There are a number of factors going on. Half of our NetCentric revenues come from the 7,600 access networks around the world that buy upstream from us. The other half come from content generators. Within those content generators, we've really seen 3 things happen. We've seen a broadening of the customer base to more midsized purchasers. Two, we have seen a slowdown in some of the aggregators that have functions as intermediaries, CDNs and many customers building their own CDNs out as opposed to relying on a third party and we're seeing continued globalization. The U.S. share of the Internet is continuing to decline. All of these factors have actually been a positive to Cogent's ability to grow our revenues and the fact that our effective price per megabit rate of decline is much more benign than the 23% headline number we've averaged over the past 20 years and the roughly 22% decline on a headline basis last quarter much in line with historical averages. We are able to get paid in 73% of the instances by senders and receivers. We're getting paid a higher effective rate because the customer base is more fragmented. And finally, we're getting more of that traffic in more expensive regions in the world that have allowed us to see our effective price decline much more slowly. Now with all of that positive news, we anticipate our NetCentric growth rate will not continue at these extremely elevated levels, but rather will revert to something closer to our long-term average of 9% year-over-year. Different than the corporate business, the NetCentric business has always experienced some seasonality and some short-term volatility, but we think that 9% growth rate is sustainable, not the 18% year-over-year growth rate that we delivered in NetCentric last quarter.

Got it. Got it. I wanted to shift gears a little bit onto your sales force. So can you speak to Cogent's ability to attract and ramp new team members given a tighter-than-normal labor market? And given some of the situation, I guess, the economic dynamic that's going on today, there's clearly some companies hiring and some companies doing the opposite. Could you walk us through how the Cogent sales force is progressing from a hiring and productivity perspective?

Yes. So we actually continue to accelerate our gross hiring throughout the pandemic. We actually hired the most people in Cogent's history in 2020 and then it was exceeded in '21, and we're on pace to continue that outpaced hiring in '22. What hurt us in the sales force was an increase in sales force turnover. We went into the pandemic with 5.2% of the sales force churning every month. That number increased to 8.9% at peak and has subsequently declined to 6.9% of the sales force per month in Q1. We expect that rate of turnover to continue to moderate and get back down to that roughly 5% on -- or 5.2% on a normalized basis. We saw our total sales force headcount decline by about 20%. We continue to have over 15 job applicants per opening. That's actually up from our historic average. Many of these applicants though are not a great fit for Cogent. And we think that our comp packages, even in this period of wage inflation, are substantially better than the prospects alternatives. Our starting comp for a rep is $80,000 a year plus a robust benefits package that is 40 of base salary and 40 a variable. That variable is guaranteed for the first 3 months. The first full month is entirely training, then followed by kind of a hybrid training and work model, allowing a rep to start to build a sales funnel. With that funnel, the rep then goes to a $40,000 base and the 40 is completely variable. We, again, think that this is sufficient, even in expensive markets like New York and San Francisco, to attract new hires. The typical Cogent hire is someone who went to a state school with a less than [indiscernible] grade point average and probably a not fully marketable area of specialization, History, English, Sociology in which there are less job opportunities. Typically, we hire someone who's had 1 or 2 selling jobs out of college before they join Cogent. It is a high activity job that requires a 100 outreaches a day. That's probably worse than a sell-side analyst, Sami, in terms of how many calls you have to make. And it's a rare individual that can keep up that pace to be successful. But we do identify those individuals, and we expect to continue to grow the sales force from the trough that we hit probably in the last couple of quarters where we've been flat by that kind of 7% to 10% rate that we were at pre-pandemic, meaning it will take about 2 years to recover the ground that we've lost in the pandemic on a net basis and then to continue to grow and expand beyond that. But we don't believe there is a candidate shortage.

Got it. Got it. We only have a little bit of time left, but I wanted to ask you a quick one. Given the market volatility that's been going on and where Cogent's stock price is, do you change any -- or would the market volatility that we're seeing compel you to actually change the way you return capital to shareholders? Is dividend still the preference? Or at what point would you consider buybacks a bit more aggressively than the last couple of quarters?

So 3 parts to the answer. One, we are totally committed to returning increasing amounts of capital, and our most recent debt raise gave us additional flexibility to do that. The second is we have used a systematic dividend growth policy as an effective way to return that capital and now have 39 sequential consecutive quarters of dividend growth. We value that, but we also, thirdly, recognize that we can monetize volatility. Our stock has declined but maybe not as aggressively as the rest of the market. I think there's probably still some market volatility that has not yet fully work through the system. So it is something we monitor daily. I'm not prepared to give investors a specific answer to that question. But I do think that over time, we will be in a position to use both tools to return capital and toggle between them where appropriate.

Got it. Got it. All right. I appreciate it, Dave. Thank you, everyone else for joining us and look forward to tuning in for the rest of the conference. Appreciate your time, Dave.

Thanks, Sami.